Bivalent RNA interference to increase isoflavone biosynthesis in soybean (Glycine max)

Bivalent RNA interference to increase isoflavone biosynthesis in soybean (Glycine max)

Yina Jiang; Yanlin Hu; Biao Wang; Tianlong Wu^*

Key Laboratory of Urban Agriculture (South) Ministry of Agriculture; School of Agriculture and Biology; Shanghai Jiao Tong University; Dongchuan Rd., Shanghai -PR -China

ABSTRACT

In this work, a bivalent RNA interference (RNAi) plant-transformation vector was constructed to silence both the flavanone 3-hydroxylase (F3H) gene and the flavone synthase II (GmFNSII) gene in soybean (Glycine max). Two further unit RNAi vectors were constructed for each of these two genes. RNAi-mediated suppression of these genes effectively regulated flavone and isoflavone production in hairy roots that arose from soybean cotyledons transformed with Agrobacterium rhizogenes ATCC15834. Notably, the bivalent RNAi vector had a significantly higher effect for increasing isoflavone production compared with the two unit RNAi vectors. The study highlighted molecular methods that could be used to enhance isoflavone production in soybean and demonstrated the challenges associated with such metabolic engineering for the production of plant natural products.

Key words: Bivalent RNA interference, Flavanone 3-hydroxylase, Flavone synthase II, Isoflavone biosynthesis, Soybean

INTRODUCTION

Flavonoids are secondary metabolites that are found throughout the plant kingdom and more than 10,000 flavonoids have been described to-date (Tahara 2007). All flavonoids are synthesized from two basic metabolites, malonyl-CoA and pcoumaroyl-CoA at a ratio of 3:1 to 15-carbon skeletons. The derived chalcone intermediates comprise two phenolic groups that are connected by an open three-carbon bridge. That linkage is part of an additional heterocyclic six-member ring that involves one of the phenolic groups on the adjacent ring. Based on the structure of their basic chalcone skeleton, various classes of flavonoids can be synthesized, including flavanones, isoflavones, flavanols, anthocyanins, flavonols and flavones (Kim et al. 2008; Fowler and Koffas 2009).

In nature, flavonoids are involved in many biological processes. Plant flavonoids are involved in defense responses to biotic and abiotic stresses. Anthocyanins are visible flavonoid pigments that give rise to the red and blue colors of certain ripe fruits and these pigments attract frugivores that can help disperse seeds (Wilson and Whelan 1990). Flavones are important for protecting against ultraviolet light, regulating auxin transport and controlling flower color (Harborne and Williams 2000). Isoflavones, the most abundant flavonoids in soybean (Glycine max), play diverse roles in plant-microbe interactions (Ferguson et al. 2010). For example, isoflavones function as preformed antibiotics, and they act as precursors for the pterocarpan phytoalexins (Ebel 1986; Rivera-Vargas et al. 1993). Isoflavones also function as signaling molecules for the induction of microbial genes involved in the infection and symbiosis between soybean and Bradyrhizobium japonicum (Dixon and Sumner 2003; Ferguson and Mathesius 2003). Furthermore, they have direct but complex effects on human nutrition and health; they can reduce cholesterol levels and prevent certain cancers (Weisshaar and Jenkinst 1998; Lee et al. 2005; Kim et al. 2006; Kim and Lee 2007). Isoflavones are especially important for preventing many hormone-dependent cancers and these compounds can improve women's health (Beecher 2003).

