Logo of plntphysLink to Publisher's site
Plant Physiol. 2007 Jan; 143(1): 504–516.
PMCID: PMC1761954

Ectopic Expression of a Basic Helix-Loop-Helix Gene Transactivates Parallel Pathways of Proanthocyanidin Biosynthesis. Structure, Expression Analysis, and Genetic Control of Leucoanthocyanidin 4-Reductase and Anthocyanidin Reductase Genes in Lotus corniculatus1,[W]


Proanthocyanidins (PAs) are plant secondary metabolites and are composed primarily of catechin and epicatechin units in higher plant species. Due to the ability of PAs to bind reversibly with plant proteins to improve digestion and reduce bloat, engineering this pathway in leaves is a major goal for forage breeders. Here, we report the cloning and expression analysis of anthocyanidin reductase (ANR) and leucoanthocyanidin 4-reductase (LAR), two genes encoding enzymes committed to epicatechin and catechin biosynthesis, respectively, in Lotus corniculatus. We show the presence of two LAR gene families (LAR1 and LAR2) and that the steady-state levels of ANR and LAR1 genes correlate with the levels of PAs in leaves of wild-type and transgenic plants. Interestingly, ANR and LAR1, but not LAR2, genes produced active proteins following heterologous expression in Escherichia coli and are affected by the same basic helix-loop-helix transcription factor that promotes PA accumulation in cells of palisade and spongy mesophyll. This study provides direct evidence that the same subclass of transcription factors can mediate the expression of the structural genes of both branches of PA biosynthesis.

Flavonoids represent one of the oldest, largest, and most diverse families of plant secondary metabolites. These compounds fulfill a multitude of functions during plant development (Winkel-Shirley, 2001) and are also of significant interest as antioxidant and anticancer agents in the human diet (Dixon et al., 2005; Tohge et al., 2005). Proanthocyanidins (PAs), also known as condensed tannins, are oligomeric and polymeric end products of the flavonoid biosynthetic pathway. They are widespread in nature, occurring in numerous plant species, including many important plant-derived food materials (Tanner, 2004).

The beneficial effects of PAs on human health and the significant pharmacological activities of PAs have been recently reviewed (Marles et al., 2003; Dixon et al., 2005). In planta, PAs act as protectants against pathogens, pests, and diseases and additionally control seed permeability and dormancy (Debeaujon et al., 2000). These compounds strongly affect plant quality traits, and the palatability and nutritive value of forage legumes are highly influenced by their concentration and structure (Barry and McNabb, 1999). High concentrations of PAs can decrease the palatability and digestibility of plants. By contrast, moderate quantities of PAs (2%–4% dry matter) in forage prevent proteolysis during ensiling and rumen fermentation, thereby protecting ruminants against pasture bloat (Tanner et al., 1995). In fact, by binding to proteins in the rumen, PAs reduce the rate of fermentation and increase the levels of proteins passing through the rumen of grazing animals. PAs therefore make the conversion of plant protein into animal protein more efficient, with reduced methane production (Kingston-Smith and Thomas, 2003). However, PAs only accumulate in the seed coats of the most valuable forage species, such as alfalfa (Medicago sativa) and clovers (Trifolium spp.), and are absent from their leaves. The environmental and agronomical benefits that could be derived from triggering the accumulation of a moderate amount of PAs in forage leaves are of considerable importance in the protection and nutrition of ruminant animals (Damiani et al., 2000).

PAs vary widely in composition, depending upon the nature (stereochemistry and hydroxylation pattern) of the flavan 3-ol starter and extension units, the position and stereochemistry of the linkage to the lower unit, the degree of polymerization, and modifications of the 3-hydroxyl group (for recent and extensive reviews, see Tanner, 2004; Dixon et al., 2005). The steps involved in PA polymerization, hydroxylation, and sequestration are still to be fully elucidated. Conversely, understanding of the biosynthesis of the building blocks of PAs has progressed significantly over the last few years. The building blocks or starter units of most PAs are the flavan-3-ols (+) catechin and (−) epicatechin, whose last common biosynthetic intermediate is leucoanthocyanidin (an intermediate in anthocyanin biosynthesis also referred to as flavan 3,4-diol), which results from the reduction of flavonols by dihydroflavonol 4-reductase (DFR). Subsequently, catechin is derived from the reduction of leucoanthocyanidin through the activity of leucoanthocyanidin reductase (LAR), an enzyme that has been purified and characterized in the legume Desmodium uncinatum (Tanner et al., 2003). Epicatechin formation occurs via anthocyanidin synthesis and reduction, two steps catalyzed by anthocyanidin synthase (ANS) and anthocyanidin reductase (ANR), encoded by leucoanthocyanidin dioxygenase (LDOX)/ANS and BANYULS (BAN)/ANR genes, respectively (Abrahams et al., 2003; Xie et al., 2003). Anthocyanidins are consequently essential for PA biosynthesis in species that accumulate only epicatechin-based PAs.

Although the PA biosynthetic pathway has been best characterized in Arabidopsis (Arabidopsis thaliana), this species accumulates only epicatechin-based starter units and lacks any obvious LAR ortholog (Abrahams et al., 2003; Tanner et al., 2003). Thus, Arabidopsis does not offer a system for the complete understanding of the genetic and environmental determinants that control PA biosynthesis in important crops where parallel catechin- and epicatechin-based pathways contribute to PA biosynthesis (Fig. 1). Additionally, Arabidopsis accumulates PAs only in seed coats, whereas these polymers accumulate in vegetative tissues of some important forage crops and pathway regulation may differ between seed coats and other nonfloral tissues, such as leaves, stems, and roots.

Figure 1.
Pathways to PA starter units (catechin type and epicatechin type) from flavan 3,4-diols in two model species. Superscript letters give references for genes encoding these enzymatic activities in given species and/or reference for structure of PA polymers. ...

The analysis of Arabidopsis seed coat mutants has yielded critical information regarding the transcriptional control of genes involved in PA biosynthesis (Lepiniec et al., 2006). A ternary transcriptional complex comprising TTG1 (encoding a WD-repeat protein), TT2 (encoding a R2R3 MYB-class protein), and TT8 (encoding a basic helix-loop-helix [bHLH] class protein) has been shown to regulate tissue-specific accumulation of PAs. Mutations in any of these three genes result in a transparent testa phenotype with either a reduced or complete absence of PA polymers. The MYB, bHLH, and WD complex also regulates anthocyanin biosynthesis, trichome initiation, and non-root hair cell specification in Arabidopsis. In these cases, the WD-repeat protein (TTG1) is the same in all of the different functional complexes, whereas the bHLH and MYB proteins are flexible in their participation. Therefore, mutations in TTG1 affect the functioning of all the pathways, whereas mutations in the genes encoding the bHLH proteins affect overlapping subsets of the pathways and the specificity in regulating the different pathways in different tissues is conferred by different MYB proteins (Zhang et al., 2003). In this context, mutations in TTG1 and TT8 also affect anthocyanin accumulation in seedlings, consistent with the expression of TT8 and TTG1 in both seed and seedlings. Conversely, TT2 is specifically expressed in the innermost cell layer of the inner seed tegument where it controls PA accumulation by regulating the expression of DFR, LDOX, and BAN. TT8 is similarly important for the expression of DFR and BAN in siliques and TTG1 controls BAN expression mainly by affecting TT8 function (Baudry et al., 2004).

The genus Lotus offers a wide range of options for studying PAs and includes species that accumulate different quantities of condensed tannins; some species accumulate these end products in the majority of plant tissues, whereas others (Lotus japonicus, Lotus tenuis) do not produce condensed tannin polymers in leaf tissues (Harborne, 1988). Another point of interest is that, in Lotus corniculatus and L. japonicus, PA polymers are synthesized from both epicatechin and catechin starter units (Foo et al., 1996). In L. corniculatus, natural or transgenic lines polymorphic for PA quantity in vegetative tissues have already been selected and characterized (Carron et al., 1994; Robbins et al., 2003). We have demonstrated previously that the ectopic expression of the maize (Zea mays) flavonoid regulator Sn, a bHLH gene, increases the levels of PAs in the vegetative tissues of L. corniculatus and regulates the expression of DFR and ANS (Paolocci et al., 2005).

