Index INTRODUCTION MATERIALS AND METHODS Plant materials SCAR primer design Optimization of PCR conditions and digestion of PCR products DNA sequencing HPLC analyses of anthocyanins RESULTS Assignment of genes involved in flavonoid biosynthesis to chromosomes 1) CHS 2) CHI 3) F3H 4) DFR 5) ANS Chromosomal assignment of gene related to glucosylation of anthocyanidin based on HPLC analysis DISCUSSION References

INTRODUCTION

Onions (Allium cepa L. Common onion group) are known to contain a large amount of flavonoids, which are present in the colored scales of the edible part (Chu et al. 2000; Bahorun et al. 2004). Quercetin-4’-glucoside and 3,4’-diglucoside account for 80% of all flavonoids in onions (Tsushida and Suzuki 1995; Rhodes and Price 1996). Red onions also contain anthocyanins, which are the largest subclass of plant flavonoids. The major anthocyanins in red onions are cyanidin 3-(6”-malonylglucoside), cyanidin 3-(6”-malonyl-3”-glucosylglucoside), cyanidin 3-(3”-glucosylglucoside), and cyanidin 3-glucoside (Terahara et al. 1994; Fossen et al. 1996; Donner et al. 1997). Cyanidin 3-(6”-malonylglucoside) constitutes more than 50% of the total anthocyanin content in three different cultivars of red onion (Fossen et al. 1996). Although the level of the anthocyanin content of red onions is <10% with respect to flavonols such as quercetin (Rhodes and Price 1996), the simultaneous intake of flavonols and anthocyanins could be regarded as particularly healthy (Gennaro et al. 2002).

The study of a complete set of Japanese bunching onion (Allium fistulosum L.) – shallot (A. cepa L. Aggregatum group) monosomic addition lines (Shigyo et al. 1996) has revealed that the genes related to the production of flavonoids are located on chromosome 5A of shallots (Shigyo et al. 1997), and there is also a possibility that their genes are located on other chromosomes. Masuzaki et al. (2006) demonstrated that the F3’H gene is located on chromosome 7A, as shown by HPLC analyses using A. fistulosum – shallot multiple alien addition lines carrying chromosome 5A and the other chromosome(s) and that the F3’H and FLS genes are located, respectively, on chromosomes 7A and 4A, as shown by SCAR analyses using the monosomic additions. Recently, studies of the relationship between bulb color and the structural genes in the flavonoid biosynthesis pathway have revealed that white onions resulted from a reduced level of transcripts of CHS genes (Kim et al. 2004c, 2005c), gold onions from mutation of the CHI gene (Kim et al. 2004c), yellow onions from inactivation of DFR gene transcription (Kim et al. 2004b), and pink onions from mutation of the ANS gene (Kim et al. 2004a, 2005d). One sequence (GenBank™ accession number, CF445052) was selected from 10 onion sequences similar to the sequence of 3GT and used for examining the expression level of 3GT (Kim et al. 2004a, 2004c, 2005c, 2005d), as the deduced amino acid sequence showed the highest homology with 3GTs of other plant species (Kim et al. 2004a). However, whether the presumed protein transfers a glucoside residue from UDP-glucose to the 3-OH position of anthocyanidin has not been demonstrated. HPLC analysis of the above-mentioned A. fistulosum – shallot multiple additions is a more valid method for the determination of the chromosomal location than SCAR analysis in the monosomic additions, because no 3GT gene has been cloned yet in A. cepa, nor the copy number described. In A. cepa, SCAR analyses have so far been used for genetic mapping of alliinase (van Heusden et al. 2000a) and chromosomal assignment of 6GFFT (McCallum et al. 2006). To utilize flavonoids effectively for human health and to achieve uniformity of the bulb colors in breeding onion cultivars, it is necessary to accumulate more genetic data regarding flavonoid biosynthesis.

We report here the chromosomal locations of seven genes involved in flavonoid biosynthesis, i.e., CHS-A, CHS-B, CHI, F3H, DFR, ANS, and 3GT, which we assigned by using A. fistulosum – shallot alien chromosome additions.

MATERIALS AND METHODS

Plant materials

Two complete sets of A. fistulosum – shallot monosomic additions (Shigyo et al. 1996) were used to assign a segment of the coding region for flavonoid genes to the chromosomes of shallots. The sets were analyzed together with their parents: the shallot (A. cepa L. Aggregatum group) and the Japanese bunching onion (A. fistulosum L. cv. ‘Kujo Hoso’). Total genomic DNA was isolated from fresh leaf tissue using a mini prep DNA-isolation method (van Heusden et al. 2000a).