Metabolic engineering tools have opened up new possibilities for studying the regulation and accumulation of plant secondary metabolites and investigating their roles in plant defenses (Subramanian et al. 2005). Successes in enhancing the production capacities have come from the manipulation of blocks of genes that encode segments of pathways, over-or under-expression of genes controlling putative rate-limiting steps in pathways, and the expression of transcription factors that regulate entire metabolic pathways (Wu and Chappell 2008). In soybean, the isoflavones daidzein I, genistein II and glycitein III are synthesized via the phenylpropanoid pathway and these compounds are stored in the vacuole as glucosyl-and malonyl-glucose conjugates (Graham 1991). Genistein II, dihydroflavonol, and apigenin are produced from the same naringenin substrate via three different branches (Fig. 1), which are mediated by isoflavone synthase, flavanone-3-hydroxylase, and flavone synthase II, respectively. Importantly, the genes encoding each of these enzymes have been identified (Steele et al. 1999; Zabala and Vodkin 2005; Jiang et al. 2010). Ralston et al. (2005) described the RNA interference (RNAi) of soybean isoflavone synthase genes using the soybean cotyledon system, and this led to enhanced plant susceptibility to Phytophthore sojae in hairy roots. Jiang et al. (2010) suggested that RNAi-mediated suppression of isoflavone synthase pathway genes could enhance isoflavone production in soybean.

In this present study, two unit RNAi planttransformation vectors were constructed to silence the flavanone 3-hydroxylase gene (F3H) and the flavone synthase II gene (GmFNSII) in soybean. In addition, a bivalent RNAi vector was constructed for both of these genes.

MATERIALS AND METHODS

Cloning full-length cDNAs of GmFNSII and F3H

GmFNSII and F3H cDNA sequences were identified from the SoyBase G. max genomic database (http://soybase.org/SequenceIntro.php). To obtain the full-length cDNAs for GmFNSII-2 (GenBank: GU568028) and F3H (GenBank: AF198451), total RNA was isolated from the leaf tissues of field-grown soybean cultivar "Hefeng 47" seedlings using Trizol Reagent (Invitrogen). First-strand cDNAs were cloned by reversetranscriptase polymerase chain reaction (RT-PCR) using the PrimeScript^TM 1st Strand cDNA Synthesis Kit (TaKaRa). These cDNAs were used as templates in the PCR for full-length GmFNSII and F3H cDNA amplification using the following primers: GmFNSII-F/R for GmFNSII and F3H-F/R for F3H (Table 1). The corresponding PCR products were cloned into the pMD18-T vector (TaKaRa) and then sequenced.

Construction of bivalent RNAi vectors

An RNAi vector was constructed for silencing the GmFNSII transcripts. A 322-bp coding region located between 897-1218 of the GmFNSII cDNA sequence was selected and amplified by the PCR using pMD18-GmFNSII as the template. The forward (5'-CTG(ATCGAT/GGTACC) AACTCGGAACCATGTCAAATC-3') and reverse (5'-CCGG(TCTAGA/CTCGAG) GTTTACGCAAACTATTGAACCT-3') primer sequences contained two different sets of restriction sites (Table 2). The PCR products were cloned into the pHANNIBAL vector in opposing orientations on either side of a PDK intron to create an invert repeat driven by the cauliflower mosaic virus (CaMV) 35S promoter (Wesley et al. 2001). Then, this GmFNSII RNAi construct was cloned into the pCAMBIA3300 plant expression vector (http://www.cambia.org/ daisy/cambia/home.html), which included the 35S: BAR fragment and was digested with Sac I/Pst I. The newly construct was named p3300-GmFNSIIi.

An RNAi vector was constructed for silencing the F3H transcripts. A 303-bp coding region located between 694-997 of the F3H cDNA sequence was selected and amplified by the PCR using pMD18-F3H as the template. The forward (5'-CTG (ATCGAT/GGTACC)TGACCTCA CTCTTGGCCT-3') and reverse (5'-CCGG (TCTAGA/CTCGAG)TCTCCTTCTCTTATCTT CAGAG-3') primer sequences contained two different sets of restriction sites (Table 2). The PCR products were cloned into the pHANNIBAL vector and then this F3H RNAi construct was cloned into pCAMBIA3300 as described above and named p3300-F3Hi.