Here, we report on the cloning and characterization of genes encoding ANR and LAR from L. corniculatus. Expression analysis of these genes in wild-type and transgenic lines showing different levels of leaf PAs and the analysis of cell types in leaf tissues, which accumulate detectable levels of PAs, have developed our understanding of the genetic control of PA biosynthesis in this forage legume. Our findings may also be of importance in understanding the genetic mechanisms controlling PA accumulation in other tanniferous plants, where both branches of the PA pathway are operating.


Cloning of ANR cDNAs

By aligning the ANR cDNA sequences from Medicago truncatula (AY184243) and Phaseolus coccineus (BN000164), degenerate primers were designed and used to amplify an orthologous leaf cDNA fragment from L. corniculatus S41, which was then extended to the 5′ and 3′ ends by RACE. The ANR amplicons resulting from end-to-end PCR on leaf cDNA were cloned in a plasmid vector and six clones were selected. Four clones harbored a fragment of about 1.2 kb, whereas the other two harbored fragments of about 1,100 bp and 500 bp, respectively. Sequence analysis showed that length differences among clones resided basically in the coding sequences. The 5′- and 3′-untranslated regions (UTR) ranged from 73 to 79 bp and from 104 to 109 bp; the longest clones, named LcANR1-1 to LcANR1-4, encoded a 338-amino acid protein and the other two, named LcANR2 and LcANR3, encoded proteins of 302 and 107 amino acids, respectively (Supplemental Fig. S1). The four LcANR1 clones differed slightly from each other, with a total of six conservative amino acid substitutions, and all showed a high (82%) sequence identity at the amino acid level to the ANR cDNAs from M. truncatula and P. coccineus. The identity of LcANR2 and LcANR3 to the above-mentioned cDNAs was still high. However, LcANR2 differed from LcANR1 and ANR cDNAs from other species, such as M. truncatula, P.coccineus, and grapevine (Vitis vinifera), because it lacked most of the coding sequence that showed homology to the NAD-dependent epimerase/dehydratase domain as per the National Center for Biotechnology Information (NCBI) conserved domain search. The same search did not show any putative conserved domain in LcANR3.

Cloning of cDNAs for LAR

A BLASTn similarity search was carried out to identify putative homologous sequences to LAR cDNA from D. uncinatum (AJ550154), the only LAR gene present in public databases when this research was initiated. This search revealed that at least five regions in the M. truncatula bacterial artificial chromosome clone mth2-103j7 (AC146570.4) displayed high nucleotide similarity with the query sequence (data not shown). From these highly homologous regions, degenerate primer pairs were designed to amplify leaf cDNA from the S41 genotype. The cDNA amplicons resulting from amplification with the primer pairs 2LarF/3larR (about 800 bp long) and 5LarF/6larR (about 500 bp long) were then used as targets for 5′- and 3′-RACE PCR amplification. Primers were designed for each set of both 5′ and 3′ ends at the extremity of the longest amplicons and used to amplify the respective full-length cDNAs. Using the different forward/reverse primer combinations, amplicons of two lengths were recovered, the longest, generated by 5Lar1cl/3larbis primers, was named LcLAR1, the other, named LcLAR2, was generated by the 14fwUTR/x3Lar primer combination.

Both LAR1 and LAR2 fragments were cloned and two clones for each gene, named LcLAR1-1, LcLAR1-2 and LcLAR2-1, LcLAR2-2, were sequenced. The 1,418-bp-long LcLAR1-1 cDNA contained 99 nucleotides upstream of the first in-frame ATG, with an open reading frame (ORF) of 1,047 bp (349 amino acids) and a 3′ UTR of 272 bp, the poly(A+) tail not included. Both LAR1-1 and LAR1-2 showed the microsatellite-like motif (TG)4(TGAG)4 in the 5′ UTR. LAR1-2 differed from LAR1-1 by two amino acid substitutions (T-S and N-K at positions 192 and 333, respectively). As regards the LcLAR2 amplification products, LAR2-1 cDNA was 1,114 bp long and contained an ORF of 339 amino acids with a 14-bp 5′ UTR and 79 nucleotides after the stop codon and before the poly(A) tail. LAR2-2 cDNA was 1,120 bp long and contained an ORF of 339 amino acids with a 19-bp 5′ UTR and 81 nucleotides after the stop codon and before the poly(A) tail. At a protein level, three amino acid substitutions were detected between the LAR2-1 and LAR2-2 cDNA clones (Q-L, I-V, and P-S at positions 37, 79, and 229, respectively). The LAR gene cloned from Desmodium was most closely related to isoflavone reductase-like proteins (Tanner et al., 2003) and all the Lotus LAR cDNA clones in this study displayed the RFLP, ICCN, and THD motifs conserved in putative LAR sequences, but absent from the related isoflavone reductase proteins (Bogs et al., 2005).

The cloned LAR1 and LAR2 sequences showed 71% sequence identity at the amino acid level when compared with each other. A BLASTP search showed that both LAR1 genes had the highest sequence identity at the amino acid level to a LAR gene cloned from Lotus uliginosus (96%) and, to a lesser extent, to the LAR genes from M. truncatula and P. coccineus (66%). LAR2-1 cDNA, by contrast, displayed a higher sequence identity to LAR genes cloned from M. truncatula and P. coccineus (78% and 76%, respectively). A phylogenetic tree showing the relation of LcLAR1 and LcLAR2 with LAR sequences from other legume species is shown in Figure 2.

Organization of ANR and LAR Genes

The primer pairs designed to amplify the full-length cDNA of ANR and of the two LAR genes were also used to amplify and clone the corresponding genomic regions from L. corniculatus to gain insight into their organization and structure within the genome. Amplification from the S41 genome using end-to-end PCR with ANR-specific primers produced two products. Alignment of the 2,860-bp-long genomic sequence, named LcANR1, with the longest cDNAs, showed the ANR coding region is organized into six exons and that the ORF of LcANR2 encodes a 302-amino acid polypeptide that lacks the N-terminal part of the second exon with respect to the longest ANR cDNAs (LcANR1). The short ANR genomic fragment, LcANR2, perfectly matched the short ANR cDNA fragment (LcANR3) and was about 540 bp long. In addition to extensive deletion on the 3′ end of the first exon, LcANR2 lacked the exons 2, 3, and 4, and the 5′ end of the fifth exon, while retaining the sixth exon in its entirety with respect to LcANR1 (Fig. 3A).

Figure 3.
Structure of ANR, LAR1, and LAR2 genes. Boxes represent exons; lines introns, thicker lines 5′ and 3′ UTRs. The figure is drawn to scale. A, Genomic structure of ANR genes. Dotted box represents the 108-bp-long deletion that characterizes ...

The LAR1-specific primer pair gave rise to amplicons highly divergent in length. Their sequencing and alignment with the respective LcLAR1 cDNA showed that two genomic fragments coded for LAR1, although with differences in terms of both intron structure and length. The shortest gene (about 2,300 bp), named LcLAR1-1, exhibited a single intron. By contrast, in the second LAR1 gene, the coding sequence was split into five exons (Fig. 3B). The microsatellite-like motif in the 5′ UTR of both genomic fragments was (TG)4(TGAG)5.

A similar approach was employed to isolate and characterize the S41 genomic fragments corresponding to LAR2. The two LAR2 genomic fragments cloned had a quite similar length (about 2,400 bp) and were organized into five exons (Fig. 3C).

Southern Analysis

To determine the gene copy numbers of LAR1, LAR2, and ANR in this tetraploid species, gene-specific probes were used to hybridize against genomic DNA restricted with XbaI and SacI restriction sites, which our sequence analysis showed were absent from the genes. To discriminate between LAR1 and LAR2 genes, the respective probes were PCR amplified from regions of intron sequences that did not show appreciable sequence homology between these two genes. Southern analysis showed the presence of multiple hybridizing bands for all probes (Supplemental Fig. S2). However, due to the high molecular weight of the hybridizing fragments, it was difficult to discriminate single from comigrant bands and determine the exact number of copies.