A. fistulosum – shallot multiple additions carrying chromosome 5A and the other chromosome(s) (2n = 2x + 2 – 2x + 7 = 18 – 23) (Masuzaki et al. 2006) were subjected to HPLC analyses in which we detected anthocyanins to reveal the chromosomal locations of the genes related to the anthocyanin biosynthesis.

SCAR primer design

Outside- and inside-primer sets to amplify a segment of the gene-encoding enzymes from CHS to DFR, with the exception of F3’H, in a possible flavonoid biosynthesis pathway in A. cepa, were designed with the software GENETYX^® 6.1.3 (Genetyx, Tokyo, Japan) based on the GenBank™ accession numbers of AY221244 (a full-length cDNA sequence of CHS-A), AY221245 (a full-length cDNA sequence of CHS-B), AY700850 (a full-length genomic sequence of CHI), AY221246 (a full-length cDNA sequence of F3H), and AY221250 (a full-length genomic sequence of DFR) (Table 1). The primer set developed by Kim et al. (2005a) was used for ANS. The primer set was as follows: forward, 5’ – TTT GCT CGA TCG TTT AGC RGA AGA AGA – 3’; reverse, 5’ – TGA GGA TGA TGA CAA AGT TAG CGG AGC A – 3’.

View Details

Table 1. Primer sequences and amplification conditions used in this study

Optimization of PCR conditions and digestion of PCR products

All PCR amplifications were done with template DNA [about 100 ng of genomic DNA for the first PCR and 5 μl of diluted template DNA (the products of the first PCR diluted 1/50 with sterile water) for the second PCR, except in the cases of CHI and F3H, in which case, 5 μl of the first PCR product was used for the second PCR], each of the primers at 1 μM, 0.25 mM dNTPs, 1 × Ex Taq buffer, and 0.5 U of Ex Taq polymerase (Takara, Shiga, Japan) in a volume of 25 μl. All PCRs were performed as follows: an initial denaturation of 3 min at 94°C; 25 cycles of PCR amplification (a 1-min denaturation at 94°C, a 1-min annealing, and a 1-min primer extension at 72°C); and a final extension for 7 min at 72°C on a Program Thermal Cycler iCycler™ (Bio-Rad, Hercules, CA, USA). The ramp times were carried out in the default conditions that adjusted temperatures at the maximum ramp rate with the minimum ramp time. The annealing temperature gradient included the temperatures between 54.0° and 68.0°C. The optimum annealing temperatures of the primer sets are given in Table 1. The second PCR products were separated by electrophoresis on 2.0% agarose gels. The analysis for ANS was done according to the procedure of Kim et al. (2005a).

If the PCR products were monomorphic on agarose gels, they were subjected to restriction analysis. Nine microliters of the second PCR products were incubated for 2 h at 37°C in a volume of 15 μl using 2 U of a restriction enzyme, and subsequently resolved by 5% denaturing polyacrylamide gel electrophoresis (PAGE) with silver staining according to the procedure of Martin et al. (2005). Restriction digestion with two enzymes, BamH I and Hae III (Toyobo, Osaka, Japan), was used in an attempt to generate polymorphisms. The two enzymes were selected using GENETYX^® 6.1.3, based on the onion sequence, which corresponded to the fragment of shallot amplified using the primer sets used in this study.

DNA sequencing

The reliability of PCR amplification was evaluated by direct sequencing of the PCR products. Homology searches were performed in the GenBank™ database using the BLAST program.

HPLC analyses of anthocyanins

The outer scale of the basal part of the leaf sheath was collected from the 14 multiple additions, along with the monosomic addition of A. fistulosum carrying shallot chromosome 5A (FF + 5A), and shallots, used as controls. Anthocyanins were extracted by the method described by Yoshitama and Yamaguchi (2004) with modification. One gram of the scale was extracted with 10 ml of 50% acetic acid for 3 days at 4°C. The extract was concentrated about 50-fold by vacuum concentration. The concentrate was filtered through a 0.5-μm filter (Katayama Chemical, Osaka, Japan) to remove pigments and used for the HPLC analyses. The HPLC separation was carried out at a flow rate of 0.8 ml per min using a LiChrospher C18 column (Merck, Darmstadt, Germany) of dimensions 4 × 250 mm. The column was equilibrated in 75% solvent A (1.5% phosphate) and 25% solvent B (1.5% phosphate, 20% acetic acid, and 25% acetonitrile). After an aliquot of filtrate (20 μl) was injected into the HPLC apparatus, a linear gradient up to 85% solvent B was applied during 40 min to elute the anthocyanin compounds (monitored at 520 nm). The apparatus for the HPLC consisted of an L-7100 pump (Hitachi, Tokyo, Japan), L-7300 column oven (Hitachi), L-7420 UV/visible detector (Hitachi), and L-7610 Degasser^® (Hitachi) to degas the filtrate.

