Crop Science
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 16 July 2007
Published in Crop Sci 47:S-113-S-124 (2007)
© 2007 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Supplemental Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zabala, G.
Right arrow Articles by Vodkin, L. O.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Zabala, G.
Right arrow Articles by Vodkin, L. O.
Agricola
Right arrow Articles by Zabala, G.
Right arrow Articles by Vodkin, L. O.
Related Collections
Right arrow Soybean
Right arrow Cell Biology & Molecular Genetics

ORIGINAL RESEARCH

A Rearrangement Resulting in Small Tandem Repeats in the F3'5'H Gene of White Flower Genotypes Is Associated with the Soybean W1 Locus

Gracia Zabala and Lila O. Vodkin*

Dep. of Crop Sciences, Univ. of Illinois, Urbana, IL 61801

* Corresponding author (l-vodkin{at}uiuc.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 
Soybean [Glycine max (L.) Merr.] flower pigmentation is controlled by six loci, W1, W2, W3, W4, Wm, and Wp, of which the functions of only three are known. None of the recessive alleles at those three loci completely abolishes flower pigmentation. On the other hand, the W1 locus is required for flower pigmentation. Based on genetic and biochemical data, the trihydroxylation reaction to form delphinidin-3-glucoside has been postulated as a likely function of the W1 locus. We report the cloning and sequencing of a flavonoid 3'5'-hydroxylase (F3'5'H) gene from the purple flower soybean line L79-908 (W1) and the recessive allele from the white flower isoline ‘Williams’ (w1). The mutation is a rearrangement leading to a small (65 bp) insertion of tandem repeats in exon 3 that truncates the translation product prematurely. The presence of this insertion in all-white flower soybean lines examined and in those of a segregating population resulting from a cross between purple and white flower lines presents compelling evidence that the F3'5'H gene isolated is likely encoded by the W1 locus of Glycine max. GmF3'5'H is a single-copy gene, expressed at very low levels in all tissues examined, including flower and seed coats, but sufficient to account for the contribution of the delphinidin-based anthocyanins and/or proanthocyanins in these tissues.

Abbreviations: aa, amino acids • ARS, Agriculture Research Service • CHS, chalcone synthase • DFR, dihydroflavonol reductase • EST, expressed sequence tag • F3H, flavanone 3-hydroxylase • FLS, flavonol synthase • PCR, polymerase chain reaction • RFLP, restriction fragment length polymorphism • RT–PCR, reverse transcriptase–polymerase chain reaction


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 
SOYBEAN [GLYCINE MAX (L.) MERR.] flower pigmentation is determined by at least six independent loci, W1, W2, W3, W4, Wm, and Wp (Palmer et al., 2004). Most of the soybean introductions (cultivars) have purple (W1W1) or white (w1w1) flowers, but there is color variation in some cultivars and wild perennial Glycine species. Genotypes that produce purple-throat (W1W1W3W3 w4w4), near-white (W1W1w3w3 w4w4), dilute-purple (w4-dp w4-dp), pink (W1W1wpwp), and magenta (wmwm) flower phenotypes have been described, and thus far, the function of only three of those loci has been identified. Cosegregation of dihydroflavonol reductase (DFR) polymorphism with W3 alleles suggested that W3 encodes DFR or is closely linked to it (Fasoula et al., 1995). The enzyme DFR is an enzyme required to convert the dihydroflavonols to flavan 3, 4-diols, a step in the synthesis of all three anthocyanin classes, delphinidins, pelargonidins, and cyanidins (Fig. 1 ). Wp encodes a flavanone 3-hydroxylase (F3H) required earlier in the pathway for synthesis of all three anthocyanin classes. A transposon insertion on the second intron of the F3H1 gene created the pink flower allele wp (Stephens and Nickell, 1991, Zabala and Vodkin, 2005). Recently, Wm was found to encode a flavonol synthase (FLS) that reduces dihydroflavonols to flavonols. A single-base deletion in the gmfls1 gene truncated the gene product of the wm allele that confers the magenta flower phenotype (Takahashi et al., 2007). Thus, the product of the Wm locus is not required for the synthesis of anthocyanins (Fig. 1).