Expression Profiles of LAR and ANR Genes in Wild-Type S41 Plants

Northern analyses were first performed to assess steady-state expression levels for LAR and ANR in different S41 plant organs harvested at late-flowering stages. Given the sequence similarity between the LAR1 and LAR2 genes, probes for differentially labeling the respective mRNAs in northern analysis were sought at the 5′ end of both cDNAs, where nucleotide similarities between the two coding sequences were only about 55%. The hybridizing signals from northern blots were faint, especially when the filters were hybridized with the LAR2 probe. A clearer picture emerged when they were probed with ANR, where a transcript of approximately 1,200 bp was detected in all tissues analyzed. Importantly, this analysis demonstrated that the shortest ANR mRNAs, identified through the initial end-to-end reverse transcription (RT)-PCR analyses, were only poorly expressed (data not shown).

Due to the very low expression levels of both LAR1 and LAR2 genes and, to a lesser extent, ANR, we took advantage of sensitivity of the real-time RT-PCR analysis to monitor their expression profiles in the different tissues of S41 plants obtained after botanical fractionation. The LAR1- and LAR2-specific primer pairs were therefore designed against the respective 5′ gene ends and the primer specificity was tested against the cloned LAR1 and LAR2 cDNAs. To test for ANR expression by real-time RT-PCR, specific primers were designed against the 5′ gene end common to all ANR cDNAs cloned from L. corniculatus. Figure 4 shows the expression profiles of LAR1, LAR2, and ANR within each plant organ using elongation factor- (EF-) as a reference. The figure shows that, regardless of the tissue, ANR was always more highly expressed than the two LAR genes and the expression of LAR2 was always and significantly lower and with a narrower expression window. The presence of ANR and LAR transcripts in all plant organs analyzed is consistent with the ability of L. corniculatus to synthesize PA in these tissues (Morris and Robbins, 1997). The highest steady-state levels of ANR and LAR1 mRNAs were shown by seedless pods, whereas stems showed the lowest levels for both LAR1 and LAR2 genes (Fig. 4). When a nuclear ITS sequence was used as a constitutively expressed control gene, the relative expression patterns of the three target genes was not altered across the samples analyzed (data not shown), indicating that EF- is a reliable housekeeping gene for normalization of the expression of flavonoid genes in different plant organs.

Figure 4.
Expression of LAR1, LAR2, and ANR genes in different organs of S41 L. corniculatus. Expression was determined by real-time RT-PCR analysis using EF- as the housekeeping gene. Values with the same letter do not differ significantly (P < ...

Expression Levels of LAR and ANR in Leaves of Wild-Type and Sn Transgenic Lines

Real-time RT-PCR analyses were performed to test the steady-state levels of ANR, LAR1, and LAR2 genes in leaves of either naturally or ectopically induced PA polymorphic genotypes. Wild-type S50 and S41 plants displayed a different pattern of PA accumulation in leaves; the S50 genotype produces much lower leaf PA levels than S41, the clonal genotype of the L. corniculatus line that constitutively accumulates the highest amount of these compounds in leaves (Robbins et al., 2003). We have previously shown that, as a result of the transformation of the S50 genotype with the maize Sn gene, transgenic lines displaying contrasting leaf PA phenotypes with respect to the control were produced and that the same regulator had only subtle effects on anthocyanin accumulation, which were restricted to the leaf midrib, leaf base, and petiole, whereas other secondary metabolites were unaffected (Robbins et al., 2003). Interestingly, Sn mRNA was never detected even by real-time RT-PCR analysis in leaves of PA-suppressed lines S50/6 and S50/9, whereas Sn was expressed in the PA-enhanced lines S50/10 and S50/11 (Paolocci et al., 2005). Wild-type, β-glucuronidase (GUS), and Sn transgenic lines were grown contemporaneously in a glasshouse and RNA was isolated from young leaves. Because lines S50/9 and S50/6 are suppressed for PA accumulation and lines S50/10 and S50/11 are enhanced in PA accumulation when compared with the host nontransformed or GUS-transformed S50 genotypes (Paolocci et al., 1999; Robbins et al., 2003), data obtained from S50/10 and S50/11 and from S50/9 and S50/6 were pooled. Similarly, data from S50 nontransformed and S50 GUS transgenic lines, as well as from S41 nontransformed and S41 GUS transgenic lines, were pooled. Expression of each target gene was normalized against the levels of the EF- gene, which had been previously shown to be stably expressed in Lotus leaves, regardless of the expression of Sn (Paolocci et al., 2005). Figure 5 shows the expression levels of LAR1, LAR2, and ANR genes in control, Sn-enhanced, and Sn-suppressed lines. Expression profiles of the suppressed lines (Fig. 5, S50S) showed a significant decrease of the steady-state levels of both LAR and ANR genes with respect to the control plants. Conversely, expression of the LAR1 and ANR genes was significantly enhanced in the PA-enhanced lines (Fig. 5, S50E) when compared to the control S50C lines. In contrast to LAR1 and ANR, LAR2 was not significantly modulated at a transcriptional level when S50C, S50E, and S41C lines were compared. Overall, the levels of LAR1 and ANR, but not LAR2, correlated with the levels of PAs in the leaves of L. corniculatus.

Figure 5.
Expression of LAR1, LAR2, and ANR genes in L. corniculatus lines, polymorphic for leaf PA accumulation. Expression was determined by real-time RT-PCR analysis using EF- as the housekeeping gene. Values are average of two S transgenic lines (S50/6 ...

Functional Analysis of ANR, LAR1, and LAR2 Genes

The catalytic function of the proteins encoded by LcANR1, LcLAR1-1, LcLAR1-2, LcLAR2-1, and LcLAR2-2 was determined by heterologous expression of the genes in Escherichia coli. Recombinant proteins were incubated with suitable 14C-labeled or unlabeled precursors, here anthocyanindins and leucoanthocyanidins, in the presence of NADPH as an absolutely required cofactor. Incubation of LcANR1 protein with cyanidin (Cy) and NADPH resulted in the formation of a novel product identified as epicatechin by cochromatography with an authentic standard and specific derivatization with dimethylcinnamaldehyde (DMACA) reagent as described in Pfeiffer et al. (2006; Fig. 6). Respective incubations using recombinant LcLAR1-1 and LcLAR1-2 in a coupled assay together with recombinant DFR catalyzed the NADPH-dependent conversion of leucocyanidin (LCy) synthesized in the first step from dihydroquercetin (DHQ) by DFR to small, but clear, detectable amounts of catechin as judged by cochromatography with authentic standard after DMACA treatment or autoradiography (Fig. 6). However, incubation with recombinant proteins obtained after expression of LcLAR2-1 and LcLAR2-2 gave no detectable product formation under assay conditions used, even after enhancement of the incubation time or protein amount used per assay (Fig. 6).

Figure 6.
ANR and LAR enzyme assays with recombinant Lotus LcANR1 and LcLAR1-1, 1-2, 2-1, and 2-2. Thin-layer chomatography separation of assay extractions was done in n-buthanol:HOAc:water. A, Activity assay for LcANR1. Product identification by cochromatography ...

Quantification and Localization of PA-Containing Cells

The number and distribution of PA-containing cells was evaluated in leaves of a number of clonal and Sn transgenic lines polymorphic for the level of leaf PAs to assess whether these differences in levels of PA were consistent with a difference in the distribution of tanniferous cells. As shown in Figure 7, the genetic transformation with Sn resulted in an altered number and distribution of PA-containing cells with respect to the control, untransformed plants. In spongy mesophyll layers, the number of cells accumulating PAs was dramatically increased or decreased and correlated with expression or suppression of the introduced transgene as measured by Robbins et al. (2003). In Sn-expressing genotypes, the number of tannin-containing cells detected in the palisade mesophyll was also significantly increased. However, PA-containing cells overlying vascular tissues (vascular mesophyll) were not substantially affected in PA-enhanced or PA-suppressed genotypes or when S50 and S41 clonal recipient genotypes were compared.