A bivalent RNAi vector was constructed for silencing the GmFNSII and F3H transcripts. The F3H RNAi construct in pHANNIBAL and p3300-GmFNSIIi were double digested using Pst I and Pvu I restriction sites and then the F3H RNAi construct was cloned into p3300-GmFNSIIi; this was named p3300-GmFNSIIi-F3Hi. The newly constructed p3300-F3Hi, p3300-GmFNSIIi and p3300-GmFNSIIi-F3Hi vectors were inserted into A. rhizogenes ATCC15834 for the subsequent transformation of the soybean cotyledons. The pCAMBIA3300 vector served as a control. pHANNIBAL, pCAMBIA3300 and A. rhizogenes were kindly gifted by Dr Tang (Fudan-SJTU-Nottingham Plant Biotechnology R&D Center).

Soybean cotyledon transformation and southern blotting

Wild-type and transformed strains of A. rhizogenes were maintained by culturing on yeast extract peptone (YEP) agar. For the transformed strains, all the media contained 50 µg mL^-1 kanamycin. Cultures for plant inoculation were grown overnight in 10 mL of YEP broth at 28ºC. The cells were pelleted by centrifugation at 5,000 ×g and 4ºC for 10 min. Cell pellets were drained briefly and gently re-suspended in quarter-strength Murashige aninSkoog medium to a final optical density of 0.5 at 600 nm before the cotyledon tissues were inoculated.

The cotyledon transformations were carried out as follows. Soybean cultivar "Hefeng 47" seeds that had been stored in a cold room were surfacesterilized in 70% ethanol for 1 min and then in 100 mL of 15% Clorox for 30 min, with occasional agitation. Afterward, ten seeds were placed into each Petri dish that contained germination medium (Murashige-Skoog medium with 0.4 mg L^-1 6-Benzyladenine). Seeds were germinated for 5-7 d at 26ºC under a 16-h photoperiod. Then, cotyledons showing no surface blemishes were sampled at 0.3 cm away from the petiole end by making small, circular (0.4 cm diameter) incisions. The cut surfaces of six cotyledons were placed on to 9-cm sterile filter paper in each plate, and this formed a wet surface on which the cotyledons were exposed to 20 µL of A. rhizogenes suspension. Plates were wrapped in Parafilm and kept in a Percival incubator at 22ºC operating a 12-h photoperiod (Subramanian et al. 2005).

After transformation for three weeks, genomic DNA (30-60 µg) of hairy roots containing the wild-type and transformed vectors were cleaved with Pst I and digested DNA was then fractionated on a 1.0% agarose gel, denatured, and transferred onto nylon positively charged membranes (Osmonics). Southern blotting followed the manufacturer's instructions (Amersham). The DNA probe (approximately 600 bp), including the BAR coding region in pCAMBIA3300 vector, which was detected using the primers BAR-F and BAR-S (Table 1), was labeled with alkaline phosphatase and the membrane exposed to film with CDPStar^TM chemiluminescent detection reagents (Amersham, RPN3690). Hybridization and stringency washes were performed at 55ºC.

Quantitative RT-PCR analysis of GmFNSII and F3H

Total RNA was isolated separately from each of fifteen transgenic hairy roots that resulted from using the wild-type and transformed vectors. Samples from the hairy roots were treated with Trizol (Invitrogen) and then DNase I to remove any contaminating genomic DNA. First-strand cDNAs were cloned by RT-PCR with a PrimeScript^TM 1st Strand cDNA Synthesis Kit (TaKaRa). Quantitative real-time (qRT)-PCR was performed with SYBR-Green and a 7300 Real-Time PCR System (Applied Biosystems). Initial transcript concentrations were estimated according to the comparative threshold cycle method (Bovy et al. 2002). Transcript abundance of the soybean Actin gene (V00450) in each RNA preparation was used as the internal standard. The PCR primers used were Actin-F and Actin-R for Actin, qRT_FNS-F and qRT_FNS-R for GmFNSII, and qRT_F3H-F and qRT_F3H-R for F3H (Table 1).