Figure 1
View larger version (23K):
[in this window]
[in a new window]

  		Figure 1. Anthocyanin metabolic pathway in Glycine max. Enzyme abbreviations and the flower and seed coat color loci (in uppercase letters) show the three branches leading to the synthesis of the purple delphinidins, pink pelargonidins, red cyanidins, and the flavonols. The flavonol kaempferol derived from dihydrokaempferol was not included to simplify the scheme. Wp, W3, and Wm are the flower color markers associated with the indicated enzymes in the pathway based on molecular and genetic evidence (Zabala and Vodkin, 2005; Fasoula et al., 1995; Takahashi et al., 2007). The suggested W1 assignment (marked by a circle) as F3'5'H was made by Buzzell et al. (1987) based on biochemical evidence. The I and T are the well-characterized loci that control seed pigmentation and that encode CHS (Todd and Vodkin, 1996) and F3'H (Toda et al., 2002; Zabala and Vodkin, 2003), respectively. PAL, phenylalanine ammonia-lyase; C4H, cinnamate 4-hydroxylase; 4CL, 4-coumarate: CoA ligase; CHS, chalcone synthase; CHI, chalcone isomerase; F3'H, flavonoid 3'-hydroxylase; F3'5'H, flavonoid 3'5'-hydroxylase; F3H, flavanone 3-hydroxylase; DFR, dihydroflavonol-4-reductase; ANS, anthocyanidin synthase (also called LDOX, leucoanthocyanidin dioxygenase); UFGT, UDP-flavonoid glucosyltransferase; FLS, flavonol synthase.

 
The visible spectrum of the delphinidins is in the blue-purple range, while the pelargonidins tend to the pinks and the cyanidins to the reds. On the other hand, flavonols (myricetin, kaempferol, quercetin), are yellow pigments that can either act as pigments per se or as copigments to the anthocyanins (Harborne, 1967). Most soybean flower colors are in the purple and pinks but not the reds; therefore, the delphinidin branch of the anthocyanin pathway must be fully active in the flowers. In contrast, seed coat pigmentation is controlled by three independent loci, I, R, and T. The I locus encodes a chalcone synthase (CHS) gene cluster manifesting a tissue-specific gene silencing in the seed coat via siRNA (Todd and Vodkin, 1996; Tuteja et al., 2004: Senda et al., 2004), and the T locus encodes a flavonoid 3'-hydroxylase (F3'H) (Toda et al., 2002; Zabala and Vodkin, 2003). The R and T loci determine the type of anthocyanins coloring the seed coats. Based on the results obtained from genetic crosses and biochemical measurements of anthocyanidins in the black-pigmented seeds, Buzzell et al. (1987) suggested that T was involved in the synthesis of cyanidin-3-glucoside and the flower color allele W1 in the trihydroxylation in the formation of delphinidin-3-glucoside. These results pointed to flavonoid 3'5'-hydroxylase (F3'5'H) as a likely candidate gene to be encoded at the W1 locus.

In this report, we present molecular evidence that W1 is associated with a F3'5'H gene of 4657 nucleotides in size that contains two introns (1 and 2), 754 and 2102 bp long, respectively. The derived amino acid sequence predicts a polypeptide of 509 amino acids (aa) with similarity to other plant species F3'5'H protein sequences and contains the signature sequence for the heme-binding domain, FxxGxxxCxG, characteristic of P450 enzymes (von Wachenfeldt and Johnson, 1995), but lacks the GGEK motif that distinguishes the F3'H enzymes from the F3'5'H sequences (Brugliera et al., 1999). In addition, the mutation in the w1 allele of ‘Williams’ and six additional lines, L62-906, L64-2139, L63-2373, L65-1077, L69-4776, and L68-2056 (Table 1) was a 65-bp insertion that created a 37-bp direct repeat separated by 10 bp and extended the size of the w1 allele by 53 bp. This insertion introduced 10 amino acid substitutions and created a stop codon that terminates the translation product prematurely. Thus, the recessive allele product is 42 aa shorter than the functional enzyme translated from the W1 allele. The cosegregation of the 65-bp insertion with the white flower phenotype in a segregating F2 population resulting from a cross between the purple flower L66-14 (W1) and the white flower L68-2056 (w1) parents was further evidence that the F3'5'H gene isolated is likely encoded by the W1 locus. Interestingly, the F3'5'H gene appears to be single copy and expressed at such low levels that its mRNA could not be detected in RNA blots in any of the tissues examined. This contrasts with the high level of expression of F3'H (T) and F3H (Wp) genes observed in the seed coats of Williams (w1iiRTWp) and LN89-5320-6 (W1iiRtWp) lines, respectively (Zabala and Vodkin, 2003, 2005). That the two alleles, W1 and w1, were expressed at similar very low levels was determined by the more sensitive reverse transcriptase–polymerase chain reaction (RT–PCR) procedure. The low level of expression of the W1 allele suggests that relatively few molecules of the F3'5'H enzyme are sufficient to drive the synthesis of the delphinidin branch of the anthocyanin pathway to account for the major type of pigments in the colored soybean flowers and to the contribution the delphinidins may make to the seed coat and other pigmented tissues.