Figure 7.
Number and distribution of PA-containing cells in leaves of PA polymorphic genotypes. Total leaf PA (mg/g dry weight) for the lines S41 (23.3 ± 0.3), S50 (1.92 ± 0.04), S50/10 (6.64 ± 0.11), and S50/9 (0.80 ± 0.07) determined ...


According to the most recent models, ANR and LAR participate in two separate pathways to PA biosynthesis in most species (Dixon et al., 2005). However, the specific contribution of each branch of this pathway to PA biosynthesis in different plant tissues and species is not clear. The isolation and characterization of ANS/ANR and LAR cDNAs from the same species has been reported in grapevine (Bogs et al., 2005) and apple (Malus domestica; Takos et al., 2006); however, whether the two branches of the PA pathway are controlled by the same or independent regulators has not yet been addressed. Here, we report the isolation and characterization of ANR and LAR cDNAs and corresponding genomic fragments from L. corniculatus, a legume species that accumulates epicatechin- and catechin-based PA in both reproductive and vegetative tissues. In addition, by means of real-time RT-PCR analyses on a range of plant lines polymorphic for leaf PA accumulation, we provide clear evidence that ANR and one LAR gene family are subject to coordinate transcriptional regulation mediated by a bHLH transcription factor and that their steady-state levels correlate with levels of leaf PAs.

The body of knowledge on the genetic regulation of late flavonoid biosynthetic genes has been gained mainly from studies on anthocyanin-defective mutants in several species, with the control of flavonoid biosynthesis being regulated principally at the transcriptional level. The model described for MYB-bHLH interaction in the regulation of anthocyanin biosynthesis in maize (Goff et al., 1992; Grotewold et al., 2000) is mirrored by the cooperation between TT2 and TT8 in regulating PA accumulation in the seed coats of Arabidopsis (Baudry et al., 2004). In Arabidopsis, the concerted action of TT2, TT8, and TTG1 regulators positively controls the transcription of DFR, LDOX (ANS), and ANR genes (Nesi et al., 2000, 2001; Debeaujon et al., 2003; Baudry et al., 2004). However, unlike the maize R (bHLH) gene, overexpression of TT8 is not sufficient to functionally complement the Arabidopsis ttg1 mutation and the developmental competence of tissues to accumulate BAN transcripts is rate limited both spatially and temporally by TT2 expression (Nesi et al., 2001). Conversely, in maize, tissue-specific pigmentation is caused in many cases by the tissue-specific expression of alleles of the bHLH genes rather than by the MYB gene (Ludwig and Wessler, 1990). In Arabidopsis, the redundancy of bHLH gene function in TTG1-dependent pathways has been reported and reviewed by Zhang et al. (2003) and the coexistence of anthocyanin and PA biosynthesis and the structural similarity between the regulators of these two pathways has raised the question of the molecular mechanisms leading to their specific activation (Baudry et al., 2004; Gachon et al., 2005). However, the mechanisms controlling the expression of the epicatechin and catechin branches of PA biosynthesis have not been investigated in species where both pathways are operating simultaneously.

We have previously shown that the ectopic expression of Sn, a maize bHLH gene, increases the steady-state levels of expression of DFR and ANS genes, two gene families committed also to the accumulation of anthocyanins (Paolocci et al., 2005). In Lotus plants overexpressing Sn, the levels of transcripts encoding Phe ammonia lyase, acetyl-CoA carboxylase, cinnamyl alcohol dehydrogenase, cinnamoyl-CoA reductase, chalcone synthase (CHS), chalcone isomerase (CHI), flavanone 3β-hydroxylase (FHT), and chalcone reductase were fundamentally unchanged (Hughes, 2002). This situation is similar to that in Petunia, where the introduction and expression of the maize Lc bHLH regulates late anthocyanin pathway genes, but has more limited effects upon early biosynthetic genes such as CHS, CHI, and FHT (Bradley et al., 1998). In Arabidopsis, the ectopic expression of Lc weakly induced the expression of the AHA10 gene, which codes for a proton pump ATPase required for accumulation of PA (Baxter et al., 2005), but did not induce BAN or accumulation of PA (Sharma and Dixon, 2005). Conversely, with reference to forage legumes, there is a report that alfalfa plants expressing the maize Lc accumulate anthocyanins and PAs in leaf tissues under conditions of environmental stress, such as cold or high light intensity (Ray et al., 2003).

Whereas Sn is not an ortholog of TT8, overexpression of this EGL3 (group IIIf) ortholog nevertheless appears to enhance the level of leaf PAs as well as anthocyanins in a limited number of tissues (Robbins et al., 2003). It is worth noting that TT8 and Sn present major functional differences in their ability to interact with C1 and in their requirement for TTG1 to induce high levels of BAN expression in Arabidopsis protoplasts (Baudry et al., 2004). Nevertheless, it has also been pointed out that, in these experiments, Sn can functionally replace TT8 in TTG1-dependent pathways (for review, see Quattrocchio et al., 2006).

The LcANR1 gene, selected as a representative of the ANR gene class most highly expressed in Lotus leaves, encodes a protein that will convert Cy, its expected major substrate, to epicatechin, confirming its biochemical activity as an ANR (Fig. 6). We also show that in L. corniculatus there are two LAR gene families that encode proteins that share 71% identity at the amino acid level and that, within LAR1 and ANR gene families, members are differently arranged in terms of the number and position of their introns, most likely as result of gene duplication events (Fig. 3). The phylogenetic analyses of LAR proteins from different legume species show that LcLAR1 is most similar to the LAR proteins from species accumulating PAs in their leaves, such as L. uliginosus and D. uncinatum. In contrast, the LAR2 sequences cluster with proteins cloned from species that lack appreciable PA biosynthesis in stem, leaf, and root tissues (i.e. M. truncatula and P. coccineus; Fig. 2). Although LcLAR1 and LcLAR2 retain the diagnostic amino acid stretches specific to LAR proteins (Bogs et al., 2005), only LcLAR1-1 and LcLAR1-2 were able to convert LCy to catechin (Fig. 6). The lack of activity shown by the products of the LAR2 genes in our in vitro tests suggests that the LAR2 genes encode enzymes with a different specificity to LAR1. Enzymes with different specificity might contribute to the production of basal levels of PAs in a range of plant tissues or to the production of other, as yet uncharacterized, metabolites. Because LAR2 genes are conserved in other legume species, it would appear to be unlikely that they encode inactive enzymes. Intriguingly, Bogs et al. (2005) reported the presence of two LAR genes actively transcribed in grapevine berries, but provided a functional evaluation of only one (VvLAR1). Sequences very similar to LcLAR1 and LcLAR2 have also been found in the L. japonicus genome and in expressed sequence tag databases (S. Sato, personal communication). The finding that in legumes there are two LAR gene families suggests a possible conserved duplication of the catechin pathway in distantly related plant species and that this branch of the PA pathway is more complex and diverse than hitherto expected. The detailed evaluation of the catalytic specificity of the LAR2 class of genes will form the basis of future investigations into functional characterization of the LAR gene family and the genetic components of the PA pathway.

Whether the LAR and ANR pathways are coordinately regulated by the same class of transcription factors is of importance in understanding the synthesis of PAs. To address these points, we first analyzed the expression profiles of LAR1, LAR2, and ANR genes in different organs of L. corniculatus and then we monitored their expression levels in leaves producing different levels of PAs. Real-time RT-PCR analyses allowed us to detect the transcripts of our target genes in all organs analyzed. High ANR expression levels combined with the observation that L. corniculatus mainly accumulates epicatechin extension units (Foo et al., 1996; Meagher et al., 2004) suggests that ANR may play a more significant role than LAR in PA synthesis. Analogously, in apple fruit skin, the pattern of abundance of MdANR and MdLAR transcripts was consistent with their roles in the synthesis of PAs, and it has been suggested that the initiation of PA synthesis is manly due to the flux through the ANR rather than the LAR branch of the pathway (Takos et al., 2006). The slight decrease of the transcript levels of the three target genes from young unexpanded trifoliate leaves to fully expanded leaves also correlates with the observation that L. corniculatus starts to accumulate PAs at early stages of leaf development.