Flavonoid analysis by high-performance liquid chromatography (HPLC)

To evaluate flavonoid composition, 100 mg of transgenic hairy roots were ground in liquid N2 and extracted for 2-h with 4 mL of 80% methanol using ultrasonic treatment (100 W). For HPLC, aliquots of each extract were analyzed at 260 nm on an Agilent 1100 series HPLC system using a Venusil MP-C18 column (2.1×150 mm, 5 µm, Agela Technologies, Inc.). Samples were separated using a linear gradient of 15-85% methyl alcohol in HPLC-grade water for 30 min at a flow rate of 1 mL min^-1. Retention times and UV spectra were compared with authentic standards where available, and the quantity of each compound was calculated by comparison with standard curves prepared using the authentic compounds (Sigma-Aldrich, St. Louis, MO). The quantuities of the isoflavone aglycones, glucoseconjugates (daidzin, genistin and glycitin) and malonyl-glucose conjugates were determined for each individual isoflavone and summed to give the total amounts for daidzein I, genistein II and glycitein III.

RESULTS

Construction of the binary RNAi vectors

To determine the roles of GmFNSII and F3H in flavonoid production and in the accumulation of isoflavone in particular, two unit RNAi vectors and one bivalent RNAi vector were constructed for silencing the F3H and GmFNSII genes in soybean. A 322-bp coding region of GmFNSII cDNA and a 303-bp coding region of F3H cDNA were amplified and cloned into the pCAMBIA3300 RNAi binary vector containing a BAR selectable marker (Fig. 2). After verification by restriction enzyme digestion and sequencing, the newly constructed p3300-GmFNSIIi, p3300-F3Hi and p3300-GmFNSIIi-F3Hi vectors were transformed successfully into soybean cotyledons by A. rhizogenes to obtain the hairy roots.

The RNAi strategy down-regulated GmFNSII and F3H expression in soybean

After transformation for three weeks, hairy roots containing the wild-type and transformed vectors were identified by southern blotting and transformed lines carrying the BAR gene were selected (Fig. 3). To investigate the effect of the new constructions, GmFNSII and F3H expression levels were analyzed by qRT-PCR. Total RNA was extracted from the transgenic hairy roots and reverse-transcribed to cDNAs. As expected, GmFNSII and F3H RNAi transgenic hairy roots showed significant reductions (to <10%) in GmFNSII and F3H expression (Fig. 4). At the same time, in the unit RNAi transgenic hairy roots, expression of GmFNSII in F3H RNAi transgenic hairy roots and expression of F3H in GmFNSII RNAi roots showed significant addition, especially for GmFNSII (about a two-fold increase compared with the control).

Silencing of GmFNSII and F3H caused high levels of isoflavone accumulation

Single hairy roots of independent positive transformation events were analyzed by HPLC to determine the effects of silencing on the production profiles of root flavones and isoflavones. Mean levels of individual isoflavones and total isoflavone levels varied between the 15 individual transgenic hairy roots. The isoflavone component profile of a control transformant (transformed with just the pCAMBIA3300 vector) showed that daidzein I was the most abundant of the isoflavones in hairy roots (Fig. 5). Genistein II was generally the next abundant component, while glycitein III was found at lowest levels.

Since genistein II, dihydroflavonol and apigenin are all produced from the same naringenin substrate, silencing any of the encoding genes should lead to increased accumulation of the other two compounds. As expected, >90% of the GmFNSII RNAi roots had significantly lower levels of apigenin (the major flavone) compared with the control roots. Some of these transgenic roots had undetectable levels of apigenin, indicating effective silencing. At the same time, genistein was found at markedly higher levels. The complete opposite result was observed in the F3H RNAi roots, which contained significantly greater levels of apigenin and genistein (Fig. 5). Daidzein I, genistein II and glycitein III levels were summed to give the total isoflavone levels. GmFNSII RNAi

DISCUSSION

Flavonoids are major secondary plant metabolites that mediate diverse biological functions in essential physiological processes and exert significant ecological impacts (Fowler and Koffas 2009). Many flavonoids are used as nutrients, colorants, flavors, fragrances and medicines (Newman and Cragg 2007, Wu and Chappell 2008). Metabolic engineering of the plants for the production of natural products can help alleviate the demands for limited natural resources. Successes in enhancing isoflavone biosynthesis production in soybean include the manipulation of the blocks of genes that encode segments of pathways, up-and down-regulation of putative rate-limiting steps in pathways, and the expression of transcription factors that regulate entire metabolic pathways (Yu et al. 2003; Subramanian et al. 2005; Lozovaya et al. 2007; Jiang et al. 2010).