RESULTS

Assignment of genes involved in flavonoid biosynthesis to chromosomes

1) CHS

The primer sets for CHS-A amplified a single PCR product with the template DNA of the shallot but not that with A. fistulosum (Fig. 1a). The amplicon of CHS-A was the same size as expected. An amplicon of the same size as in the shallot was observed only in the monosomic addition FF + 2A. The amplicon of shallot was sequenced. The BLASTn of the sequences showed 100% homology with the original sequence (AY221244). The primer set specific for the CHS-A gene of A. cepa designed in this study could not amplify a CHS-A gene of A. fistulosum. In the monosomic additions, the primer set could not amplify a CHS-A gene in the monosomic additions, except for FF + 2A. These results indicated that only FF + 2A had the CHS-A gene derived from the shallot. Therefore, we assigned the CHS-A gene of the shallot to chromosome 2A.

View Details

Fig. 1. Electropherograms revealing the chromosomal locations of flavonoid biosynthesis genes in shallot. FF, A. fistulosum; AA, shallot; 1A-8A, two complete sets of A. fistulosum - shallot monosomic addition lines. M, molecular size marker (100-bp DNA ladder). (a) CHS-A. (b) CHS-B. Arrows indicate shallot-specific bands. (c) CHI. (d) F3H. (e) DFR. (f) ANS.

The primer sets of CHS-B amplified a single fragment of the expected size with the DNA of the shallot, and a comigrating fragment in A. fistulosum as well (data not shown). When the fragments of A. fistulosum and shallot were digested with the enzyme BamH I, they both showed the same digestion patterns on 5% denaturing PAGE (data not shown). After both PCR products were digested with Hae III, a polymorphism between A. fistulosum and the shallot was detected via 5% denaturing PAGE. In the two complete sets of the monosomic additions, the same restriction fragment length polymorphism (RFLP) patterns as in the shallot were present only in FF + 4A (Fig. 1b). BLASTn of the sequence data from the amplicon of shallot had homology of 99% with the original sequence (AY221245). The primer set specific to the CHS-B gene of A. cepa amplified CHS-B genes in both A. fistulosum and shallot. The Hae III – RFLP patterns showed that FF + 4A had the CHS-B gene from the shallot genome and that the other monosomic additions had only a CHS-B gene from A. fistulosum. These results revealed that the CHS-B gene of the shallot was located on chromosome 4A.

2) CHI

The primer sets of CHI amplified a single fragment of the expected size in the shallot and FF + 3A (Fig. 1c). The BLASTn of the amplicon showed 100% homology with the original sequence (AY700850). The four different alleles of the onion CHI gene were designated CHI-b (AY700850), CHI-t (AY700851), CHI-h (AY700852), and CHI-g (AY700853), and unusual gold-colored onions result from CHI-g, which is the mutant CHI allele (Kim et al., 2004c). There was a possibility that the CHI gene of the shallot used in this study was the functional gene, as the sequence of CHI obtained in this study showed 100% homology (no gaps) with CHI-b and had a homology of 98% (one gap) to CHI-g being the mutant allele. These results revealed that the activated CHI gene might be located on chromosome 3A.

3) F3H

The BLASTn of a single fragment amplified from the DNAs of the shallot and FF + 3A using the primer sets of F3H (Fig. 1d) showed 99% homology with the original sequence (AY221246). These results revealed that the F3H gene was located on chromosome 3A. The amplicon of F3H was larger than expected. Alignment between the original sequence (AY221246) and our sequence data indicated that a 159-bp indel existed as an intron (data not shown). The fact that the amplicon was larger than the expected size resulted from this indel. This sequence has been deposited with GenBank™ (Accession number, DQ394303).

4) DFR

The primer sets for DFR amplified a fragment of the expected size in the shallot and FF + 7A (Fig. 1e). The result of the homology search of the amplicon had a higher homology of 99% to a DFR gene that produces red pigment (AY221250) compared with 94% to a DFR pseudogene that does not produce red coloring (AY221251) (Kim et al. 2005b). These results indicated that the DFR gene that directs the production of red coloring might be located on chromosome 7A.