Given that all three gene families (i.e. ANR, LAR1, and LAR2) are transcribed in Lotus leaves, we sought to gain information on their genetic control by comparing their relative expression levels in leaves of natural and ectopically induced PA polymorphic L. corniculatus lines. The overexpression and silencing of the maize bHLH gene Sn in the S50 background results in a significant up- and down-regulation, respectively, of leaf PAs (Paolocci et al., 1999; Robbins et al., 2003). S41 and S50 control lines also accumulate different levels of leaf PAs and the basis for this natural, rather than engineered, difference in PA levels is currently unclear. The combined analysis of ANR, LAR1, and LAR2 levels in S41 and S50 control lines, as well as in S50-Sn transgenic lines, clearly showed that the expression of ANR and LAR1 correlates with PA accumulation and with the expression/silencing of Sn in transgenic lines. These data indicate that genes of the bHLH class control the level of PAs in the leaves of L. corniculatus. In summary, a comparison of the target metabolic genes controlled by TT8 in Arabidopsis and by the ectopic expression of Sn in Lotus leaves is displayed in Figure 8.

Figure 8.
Comparison of the effects of TT8 in Arabidopsis and Sn expression in Lotus on genes of PA biosynthesis. Gray line indicates the putative activity of the protein coded by LcLAR2.

To understand whether the ranges of PA-specific mRNA transcripts and total amounts of leaf PAs correlated with the presence of detectable tannin-accumulating cells, we recorded the number and distribution of PA-containing cells in S41, S50, S50/10 (PA-enhanced), and S50/9 (PA-suppressed) lines. The increase of steady-state levels of ANR and LAR1 mRNAs in S41 and S50/10 was paralleled by an increase in the number of PA cells in palisade and spongy mesophyll tissues when compared with the S50 control line (Fig. 6). Similarly, there were less PA cells in palisade and, most significantly, in spongy mesophyll of the S50/9 line than in S50. In stark contrast, the number of PA cells adjacent to veins remained unaltered in all genotypes. Thus, it is tempting to speculate that Sn activates the expression of both ANR and LAR1 in palisade and spongy mesophyll and that the high levels of PAs in the S41 genotype are due to constitutive expression of a gene encoding a protein functionally equivalent to Sn throughout the leaf blade. The significant reduction of levels of DFR, ANS, ANR, and LAR1 transcripts and PAs in leaves of Sn-suppressed lines (Paolocci et al., 2005; this study) suggests that, in these lines, the transgene has caused the silencing of an endogenous bHLH regulator. However, unlike Sn, the bHLH protein endogenous to Lotus may also participate in controlling LAR2 expression in leaves because this target gene was significantly down-regulated in S50 Sn-suppressed lines with respect to the control S50 (Fig. 5). Therefore, these studies on L. corniculatus indicate that it is likely that the accumulation of PAs in leaves is limited by the expression of a bHLH gene.

This model is in contrast to the proposal that TTG1 is the rate-limiting regulatory factor for the expression of phenylpropanoid pathways in Arabidopsis (Zimmermann et al., 2004). The observation that the ectopic expression of AtTTG1 in S50 has no obvious chemical or developmental phenotype, whereas a range of bHLH genes modify the expression of anthocyanin and PA pathways (G. Allison and M.P. Robbins, unpublished data) lends further support to our thesis.

The successful engineering of PAs in the leaves of forage plants requires a deep understanding of the network of structural and regulatory genes involved. Here we show the presence of two LAR gene families in L. corniculatus and provide evidence that LAR1 and ANR provide two parallel and coregulated routes for PA biosynthesis in leaves and that this regulation is at least partly under the control of a bHLH transcription factor. Analysis of gene promoters and DNA-binding proteins combined with mutant analysis in the related genetic model L. japonicus will assist future studies relating to the mechanisms that control the expression of PAs in the leaves of forage legumes.


Plant Material

Lotus corniculatus genotypes S50 and S41, isolated from the cultivar Leo and characterized for their different levels of PA accumulation in leaves, have been described previously (Carron et al., 1994). S50 and S41 plants transformed with the maize (Zea mays) bHLH regulatory gene Sn or with the GUS reporter gene were obtained and characterized as reported previously (Paolocci et al., 1999, 2005; Robbins et al., 2003).

Isolation of RNA

For RNA isolation, the following organs were collected from S41 clonal plants at late-flowering stage: flowers, trifoliate unexpanded leaves, fully expanded leaves, stems, roots, flower pods, and flower pods deprived of the immature seeds (seedless fruits). For RNA isolation from leaves of Sn and GUS transgenic lines in both S50 and S41 genetic backgrounds and from control nontransformed S50 and S41 lines, plants were grown contemporaneously in a glasshouse under identical outdoor conditions and only healthy young (unexpanded) trifoliate leaves were harvested. RNA was isolated using the Nucleo Spin RNA plant isolation kit (Macherey-Nagel) according to the supplier's instructions, after which a further DNase treatment was added. The quality and quantity of RNA was verified by agarose gel electrophoresis and spectrophotometric analysis.

Gene Cloning

Cloning of ANR Gene Sequences from L. corniculatus

The primer pair Bn1F (5′-aagaaggcatgtgtgatMggtggMaStg-3′)/Bn1R (5′-tgttcaaRtactccacagtYtg-3′) and the nested primer pair Bn2F (5′-ttgcttSaMaagggttatgctgt-3′)/Bn2R (5′-ccttctttSaYaagctttYcagaagagatt-3′) were designed and used in a 25-μL volume of PCR reaction to amplify a partial ANR cDNA fragment on first-strand cDNA from S41 leaves obtained using the SMART RACE cDNA amplification kit (CLONTECH). The resulting RT-PCR fragment was sequenced and then extended toward the 5′ and 3′ ends using the above-mentioned kit by using the adaptor primers provided with the kit in combination with ANR gene-specific primers 5′-ggtgtagcgagttggaagacaagttcagagcctgct-3′ and 5′-tgtggcggagaaacaatcagcttctgg-3′ for the 5′ and 3′ RACE, respectively, according to the manufacturer's instructions. Primers were then designed proximally to the 5′ (5′-acagtgacaaaagaaaccgtata-3′) and 3′ (5′-gtgtccttgatttataattattcgc-3′) sequence ends to amplify the full-length cDNA, as well as the ANR gene from L. corniculatus, which were then cloned into the pGEM T-Easy vector (Promega).

Cloning of LAR Genes from L. corniculatus

By aligning the Desmodium uncinatum LAR sequence with the Medicago truncatula bacterial artificial chromosome clone mth2-103j7 (AC146570.4; see “Results”), the primer pairs 1LarF (5′-ggagSaactggKttcatRgg-3′) and 4larR (5′-catctatSYtgaagttaacttgRcaWccRtt-3′) and the nested primer pairs 2LarF (ggWMaRttYRtaactRaggcaag-3′) and 3larR (5′-ttgatgaaWatatcatgggtga-3′) and 5LarF (tgccttcagaRtttggKcacgatgtgg) and 6Lar (cagttgStMgagggtcgaaaatgaaca) were used to amplify LAR sequences from the S41 leaf cDNA (as above). The resulting partial cDNA fragments were sequenced and then extended toward the 5′ and 3′ ends using the SMART 5′-3′ RACE kit (CLONTECH) by using as gene-specific primer either the reverse primer 7lar (5′-tgtcttgaaagcttttgaYaatgRcagcctgga-3′) or 8Lar (5′-tagacctggctccacaggatctgccctg-3′) for the 5′ RACE or the forward primer 11Lar (5′-agcttactttgttgatggctttgatattggg-3′) or 13Lar (5′-gaacaaaaatgttcattttcgacccYc-3′) for the 3′ RACE.

The forward primers 14FxUTR (5′-agtgcaRtgctgcaaaaccatgg-3′) and 5Lar1cl (5′-acYgcataccatacactcagtaa-3′) and the reverse primers x3Lar (5′-catcaacatggggtccatatc-3′) and x3Larbis (5′-caaccatacatgcatgtggagcatatgatcc-3′) were then designed at the extreme 5′ and 3′ cDNA ends to amplify the full-length LAR cDNAs and the related genes (see “Results”), which then were cloned into the pGeM T-Easy vector.