In this present study, RNAi-mediated suppression of GmFNSII or/and F3H in the hairy roots of soybean that arose from cotyledons transformed with A. rhizogenes was tested to define the roles of these genes in the accumulation of flavones and isoflavones. qRT-PCR data showed significant reductions of GmFNSII and F3H expression in transgenic hairy roots, which confirmed that the new constructions were very effective in downregulating the two target genes. These RNAi constructions were expressed under the control of the constitutive CaMV 35S promoter, and therefore it was expected that the transgene effect would not be restricted to the hairy roots but would also influence the flavonoid pathway in other parts of the soybean.

HPLC analysis showed that down-regulation of F3H had a greater effect for increasing the total isoflavone production compared with downregulation of GmFNSII. One possible reason for this was the low expression activation of GmFNSII. The expression levels of GmFNSII were only approximately one-fifth of the expression of F3H in the vector-control hairy roots. However, since both the genes showed tissue-specific expression patterns and differential transcript levels in all the tissues (Ralston et al. 2005; Jiang et al. 2010), it was not possible to speculate whether the results would be repeated if different tissues were tested in transgenic plants.

CONCLUSION

This is the first time that a bivalent RNAi planttransformation vector has been constructed for silencing both the flavanone 3-hydroxylase gene and the flavone synthase II gene in soybean. This study showed that the bivalent RNAi roots were significantly more effective for increasing isoflavone production than the two unit RNAi roots. The attempt to further enhance isoflavone accumulation in soybean reinforced the knowledge on the relationships between different genes and highlighted an effective combination of different approaches in one method.

ACKNOWLEDGEMENTS

This work was supported by the grants from the National Natural Science Fund (31071435), the National Fund on "Breeding of New Cultivars" (2011ZX08004-003), Ministry of Education Fund on New Staff (20100073120059), and the Shanghai Horticulture Key Discipline Construction Fund (B209).

Received: June 20, 2012

Accepted: August 12, 2013.