5) ANS

The primer set for ANS described by Kim et al. (2005a) amplified a single PCR product with the DNAs of the shallot and FF + 4A but not with A. fistulosum or the other monosomic additions (Fig. 1f). The size of these amplicons was approximately 430-bp. This primer set, which was designed to distinguish between the alleles of the ANS gene, amplified an 828-bp fragment in the pink allele and a 428-bp fragment in the red (Kim et al. 2005a). This result revealed that the red allele of the ANS gene was located on chromosome 4A.

Chromosomal assignment of gene related to glucosylation of anthocyanidin based on HPLC analysis

Several peaks attributable to anthocyanins were observed in the HPLC profiles of the shallot and FF + 5A, and three compounds were identified as cyanidin 3-O-glucoside (Cy 3-glc), cyanidin 3-O-(6”-malonylglucoside) [Cy 3-(6”-Maglc)], and cyanidin aglycone (Cy) by their retention times and absorption spectra (Fig. 2a). In the shallot, all three compounds appeared, and the content of Cy 3-(6”-Maglc) was the highest. In FF + 5A, almost the same contents of Cy 3-(6”-Maglc) and Cy were present, but Cy 3-glc was not detected. We divided the 14 multiple additions (Fig. 2a) into two groups according to the component ratio of Cy (Fig. 2b). U121 (2n = 2x + 7 = FF + 1A + 2A + 3A + 5A + 6A + 7A + 8A) showed the highest component ratio of Cy (21.8%) among the five multiple additions possessing 23 chromosomes, including seven out of eight kinds of extrachromosomes. We presumed that this phenomenon was caused by an absence of one extrachromosome carrying the genes for glucosylation of Cy. Therefore, we made a demarcation at a 21.8% component ratio of Cy to evaluate whether the multiple additions had a little Cy, i.e., whether they had the genes for glucosylation of Cy. Group I had a ratio of Cy less than 21.8%, while Group II had a ratio of Cy more than 21.8%. Comparison of the genomic constitution between the two groups revealed that chromosome 4A was found only in Group I. The gene responsible for the glucosylation of anthocyanidin is 3GT, which encodes the enzyme that converts anthocyanidin into anthocyanin 3-O-glucoside (Kamsteeg et al. 1978; Holton and Cornish 1995) (Fig. 3). Therefore, a 3GT gene might be located on chromosome 4A.

View Details

Fig. 2. High-performance liquid chromatography profiles of acetic acid extracts (detection: 520 nm) (a) and component ratio of anthocyanins (b) in shallot (18-5), a monosomic addition FF + 5A (24) and multiple alien addition lines. (a) Cy 3-glc, cyanidin 3-O-glucoside (Rt 12.14 min); Cy 3-(6”-Maglc), cyanidin 3-O-(6”-malonylglucoside) (Rt 18.24 min); Cy, cyanidin aglycone (Rt 23.23 min). (b) The genomic constitution shows the shallot chromosome which each of the plants possessed. C1 and C2 in the group represent control plants carrying all the chromosomes of shallot and only chromosome 5A, respectively. Group I has Cy < 21.8% as the component ratio of anthocyanins, and II has Cy > = 21.8%.

View Details

Fig. 3. Putative flavonoid biosynthesis pathway in A. cepa and chromosomal locations of the gene for each enzyme. The numbers in parentheses stand for the chromosome numbers of shallot on which each enzyme gene is located. The italics show that the chromosomal locations were determined by SCAR analyses. Asterisks indicate the genes assigned to chromosomes by Masuzaki et al. (2006). CHS, chalcone synthase; CHI, chalcone isomerase; F3H, flavanone 3-hydroxylase; F3’H, flavonoid 3’-hydroxylase; DFR, dihydroflavonol 4-reductase; ANS, anthocyanidin synthase; 3GT, UDP glucose:flavonoid 3-O-glucosyltransferase; 3MaT, malonyl coenzyme-A:anthocyanidin 3-O-glucoside-6”-O-malonyltransferase; FLS, flavonol synthase.