Phylogenetic Analysis

Full-length LAR cDNA sequences from legume species were retrieved from public databases and two grapevine (Vitis vinifera) LAR sequences (Bogs et al., 2005) were used as outgroups to build a phylogenetic tree. The Clustal IV multiple sequence alignment was formed using the default parameters of the MEGA package (Kumar et al., 2004). The tree was constructed from the Clustal IV alignment using the neighbor-joining method by the MEGA program. One thousand bootstrapped datasets were used to estimate the confidence of each tree clade.

Expression Analysis of ANR, LAR1, and LAR2 in Organs of S41 Genotype and in Leaves of L. corniculatus Lines Showing Contrasting PA Phenotypes

About 20 μg of total RNA from a number of different organs of the L. corniculatus S41 genotype were run and blotted on Hybond N membrane (Amersham Biosciences) according to the supplier's instructions. The ANR probe (about 1,200 bp long) was obtained by amplifying the clone containing LcANR1-1 cDNA using the 5′ and 3′ extreme primer pairs (see above). Similarly, both the LAR1 (about 239 bp) and LAR2 (128 bp) probes were obtained by amplifying the respective cDNAs from plasmids carrying them with the primer pairs 5Lar1cl/lar1qRn (5′-gaccctggccttagaagcagat-3′) and 14FwUTR/larQ2Rn (5′-ggcggacRagcaaataggt-3′) for the former and the latter gene, respectively. RNA hybridization and washes were performed as reported in Damiani et al. (1998). For real-time RT-PCR analysis, cDNAs from the different Lotus organs analyzed and from leaves of S50, S50.121.1 (cauliflower mosaic virus-GUS control), S41, S41.121.1 (cauliflower mosaic virus-GUS control), S50/10 and S50/11 (Sn-enhanced), and S50/9 and S50/6 (Sn-suppressed) lines were synthesized from about 5 μg of total RNA using SuperScriptIII H-reverse transcriptase (Invitrogen) and 100 pmol of random hexamers (Pharmacia Biotech) according to the enzyme's supplier. The primer pairs used to amplify fragments relative to ANR, LAR1, LAR2, and ITS cDNAs were designed with the help of the OligoExpress Software (Applied Biosystem), the sequences of EF--specific primers were reported previously (Paolocci et al., 2005). The primer sequences and the size of each amplicon are given in the Supplemental Table S1. For each plant sample, two independent RNA isolations were performed and for each RNA sample two RT steps were performed and then pooled. An aliquot of 4 μL of 1:10 diluted pools of cDNA was used in the PCR reaction that was made up using the SYBER-Green PCR core mix (Applied Biosystems) according to the supplier's instructions in 25 μL of final volume in the presence of 2.5 pmol of each primer. Cycling parameters were two initial steps of 50°C for 2′ and 95°C for 2′, a two-step cycle of 95°C for 15 s and 60°C for 1′ repeated 50 times, a final step of 10′ at 60°C. Afterward, the dissociation protocol was performed. Amplifications were performed on ABI PRISM 5700 SDS apparatus (Applied Biosystems). Four replicates for each RNA sample per gene were amplified and two independent real-time PCR experiments were carried out for each RNA sample. For each transcript, the average threshold cycle (Ct) was determined. Ct is defined as the point at which fluorescence rises appreciably above the background. Standard curves for target genes and housekeeping genes (EF- and ITS) were obtained by the amplification of a serially diluted mixture of cDNA samples, with six dilution points, each one replicated four times to calculate the amplification efficiency of each primer pair. The gene quantification method based on the relative expression of the target gene versus the reference genes EF- or ITS, was adopted according to the ABI PRISM 5700 user bulletin number 2 (Applied Biosystems).

Final figures from this analysis were then treated in the same manner as quantitative traits and were statistically analyzed accordingly. ANOVA and Duncan tests for mean comparison were performed using the GLM procedure of the SAS program (SAS Institute). Due to the high value of the interactions, tissue × mRNA and genotype × mRNA, only the results of one-way ANOVA are reported.

Southern Analysis

Isolation, blotting, and Southern hybridization of genomic DNA were performed as in Damiani et al. (1999). Aliquots of approximately 8 μg (3.5 × 106 genomes) of genomic DNA were restricted independently with the following endonucleases: XbaI and SacI (New England Biolabs) according to the supplier's instructions. Digestions were run in parallel to obtain replicated filters, each one containing XbaI- and SacI-digested DNA. Each filter was hybridized with the ANR, LAR1, or LAR2 probes. The LAR1 (970 bp long) and LAR2 (650 bp long) probes were obtained by amplifying the cloned LAR1-1 and LAR2 genes (see “Results”) with the primer pairs 5′-tgcggtatttgtagagaattct-3′/5′-atgcccaaactctgaaggcaa-3′ and 5′-tgggttggggtagaggattg-3′/5′-gatgctacaatgctctctgg-3′, respectively. Similarly the ANR genomic probe was obtained by amplifying the longest ANR gene (see “Results”) with the 5′ and 3′ extreme primers (see above).

Expression of ANR and LAR Genes in Escherichia coli

The coding regions of LcANR1-1, LcLAR1-1, LcLAR1-2, LcLANR2-1, and LcLANR2-2 were amplified from the clones harboring the corresponding full-length cDNAs with Pfx50 DNA polymerase (Invitrogen Life Technology) in the presence of the primer pairs exANRF (5′-caccatggcaagcatcaagccaatttatg)/exANRR (5′-ttaattcttcagggccccctttgtcttcaa); exLAR1F (5′-caccatggtgtcaacggcagccact)/exLAR1R (5′-tcagcccatagcagtgattggcgcta); and exLAR2F (5′-caccatggcaaccaagggcagagtcctg)/exLAR2R (5′-tcaacaggaagctgtgattggcactgctt), respectively. The resulting amplicons were cloned into the pET160/GW/D/TOPO vector (Invitrogen) and sequenced to confirm their identity. Expression and protein isolation was done according the manufacturer's instructions (Invitrogen).

Recombinant Enzyme Assay

Catechin, epicatechin, and Cy were obtained from Roth. [14C]DHQ was synthesized using recombinant CHS, CHI, FHT, and flavonoid 3′-hydroxylase from various plant sources as described (Martens et al., 2003). Expression of Gerbera DFR for first-step synthesis of LCy in coupled assays with LAR was carried out as described (Martens et al., 2002). Chemical synthesis of unlabeled LCy from the respective dihydroflavonol, DHQ (TransMIT GmbH Flavonoidforschung), was performed basically according to Tanner and Kristiansen (1993).

The coupled DFR/LAR assay (500 μL in total; 0.1 m Tris-HCl buffer, pH 7.5) contained either 14C-labeled (0.03 nmol) or unlabeled DHQ (1 mmol), 25 μL 20 mmol/L NADPH, and recombinant DFR and LAR (approximately 200–500 μg total protein of each determined according to Bradford, 1976). ANR assay (500 μL in total; 0.1 m KPi buffer, pH 6.0) contained 50 μL saturated Cy solution (1 mg/mL in KPi buffer), 25 μL 20 mmol/L NADPH, recombinant ANR (approximately 300 μg total protein). After incubation at 30°C (LAR) or 45°C (ANR) for 60 min the reaction was stopped and extracted twice with 500 μL ethylacetate. The pooled and concentrated extractions were spotted on cellulose thin-layer chromatography plates (Merck) and developed in n-buthanol:HOAc:water; 6:2:1). After drying, the plates were either exposed for autoradiography (Phosphor Imager; Fuji; software package AIDA, Raytest) or sprayed with DMACA solution (0.1% in ethanol/6 n HCl; 1:1 [v/v]) to visualize the flavonoids. The identity of all substrates and products were done by cochromatography with authentic standards as described (Pfeiffer et al., 2006).