View this table:
[in this window]
[in a new window]

  		Table 1. Genotypes and phenotypes of soybean cultivars and backcrossed isolines.

 

   Materials and Methods
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
Results
 Discussion
 Conclusions
 REFERENCES
 
Plant Material and Genotypes
The Glycine max cultivars and isolines used for this study are described in Table 1. Seeds were obtained from the USDA Soybean Germplasm Collections (Department of Crop Sciences, USDA Agriculture Research Service (ARS), University of Illinois, Urbana, IL). The genotypes and flower phenotypes of the lines used are shown also in Table 1 and Fig. 2 . All lines are homozygous, and only one of the alleles is shown for brevity in the tables and text. Further information on the accessions can be found at the USDA–ARS National Plant Germplasm System (http://www.ars-grin.gov/npgs/acc/acc_queries.html; verified 11 June 2007).


Figure 2
View larger version (63K):
[in this window]
[in a new window]

  		Figure 2. Illustration of the effect of w1 on flower and seed coat phenotypes. (A) Purple flower of plants with W1 genotype (left panel) or white flower of plants with w1 genotype (right panel) in lines L79-908 (W1,ii,R,T,Wp) and Williams (w1,ii,R,T,Wp) respectively, both of which have yellow seed coats because of silenced ii allele. (B) Seed coats of plants L83-930 with W1,i,r,t,Wp genotype (left panel) compared with seed coats of plants of the isoline L82-2669 with w1,i,r,t,Wp genotype (right panel). It shows little difference in color revealing the very small contribution delphinidins and prodelphinidins make to seed coat pigmentation.

 
Plants were grown in the greenhouse. Roots, hypocotyls, unifoliates, and shoot tips (meristems surrounded by primordial leaves) were dissected from two-week-old plants. Flower buds were of early developmental stages and seed coats and cotyledons dissected from seeds at varying stages of development. For developing seed coats and cotyledons, seeds were divided into the following stages according to the fresh weight of the entire seed: 10 to 25 mg, 25 to 50 mg, and 50 to 75 mg. Upon collection, all tissues were frozen in liquid nitrogen, freeze dried (Multi-dry lyophilizer; FTS Systems, Stone Ridge, NY), and stored at –20°C.

DNA Isolation and DNA Gel-Blot Analysis
Genomic DNA was isolated from soybean freeze-dried shoot tips using the methods of Dellaporta (1993) with minor modifications (Zabala and Vodkin, 2003). Genomic DNA (12 µg) was digested with restriction endonucleases HindIII and/or PstI for ~2 h at 37°C and electrophoresed in a 0.7% agarose gel (Sambrook et al., 1989). Size-fractionated DNAs were transferred to Optitran-supported nitrocellulose membrane (Midwest Scientific, Valley Park, MO) by capillary action as described in Sambrook et al. (1989) and cross-linked in a UV stratalinker (Stratagene, La Jolla, CA). Nitrocellulose DNA gel blots were prehybridized, hybridized, washed, and exposed to Hyperfilm (Amersham, Arlington Heights, IL) as described by Todd and Vodkin (1996).