• Beecher GR. Overview of dietary flavonoids: nomenclature, occurrence and intake. J Nutr 2003; 133:3248S-3254S.
• Bovy A, de Vos R, Kemper M, Schijlen E, Pertejo MA, Muir S, et al. High-flavonol tomatoes resulting from the heterologous expression of the maize transcription factor genes LC and C1. Plant Cell 2002; 14:2509-2526.
• Dixon RA, Sumner LW. Legume natural products: understanding and manipulating complex pathways for human and animal health. Plant Physiol 2003;131:878-885.
• Ebel J. Phytoalexin synthesis: the biochemical analysis of the induction process. Annu Rev Phytopathol 1986;24:235-264.
• Ferguson BJ, Indrasumunar A, Hayashi S, Lin MH, Lin YH, Reid DE, et al. Molecular analysis of legume nodule development and autoregulation. J Integr Plant Biol 2010;52:61-76.
• Ferguson BJ, Mathesius U. Signaling interactions during nodule development. J Plant Growth Regul 2003;22: 47-72.
• Fowler ZL, Koffas MAG. Biosynthesis and biotechnological production of flavanones: current state and perspectives. Appl Microbiol Biotechnol 2009;83:799-808.
• Graham TL. Flavonoid and isoflavonoid distribution in developing soybean seedling tissues and in seed and root exudates. Plant Physiol 1991;95,594-603.
• Harborne JB, Williams CA. Advances in flavonoid research since 1992. Phytochem. 2000;55:481-504.
• Jiang YN, Wang B, Li H, Yao LM, Wu TL. Flavonoid Production is effectively regulated by RNAi interference of two flavone synthase genes from Glycine max. J Plant Biol 2010;53:425-432.
• Kim JD, Liu L, Guo W, Meydani M. Chemical structure of flavonols in relation to modulation of angiogenesis and immune-endothelial cell adhesion. J Nutr Biochem 2006;17:165-176.
• Kim JH, Cheon YM, Kim BG, Ahn JH. Analysis of Flavonoids and Characterization of the OsFNS Gene Involved in Flavone Biosynthesis in Rice. J Plant Biol 2008;51:97-101.
• Kim YH, Lee YJ. TRAIL apoptosis is enhanced by quercetin through Akt dephosphorylation. J Cell Biochem 2007; 100:998-1009.
• Lee BH, Jung SM, Lee JH, Kim JH, Yoon IS, Lee JH, et al. Quercetin inhibits the 5-hydroxytryptamine type 3 receptor-mediated ion current by interacting with pretransmembrane domain I. Mol Cells 2005;20:69-73.
• Lozovaya VV, Lygin AV, Zernova OV, Ulanov AV, Li SX, Hartman GL, et al. Modification of phenolic metabolism in soybean hairy roots through down regulation of chalcone synthase or isoflavone synthase. Planta 2007;225:665-679.
• Newman DJ, Cragg GM. Natural products as sources of new drugs over the last 25 years. J Nat Prod 2007;70:461-477.
• Ralston L, Subramanian S, Matsuno M, Yu O. Partial reconstruction of flavonoid and isoflavonoid biosynthesis in yeast using soybean type I and type II chalcone isomerases. Plant Physiol 2005;137:17751788.
• Rivera-Vargas LI, Schmitthenner AF, Graham TL. Soybean flavonoid effects on and metabolism by Phytophthora sojae. Phytoche 1993;32:851-857.
• Steele CL, Gijzen M, Qutob D, Dixon RA. Molecular characterization of the enzyme catalyzing the aryl migration reaction of isoflavonoid biosynthesis in soybean. Arch Biochem Biophysiol. 1999;367:146-150.
• Subramanian S, Graham MY, Yu O, Graham TL. RNA interference of soybean isoflavone synthase genes leads to silencing in tissues distal to the transformation site and to enhanced susceptibility to Phytophthora sojae. Plant Physiol 2005;137:1345-1353.
• Tahara S. A journey of twenty-five years through the ecological biochemistry of flavonoids. Biosci Biotechnol Biochem 2007;71:1387-1404.
• Weisshaar B, Jenkinst GI. Phenylpropanoid biosynthesis and its regulation. Curr Opin Plant Biol 1998;1:251-257.
• Wesley SV, Helliwell CA, Smith NA, Wang MB, Rouse DT, Liu Q, et al. Construct design for effcient, effective and high-throughput gene silencing in plants. Plant J 2001;27:581-590.
• Wilson MF, Whelan CJ. The evolution of fruit color in fleshy-fruited plants. Am Nat 1990;136:790-809.
• Wu SQ, Chappell J. Metabolic engineering of natural products in plants; tools of the trade and challenges for the future. Curr Opin Biotechnol 2008;19(2):145-152.
• Yu O, Jung W, Shi J, Croes RA, Fader GM, McGonigle B, et al. Production of the isoflavones genistein and daidzein in non-legume dicot and monocot tissues. Plant Physiol 2000;124:781-794.
• Yu O, Shi J, Hession AO, Maxwell CA, McGonigle B, Odell JT. Metabolic engineering to increase isoflavone biosynthesis in soybean seed. Phytochem 2003;63:753-763.
• Zabala G, Vodkin LO. The wp mutation of Glycine max carries a gene-fragment-rich transposon of the CACTA superfamily. Plant Cell 2005;17:2619-2632.

Publication Dates

• Publication in this collection
  26 Nov 2013
• Date of issue
  Apr 2014

History

• Received
  20 June 2012
• Accepted
  12 Aug 2013