Cell Counts and Tissue Location of PA-Containing Cells in Leaves of L. corniculatus

Lotus PA cell counts were carried out on leaves of S41, S50, S50/10, and S50/9 genotypes essentially as described by Robbins et al. (2003). Trifoliate leaves were harvested randomly from selected lines and decolorized overnight in ethanol. After discarding the ethanol, samples were stained for 1 h using 0.3% (w/v) DMACA (Li et al., 1996) and then subjected to three changes of deionized water. Cells containing PA polymers stained blue and cell counts were performed by focusing on adaxial and abaxial surfaces in turn using an Olympus BH light microscope (15× magnification). Results are from 10 leaves per line and expressed ±sem.

Sequence data from this article can be found in the GenBank/EMBL data libraries under accession numbers DQ349108 to DQ349112 and DQ349114 (ANR cDNAs), DQ349113 and DQ349115 (ANR genes), DQ349100 and DQ349101 (LcLAR1 cDNAs), DQ349102 and DQ349103 (LcLAR1 genes), DQ349104 and DQ349105 (LcLAR2 cDNAs), and DQ349106 and DQ349107 (LcLAR2 genes).

Supplemental Data

The following materials are available in the online version of this article.

  • Supplemental Figure S1. Alignment of L. corniculatus ANR amino acid sequences.
  • Supplemental Figure S2. Southern analysis of LcANR, LcLAR1, and LcLAR2.
  • Supplemental Table S1. Primers used in real-time RT-PCR analysis.

Supplementary Material

[Supplemental Data]


We thank Dr. Andrea Rubini (Plant Genetics Institute) and Professor Phil Morris and colleagues in the Plant Cell Biology Team at the Institute of Grassland and Environmental Research. We also acknowledge Dr. John-Wayne Hughes for access to unpublished data and Dr. Greg Tanner and colleagues (Commonwealth Scientific and Industrial Research Organization, Canberra) and Professor Clint Chapple (Purdue University) for their assistance in the work of the Institute of Grassland and Environmental Research laboratory on PAs in legumes and Arabidopsis, respectively.


1This work was supported by the Italian Ministry of University and Research (Fondo per gli Investimenti della Ricerca di Base project “Post genomica di leguminose foraggere” code n.RBAU01BKAN to the Plant Genetics Institute) and by the Biotechnology and Biological Science Research Council (to the Institute of Grassland and Environmental Research for work in the area of plant natural products). Travel funding from Consiglio Nazionale della Ricerche to the Institute of Grassland and Environmental Research was supplied by the Consiglio Nazionale delle Ricerche Short-Term Mobility Program. This is contribution number 77 from Plant Genetics Institute-Consiglio Nazionale delle Ricerche.

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Francesco Paolocci (ti.rnc.vgi@iccoloap.ocsecnarf).

[W]The online version of this article contains Web-only data.