Primer Synthesis, PCR Reaction Conditions, and DNA Sequencing
Oligonucleotide primers were synthesized on an Applied Biosystems (Foster City, CA) model 394A DNA synthesizer at the Keck Center, a unit of the University of Illinois Biotechnology Center. Multiple primer pairs were designed and synthesized to span the F3'5'H genomic sequences from L79-908 (W1) and Williams (w1) isolines. Copies of genomic DNA fragments encoding the F3'5'H genes were obtained via polymerase chain reaction (PCR) using the following conditions: an initial denaturation step at 94°C for 2 min followed by 30 cycles of denaturing at 94°C for 30 sec, annealing at 56°C for 1 min, extension at 68°C for 9 min, to end with a 10 min extension at 72°C. High-fidelity and -efficiency ExTaq (Takara Bio Inc. Otsu, Japan) polymerase was used at 0.75 units per 50-µL reaction. Amplified DNAs were separated from oligonucleotides with a QIAquick PCR Purification kit (QIAGEN, Valencia, CA), cloned into pGem-T easy and sequenced in an ABI 3730 x l (Applied Biosystems, Inc. Foster City, CA) at the Keck Center.

Sequence data from this article have been deposited with the EMBL/GenBank data libraries under accession numbers [GenBank: EF174665] (F3'5'H genomic sequence L79–908-W1); [GenBank: EF174666] (F3'5'H genomic sequence Williams-w1).

Polymerase chain reaction conditions used to amplify the smaller DNA fragments close to the site of the insertion that created the w1 mutation were the following: an initial denaturation step at 96°C for 2 min followed by 39 cycles of denaturing at 96°C for 20 sec, annealing at 55°C for 1 min and polymerization at 72°C for 2 min, to end with a 7-min extension at 72°C. The primers used were 5'-TAG AAA GCA CCC TTC AAC AC-3' (F35H-3980F) and 5'-TTT ATG TAG CCA CAG CCA CA-3' (F35H-4558R). The resulting PCR amplification products were 526-bp and a 579-bp fragments from DNAs of the W1 or w1 lines, respectively. To visualize the 53 bp difference in size between those two PCR fragments, 8-µL aliquots of the 50 µL PCR reactions were fractionated in 3% NuSieve 3:1 agarose gels (FMC BioProducts, Rockland, ME).

RNA Extraction, Purification and RNA Gel-Blot Analysis
Total RNA was isolated from all mentioned soybean tissues using a phenol–chloroform and lithium chloride precipitation method (McCarty, 1986; Wang et al., 1994) and stored at –70°C until used. RNA (10 µg/sample) was electrophoresed in a 1.2% agarose–3% formaldehyde gel (Sambrook et al., 1989). Size-fractionated RNAs were transferred to Optitran-supported nitrocellulose membrane as described for DNA gel blots.

cDNA Synthesis
cDNA copies of the F3'5'H gene from the three lines, L79-908 (W1,ii,R,T,Wp), Williams (w1,ii,R,T,Wp), and L76-2023 (w1,ii,R,t,Wp), were amplified from a first-strand cDNA pool synthesized using 1 µg of flower bud total RNA and the Superscript first-strand synthesis system for RT–PCR (Invitrogen, San Diego, CA). The total RNAs used for these RT–PCRs were treated with DNAaseI using Ambion's DNA-free kit and concentrated in Microcon YM-30 columns (Millipore, Bedford, MA). For each RNA sample, parallel reactions were allowed in the absence of superscript (negative controls) to assess the extent of DNA contamination. The sequences of the two primers used were: 5'- TGATCACTCGTCTCTCCATTCA-3' (81F) and 5'-AATCTAAGCCTTGCAGGTGGTG-3' (1731R). The primer's numbering was based on the sequence of a cDNA with the accession number AY117551.

Probes for DNA and RNA Gel Blots
Cloned DNAs used as probes were digested from their vectors or PCR amplified, electrophoresed, and purified from the agarose using the QIAquick gel extraction kit (QIAGEN, Valencia, CA). DNA concentration of the final eluate was determined with a NanoDrop instrument (NanoDrop Technologies, Inc., Rockland, DE). Purified DNA fragments (25–250 ng) were labeled with [{alpha}-32P]dATP by random primer reaction (Feinberg and Vogelstein, 1983).

Primers used to amplify the genomic fragment from the L79-908 W1 allele that was used as probe for the DNA gel blot were 5'-GCCCAAATTCGAGATGAAGAGA-3' (F35H-470F) and 5'-AATCTAAGCCTTGCAGGTGGTG-3' (F35H-1731R). The size of gene fragment amplified was 3367 bp and included the majority of exon 2, intron 2, and most of exon 3.