DISCUSSION

The eight genes coding the enzymes from CHS to ANS in the flavonoid biosynthesis pathway in shallots were assigned to four different chromosomes in the previous and present studies (Fig. 3). Shigyo et al. (1997) reported that the flavonoid and anthocyanin compounds were produced only in FF + 5A out of the set of monosomic additions, but none of the flavonoid genes have thus far been allocated to chromosome 5A. In a previous report (Masuzaki et al. 2006), we demonstrated that a regulatory gene or CHS might be located on 5A. This study provides considerable support for the possibility that the regulatory gene might be located on 5A. Although the level of expression of the F3’H gene of A. fistulosum may be low, as indicated by an excess content of kaempferol compared with quercetin in FF + 5A (Shigyo et al. 1997; Masuzaki et al. 2006), the fact that flavonoids, including anthocyanins, were produced in FF + 5A supports the notion of a backup expression of the flavonoid genes of A. fistulosum in FF + 5A. Therefore, the regulatory gene for the flavonoid biosynthesis must be inactivated in A. fistulosum. Kim et al. (2005c) mentioned that the regulatory gene of the flavonoid biosynthesis in onions seemed to control the expression of all the structural genes. It is important to identify the structural genes controlled by the regulatory gene in A. cepa. We attempted to determine the copy number of the structural genes by Southern blotting with non-radioisotopes (data not shown), but no signals were detected under the genome background of 15.290 Mbps per 1C nucleus in A. cepa (Arumuganathan and Earle 1991). This result indicated the possibility that the structural genes involved in flavonoid biosynthesis in shallot as well as in other plants might be present as a single copy gene or a small multigene family (Sparvoli et al. 1994; Pelletier et al. 1997; Woo et al. 1999; Moriguchi et al. 2001).

Anthocyanins exist as glucosylated, acylated, and/or methylated forms in vacuoles (Strack and Wray 1994). It has been proposed that aliphatic acylation, including malonylation, of anthocyanins enhances the pigment solubility in water, protects glycosides from enzymatic degradation, stabilizes anthocyanin structures, or enhances the uptake of anthocyanins into vacuoles (Heller and Forkmann 1994). In addition, Suzuki et al. (2002) reported that malonylation of anthocyanins also prevented anthocyanins from the attack of β-glucosidase and that malonylation should enhance the pigment stability and thereby play a very important role in maintaining the color of anthocyanins in dahlia flowers. Among the three cyanidin compounds identified in this study, Cy 3-(6”-Maglc) in the outer scale of the shallot accounted for nearly 90% of the cyanidin, and among five anthocyanin compounds, two kinds of malonylated cyanidin derivatives in the skin of the shallot constituted approximately 80% of the anthocyanin (Shigyo et al. 1997). Shallots, like red onions, contain a large amount of malonylated anthocyanins, and malonylation of anthocyanins might also serve as a strategy for pigment stabilization in A. cepa. Cy 3-(6”-Maglc) is synthesized from Cy 3-glc by 3MaT (Yamaguchi et al. 1999; Suzuki et al. 2002) (Fig. 3). The chromosomal locations of malonylation genes will be determined by SCAR analysis in the future.

It seems that at least five different chromosomes, i.e., chromosomes 2A, 3A, 4A, 5A, and 7A, are required for the expression of red bulb color in shallot. Unfortunately, the four multiple additions (S₂27, U82, U120, and U138) that fulfilled this condition did not always show a red outer scale (data not shown). S₂27 and U138 had a red outer scale, but U120 and U82 had light red and yellow scales, respectively. The scale color of the latter two plants might result from a lower component ratio of Cy 3-(6”-Maglc) than those of the former two plants. However, in group II, which did not fulfill the prerequisite condition, the outer scale of U83 turned red. These results indicated that the color of the scale in the multiple additions might not be determined by a simple combination of extrachromosome(s) of the shallot. Consequently, to identify the major factor causing the red scale in the multiple additions, investigations must be made into correlations among the scale color, genomic constitution, flavonoid and anthocyanin constitution, and expression level of the flavonoid biosynthesis genes derived from shallot and A. fistulosum in each of the multiple additions.

We assigned all structural genes from CHS to ANS involved in flavonoid biosynthesis to the chromosomes of A. cepa. In A. cepa, linkage maps based on AFLPs (van Heusden et al. 2000a, 2000b) and SSRs, SNPs, RAPDs, and RFLPs (King et al. 1998; Martin et al. 2005) have been produced. The flavonoid genes whose chromosomal locations have been revealed will be represented on these two maps hereafter, as that will be indispensable for further breeding of onion.

The authors sincerely thank Mr. Toyoshi Iwata, Shippo Seed Co., LTD, Kagawa, Japan, for providing financial support for this study. We are also grateful to Dr. Takejiro Takamura of Kagawa University for technical suggestions on HPLC analysis. We would like to acknowledge the technical expertise of The DNA Core Facility of the Center for Gene Research, Yamaguchi University, supported by a grant-in-aid from the Ministry of Education, Science, Sports and Culture of Japan.