  • Abrahams S, Lee E, Walker AR, Tanner GJ, Larkin PJ, Ashton AR (2003) The Arabidopsis TDS4 gene encodes leucoanthocyanidin dioxygenase (LDOX) and is essential for proanthocyanidin synthesis and vacuole development. Plant J 35 624–636 [PubMed]
  • Barry TN, McNabb WC (1999) The implications of condensed tannins on the nutritive value of temperate forages fed to ruminants. Br J Nutr 81 263–272 [PubMed]
  • Baudry A, Heim MA, Dubreucq B, Caboche M, Weisshaar B, Lepiniec L (2004) TT2, TT8 and TTG1 synergistically specify the expression of BANYULS and proanthocyanidin biosynthesis in Arabidopsis thaliana. Plant J 39 366–380 [PubMed]
  • Baxter IR, Young JC, Armstrong G, Foster N, Bogenschutz N, Cordova T, Peer WA, Hazen SP, Murphy AS, Harper JF (2005) A plasma membrane H+-ATPase is required for the formation of proanthocyanidins in the seed coat endothelium of Arabidopsis thaliana. Proc Natl Acad Sci USA 102: 2649–2654; erratum Baxter IR, Young JC, Armstrong G, Foster N, Bogenschutz N, Cordova T, Peer WA, Hazen SP, Murphy AS, Harper JF (2005) Proc Natl Acad Sci USA 102: 5635 [PMC free article] [PubMed]
  • Bogs J, Downey MO, Harvey JS, Ashton AR, Tanner GT, Robinson SP (2005) Proanthocyanidin synthesis and expression of genes encoding leucoanthocyanidin reductase and anthocyanidin reductase in developing grape and grapevine leaves. Plant Physiol 139 652–663 [PMC free article] [PubMed]
  • Bradford MM (1976) A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein dye binding. Anal Biochem 72 248–254 [PubMed]
  • Bradley JM, Davies KM, Deroles SC, Bloor SJ, Lewis DH (1998) The maize Lc regulatory gene up-regulates the flavonoid biosynthetic pathway of petunia. Plant J 13 381–392
  • Carron TR, Robbins MP, Morris P (1994) Genetic modification of condensed tannin biosynthesis in Lotus corniculatus. 1. Heterologous antisense dihydroflavonol reductase down-regulates tannin accumulation in “hairy root” cultures. Theor Appl Genet 8 1006–1015 [PubMed]
  • Damiani F, Paolocci F, Cluster PD, Arcioni S, Tanner GJ, Joseph RG, Li YG, de Majnik J, Larkin PJ (1999) The maize Sn regulatory transgene alters proanthocyanidin production in Lotus corniculatus in some tissues by transactivation and in others by co-suppression of an endogenous regulator. Aust J Plant Physiol 26 159–169
  • Damiani F, Paolocci F, Consonni G, Crea F, Tonelli C, Arcioni S (1998) A maize anthocyanin transactivator induces pigmentation in hairy roots of dicotyledonous species. Plant Cell Rep 17 339–344
  • Damiani F, Paolocci F, Turchetti V, Arcioni S (2000) Toward the identification of the genes for the synthesis of condensed tannins in forage legumes. In GE de Vries, K Metzlaff, eds, Phytosphere ′99 Highlights on European Plant Biotechnology Research and Technology Transfer. Elsevier, Amsterdam, pp 257–263
  • Debeaujon I, Léon-Kloosterziel KM, Koornneef M (2000) Influence of the testa on seed dormancy, germination, and longevity in Arabidopsis. Plant Physiol 122 403–414 [PMC free article] [PubMed]
  • Debeaujon I, Nesi N, Perez P, Devic M, Grandjean O, Caboche M, Lepiniec L (2003) Proanthocyanidin-accumulating cells in Arabidopsis testa, regulation of differentiation and role in seed development. Plant Cell 15 2514–2531 [PMC free article] [PubMed]
  • Devic M, Guilleminot J, Debeaujon I, Bechtold N, Bensaude E, Koornneef M, Pelletier G, Delseny M (1999) The BANYULS gene encodes a DFR-like protein and is a marker of early seed coat development. Plant J 19 387–398 [PubMed]
  • Dixon RA, Xie D, Sharma SB (2005) Proanthocyanidins—a final frontier in flavonoid research? New Phytol 165 9–28 [PubMed]
  • Foo LY, Newman R, Waghorn G, McNabb WC, Ulyatt MJ (1996) Proanthocyanidins from Lotus corniculatus. Phytochemistry 41 617–624
  • Gachon CMM, Langlois-Meurinne M, Henry Y, Saindrenan P (2005) Transcriptional co-regulation of secondary metabolism enzymes in Arabidopsis: functional and evolutionary implications. Plant Mol Biol 58 229–245 [PubMed]
  • Goff SA, Cone KC, Chandler VL (1992) Functional analysis of the transcriptional activator encoded by the maize B gene: evidence for a direct functional interaction between two classes of regulatory proteins. Genes Dev 6 864–865 [PubMed]
  • Grotewold E, Sainz MB, Tagliani L, Hernandez JM, Bowen B, Chandler VL (2000) Identification of the residues in the Myb domain of maize C1 that specify the interaction with bHLH cofactor R. Proc Natl Acad Sci USA 97 13579–13584 [PMC free article] [PubMed]
  • Harborne JB (1988) The Flavonoids: Advances in Research Since 1988. Chapman and Hall, London
  • Hughes JW (2002) Resource allocation into secondary end products in the forage legume Lotus corniculatus L. (bird's foot trefoil). PhD thesis. University of Wales, Aberystwyth, Wales
  • Kingston-Smith AH, Thomas HM (2003) Strategies of plant breeding for improved rumen function. Ann Appl Biol 142 13–24
  • Kumar S, Tamura K, Nei M (2004) MEGA3: integrated software for molecular evolutionary genetics analysis and sequence alignment. Brief Bioinform 5 150–163 [PubMed]
  • Lepiniec L, Debeaujon I, Routaboul J-M, Baudry A, Pourcel L, Nesi N, Caboche M (2006) Genetics and biochemistry of seed flavonoids. Annu Rev Plant Biol 57 405–430 [PubMed]
  • Li YG, Tanner GJ, Larkin PJ (1996) The DMACA-HCl protocol and the threshold proanthocyanidin content for bloat safety in forage legumes. J Sci Food Agric 70 89–101
  • Ludwig SR, Wessler SR (1990) Maize R gene family: tissue-specific helix-loop-helix proteins. Cell 62 849–851 [PubMed]
  • Marles MA, Ray H, Gruber MY (2003) New perspectives on proanthocyanidin biochemistry and molecular regulation. Phytochemistry 64 367–383 [PubMed]
  • Martens S, Metho M, Teeri T, Forkmann G (2002) Heterologous expression of dihydroflavonol 4-reductase from Gerbera hybrids. FEBS Lett 531 453–458 [PubMed]
  • Martens S, Wellmann F, Britsch L, Forkmann G, Matern U, Lukacin R (2003) Divergent evolution of flavonoid 2-oxoglutarate-dependent dioxygenases in parsley. FEBS Lett 544 93–98 [PubMed]
  • Meagher LP, Lane G, Sivakumaran S, Tavendale MH, Fraser K (2004) Characterisation of condensed tannins from Lotus species by thiolytic degradation and electrospray mass spectrometry. Anim Feed Sci Technol 117 151–163
  • Morris P, Robbins MP (1997) Manipulating condensed tannins in forage legumes. In BD McKersie, DCW Brown, eds, Biotechnology and the Improvement of Forage Legumes. CAB International, Oxford, pp 147–173
  • Nesi N, Debeaujon I, Jond C, Pelletier G, Caboche M, Lepiniec L (2000) The TT8 gene encodes a basic helix-loop-helix domain protein required for expression of DFR and BAN genes in Arabidopsis siliques. Plant Cell 12 1863–1878 [PMC free article] [PubMed]
  • Nesi N, Jond C, Debeaujon I, Caboche M, Lepiniec L (2001) The Arabidopsis TT2 gene encodes an R2R3 MYB domain protein that acts as key determinant for proanthocyanidin accumulation in developing seed. Plant Cell 13 2099–2114 [PMC free article] [PubMed]
  • Paolocci F, Bovone T, Tosti N, Arcioni S, Damiani F (2005) Light and an exogenous transcription factor qualitatively and quantitatively affect the biosynthetic pathway of condensed tannins in Lotus corniculatus leaves. J Exp Bot 56 1093–1103 [PubMed]
  • Paolocci F, Capucci R, Arcioni S, Damiani F (1999) Birdsfoot trefoil, a model for studying the synthesis of condensed tannins. In GG Gross, RW Hemingway, T Yoshida, eds, Plant Polyphenols 2. Chemistry, Biology, Pharmacology, Ecology. Kluwer Academic/Plenum Press, New York, pp 343–356
  • Pfeiffer J, Kühnel C, Brandt J, Duy D, Punyasiri PAM, Forkmann G, Fischer TC (2006) Biosynthesis of flavan 3-ols by leucoanthocyanidin 4-reductases and anthocyanidin reductases in leaves of grape (Vitis vinifera L.), apple (Malus × domestica Borkh.) and other crops. Plant Physiol Biochem 44 323–334 [PubMed]
  • Quattrocchio F, Baudry A, Lepiniec L, Grotewold E (2006) The regulation of flavonoid biosynthesis. In E Grotewold, ed, The Science of Flavonoids. Springer Science, New York, pp 97–122
  • Ray Y, Yu M, Auser P, Blahut-Beatty L, McKersie B, Bowley S, Westcott N, Coulman B, Lloyd A, Gruber MY (2003) Expression of anthocyanins and proanthocyanidins after transformation of alfalfa with maize Lc. Plant Physiol 132 1448–1463 [PMC free article] [PubMed]
  • Robbins MP, Paolocci F, Hughes JW, Turchetti V, Allison G, Arcioni S, Morris P, Damiani F (2003) Sn, a maize bHLH gene, transactivates anthocyanin and condensed tannin pathways in Lotus corniculatus. J Exp Bot 54 239–248 [PubMed]
  • Sharma SB, Dixon RA (2005) Metabolic engineering of proanthocyanidins by ectopic expression of transcription factors in Arabidopsis thaliana. Plant J 44 62–75 [PubMed]
  • Takos AM, Ubi BE, Robinson S, Walker AR (2006) Condensed tannin biosynthesis genes are regulated separately from other flavonoid biosynthesis genes in apple fruit skin. Plant Sci 170 487–499
  • Tanner GJ (2004) Condensed tannins. In KM Davies, ed, Plant Pigments and Their Manipulation. Annual Plant Reviews, Vol 12. Blackwell Publishing-CRC Press, Boca Raton, FL, pp 150–184
  • Tanner GJ, Francki KT, Abrahams S, Watson JM, Larkin PJ, Ashton AR (2003) Proanthocyanidin biosynthesis in plants: purification of legume leucoanthocyanidin reductase and molecular cloning of its cDNA. J Biol Chem 278 31647–31656 [PubMed]
  • Tanner GJ, Kristiansen KN (1993) Synthesis of 3,4-cis-[3H]leucocyanidin and enzymatic reduction to catechin. Anal Biochem 209 274–277 [PubMed]
  • Tanner GJ, Moate PJ, Davis LH, Laby RH, Li YG, Larkin PJ (1995) Proanthocyanidins (condensed tannin) destabilize plant protein foams in a dose dependent manner. Aust J Agric Res 46 1101–1109
  • Tohge T, Matsui K, Ohmi-Takagi M, Yamazaki M, Saito K (2005) Enhanced radical scavenging activity of genetically modified Arabidopsis seeds. Biotechnol Lett 27 297–303 [PubMed]
  • Winkel-Shirley B (2001) Flavonoid biosynthesis: a colorful model for genetics, biochemistry, cell biology, and biotechnology. Plant Physiol 126 485–493 [PMC free article] [PubMed]
  • Xie DY, Sharma SB, Paiva NL, Ferreira D, Dixon RA (2003) Role of the anthocyanidin reductase, encoded by BANYULS in plant flavonoid biosynthesis. Science 299 396–399 [PubMed]
  • Zhang F, Gonzalez A, Zhao M, Payne CT, Lloyd A (2003) A network of redundant bHLH proteins functions in all TTG-1 dependent pathways of Arabidopsis. Development 130 4859–4869 [PubMed]
  • Zimmermann IM, Heim MA, Weisshaar B, Uhrig JF (2004) Comprehensive identification of Arabidopsis thaliana MYB transcription factors interacting with R/B-like BHLH proteins. Plant J 40 22–34 [PubMed]

Articles from Plant Physiology are provided here courtesy of American Society of Plant Biologists
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem chemical compound records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records. Multiple substance records may contribute to the PubChem compound record.
  • MedGen
    Related information in MedGen
  • Nucleotide
    Primary database (GenBank) nucleotide records reported in the current articles as well as Reference Sequences (RefSeqs) that include the articles as references.
  • Protein
    Protein translation features of primary database (GenBank) nucleotide records reported in the current articles as well as Reference Sequences (RefSeqs) that include the articles as references.
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem chemical substance records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records.
  • Taxonomy
    Taxonomy records associated with the current articles through taxonomic information on related molecular database records (Nucleotide, Protein, Gene, SNP, Structure).
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...