   Results
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
Discussion
 Conclusions
 REFERENCES
 
Identification and Isolation of a Glycine max Flavonoid 3'5'-Hydroxylase Gene from a Purple Flower L79-908 (W1) Line
Recent attempts at identifying and characterizing several of the seed coat and flower color genetic markers were helped by exploration of the recently developed large soybean expressed sequence tag (EST) collection and the use of soybean cDNA microarrays (Shoemaker et al., 2002; Vodkin et al., 2004). Several ESTs representing the F3'H gene and the F3H genes were found in the soybean EST collection entered in GenBank (Zabala and Vodkin, 2003, 2005). However, none of the soybean ESTs annotated as flavonoid hydroxylases proved to detect polymorphisms associated with the W1 locus recessive allele, w1. Only when we used primers designed from a cDNA sequence with accession number AY117551 were we able to amplify, clone, and sequence the F3'5'H gene from the L79-908 (W1) purple flower line and the mutant allele from the white flower isoline, Williams (w1) (see Table 1). The cDNA with accession number AY117551 was obtained from young seed coats of the cultivar Chin-Ren-Woo-Dou (CRWD) and entered in GenBank by R.-M. Liao, T.-M. Chu, and C.-S. Wang in 2003 (http://www.ncbi.nlm.nih.gov).

The F3'5'H genomic sequence of L79-908 (W1) isolated was 4657 nucleotides long with two introns of 754 and 2102 bp in length, respectively (Fig. 3 and Supplemental Fig. 1). The corresponding cDNA was 1801 bp long, and its derived amino acid sequence of 509 aa was 99% similar to the one derived from the F3'5'H cDNA (accession number AY117551). It only differed by one aa addition (S) at the 181 position and four aa substitutions (Fig. 4 ). These differences may represent variability between Williams and the Chin-Ren-Woo-Dou (CRWD) cultivar from which the AY117551 cDNA was isolated. A tblastn search found the Glycine max F3'5'H aa sequence highly similar to several other F3'5'H sequences, with Clitoria ternatea (AB234897) being the most similar (80% identity), followed by Vitis vinifera (AJ880359, 77% identity), Gossypium hirsutum (AY275430, 75% identity), Petunia hybrida (Z22544, 75% identity), and Lycianthes ratonnei (AF313490, 74% identity) (see amino acid sequence alignment of these closely related F3'5'H sequences in Supplemental Fig. 2). The Glycine max F3'5'H aa sequence and those related sequences have near the C terminus the conserved aa motif (FxxGxxxCxG) that surrounds the heme-binding cysteine residue and distinguishes the P450s from other heme enzymes (Fig. 4 and Supplemental Fig. 2) (von Wachenfeldt and Johnson, 1995). These results indicate that the genomic sequence amplified from the L79-908 line (W1) encodes a F3'5'H enzyme.


Figure 3
View larger version (22K):
[in this window]
[in a new window]

  		Figure 3. (A) Schematic representation of F3'5'H genomic sequences. Top graphic represents the components of the F3'5'H genomic sequence isolated from the L79-908 (W1) line. The three solid blocks are three exons predicting a cDNA 1801 bp in length. The two blocks representing introns of 754 and 2102 bp in length are shown in hatched lines. Bottom scheme is a graphic summary of the F3'5'H genomic sequence isolated from the ‘Williams’ isoline with the w1 allele. The three solid blocks represent the three exons predicting a slightly larger cDNA of 1854 bp. Exon 3 is 53 bp longer in the F3'5'H gene of the w1 white-flowered line than in the F3'5'H gene of the W1 purple-flowered line. The small block above exon 3 represents the extra 53 bp that contains part of the tandem repeats (arrows) resulting from the 65-base insertion (53 additional bases and 12 bases of sequence substitution). The two introns (755 bp and 2104 bp respectively) in the F3'5'H alleles isolated from both the w1 and W1 isolines are very similar in size and sequence. (B) Small tandem repeats in the F3'5'H gene of plants with the w1 white flower allele. Nucleotide numbering is from the Williams (w1) genomic sequence. The nucleotide sequence and translated amino acids flanking the insertion are shown. Nucleotides in lowercase italic letters denote the 53 extra bases of the insertion. Highlighting denotes two identical 37-bp repeats found at positions 4242 to 4278 and 4289 to 4325 that are separated by 10 nucleotides. Bold letters indicate the 12 nucleotides of the uninterrupted F3'5'H allele that are replaced by the 65-bp insertion. Also marked is the 21 nucleotide repeat found at position 4189 to 4209 in both the F3'5H alleles of the W1 line and w1 isolines that matches a portion of the tandem repeat unit of the insertion. The conserved P450 heme-binding domain is underlined. * denotes the premature stop codon created by the insertion.

 

Figure 4
View larger version (66K):
[in this window]
[in a new window]

  		Figure 4. Alignment of three Glycine max F3'5'H cDNAs derived amino acid sequences. The three amino acid sequences were derived from (i) the cDNA annotated as F3'5'H in GenBank with accession number AY117551 and isolated from the cultivar Chin-Ren-Woo-Dou (W1) by R.-M. Liao, T.-M. Chu, and C.-S.Wang (http://www.ncbi.nlm.nih.gov); (ii) the cDNA sequence predicted from the L79-908 (W1) F3'5'H genomic sequence reported here; and (iii) the cDNA sequence predicted from the ‘Williams’ (w1) genomic sequence described here. The alignment was done with Multalin, a multiple sequence alignment on-line tool by Corpet (1988). The consensus sequence shows identical amino acids in the three sequences; divergences between two of the amino acid sequences are highlighted. The amino acid sequence derived from the Williams w1 recessive allele shows the amino acid substitutions introduced by the 65-bp insertion. This sequence terminates earlier (stop codon indicated by *) and is missing the last 44 amino acids. Overall, it is a predicted 42 amino acids shorter than the polypeptide derived from the wild-type sequence in the dominant W1 line. Underlined is the motif that distinguishes P450s from other heme-binding enzymes.

 
Isolation and Characterization of an F3'5'H Mutant Genomic Sequence from the Williams Isoline Containing the w1 Allele
With the same set of primers used to amplify the F3'5'H genomic sequence from the L79-908 (W1) line, we were able to amplify a related genomic sequence from the Williams (w1) line (Supplemental Fig. 1). The F3'5'H gene in the w1 genotype was 4713 bp long with two introns of 755 and 2104 bp in length and a predicted cDNA 1854 bp long (Fig. 3A). The mutation resulted from a 65-bp insertion in exon 3 that apparently substituted 12 nucleotides and that also created a directly repeated sequence 37 bp long, with the following sequence, TTTGAGCTTATTCCATTTGGTTCACTACATTTTGGGC, separated by 10 nucleotides (Fig. 3B and Supplemental Fig. 1). As a result of the 65-bp insertion and 12-bp substitution, the mutated sequence is 53 bp longer than that of the wild-type allele. Alignment of the derived amino acid sequences revealed the introduction of 10 aa substitutions and a premature stop codon in the Williams (w1) sequence that shortened the translation product by 42 aa (Fig. 4). The isolation of this F3'5'H gene mutation in the white flower Williams (w1) isoline is evidence that the W1 locus encodes a F3'5'H.

The F3'5'H Small Insertion Is Found in Genotypes with the w1 White Flower Phenotype and Cosegregates with the w1 Locus
To further strengthen the assertion that the W1 could encode a F3'5'H, we analyzed a series of Glycine max cultivars and lines in which the W1 or w1 alleles had been introduced independently from diverse sources. Table 1 show pairs of existing isolines for the W1 locus that were previously created in three different genetic backgrounds, Williams, ‘Harosoy’, and ‘Clark’, and that are part of the USDA germplasm collection. For example, the cultivar Williams has white flowers (w1w1), but by crossing originally to ‘Lee’ with purple flowers (W1W1) and selecting for purple flowers through six backcrosses with the recurrent parent Williams in each generation, a backcrossed isoline L79-908 was established in Williams that has purple flowers (W1W1). As we have shown above by isolating the corresponding W1 and w1 alleles from the L79-908 (W1) and Williams (w1) isolines, the W1 allele originating from Lee does not contain the insertion, but the w1 allele in the standard Williams cultivar does contain the 65 base insertion as determined by direct sequence analysis.

Using two primers whose sequences spanned the 65-bp insertion of the mutant allele and that would amplify 526-bp and 579-bp fragments from DNAs of the W1 or w1 lines, respectively, it became feasible to distinguish the size difference (53 bp) between those fragments in a NuSieve agarose gel. The results of this analysis are shown in Fig. 5 and clearly display a perfect correlation between the 579-bp larger fragments and the lines with the w1 allele and white flower phenotype (Table 1). Also, it shows perfect correlation between the 526-bp smaller fragment and the lines with the W1 purple flower phenotype. In all, eight independent backcrosses in three different genetic backgrounds showed perfect correlation of the molecular difference with the flower phenotype.


Figure 5
View larger version (33K):
[in this window]
[in a new window]

  		Figure 5. Correlation between the F3'5'H 65-bp insertion and independently introduced w1 alleles. Polymerase chain reaction (PCR) fragments of 579 and 526 bp amplified from genomic DNAs isolated from multiple soybean lines (Table 1) with independently introduced W1 alleles were separated in a 3% NuSieve 3:1 agarose gel. The 53-bp longer PCR fragment was amplified from DNAs of cultivars and isolines with the w1 allele while the smaller 526 PCR fragment correlated with cultivars and lines with the W1 allele. Primers used were F35H-3980F and F35H-4558R. (A) Ethidium bromide stained gel showing the PCR fragments amplified from DNAs isolated from two independent plants of two isolines, L68-2056 and L66-14, which were used as parents for the segregating population shown in Table 2. (B) Ethidium bromide stained gel with PCR products amplified from DNAs isolated from 10 different soybean lines varying at the W1 locus (Table 1).

 

View this table:
[in this window]
[in a new window]

  		Table 2. Cosegregation of W1 with a small insertion in GmF3'5'H.

 
The same type of analysis was performed on DNAs isolated from an F2 population (40 individual plants) segregating for flower color resulting from a cross between the L66-14 (W1, purple flower) and L68-2056 (w1, white flower) parents (Table 1). The phenotype of the segregating plants was scored in the field and leaf tissue harvested and freeze-dried before DNA extraction. The genotypes of the plants were expected to be homozygous (W1W1) or heterozygous (W1w1) for the purple flower phenotype or homozygous (w1w1) for the white flower phenotype. The GmF3'5'H PCR fragments corresponding to the two parents, A (526 bp) and B (579 bp) and the heterozygous H pattern (526 and 579 bp) were amplified from the DNAs of the segregating plants. Table 2 shows a summary of the fragment patterns and phenotypes of the F2 plants. As expected for a single locus with codominant alleles, the F3'5'H fragments (526 and 579 bp) amplified by PCR segregated in a 1:2:1 ratio. The purple flower color phenotype segregated 3:1 as expected for a single gene with dominant–recessive inheritance. The recombination analysis showed clear cosegregation of the F3'5'H PCR fragments with flower color (Table 2). These results are additional evidence that the F3'5'H gene isolated is associated with the W1 locus.

The Flavonoid 3'5'-Hydroxylase at the W1 Locus Is a Single-Copy Gene
It is common in soybean to find several genes, members of a family, with very similar sequence and function. To investigate if there were more than one copy of the F3'5'H gene, we analyzed the hybridization pattern in a blot containing DNAs isolated from three pairs of isolines differing at the W1 locus (L79-908, Williams; L92-7600, L64-2139; L66-14, L68-2056) (Table 2). The DNAs were digested with Hind III and Pst I singly or in a double digest. Hind III cuts the F3'5'H genomic sequence 298 nucleotides downstream of the 65-bp insertion, dividing the allelic genomic sequences into two fragments, one of them including the 65-bp insertion. Pst I does not cut the isolated F3'5'H sequences. The probe used was a 3.4-kb genomic fragment that extended almost the entire length of exon 2, intron 2, and exon 3 of the W1 allele (see "Materials and Methods" for details). Figure 6 shows a very simple hybridization pattern of only two Hind III fragments in all DNA samples with the smaller fragment (~3 kb) detecting the size polymorphism due to the 65-bp insertion in the lines with the w1 allele. Only one high molecular band hybridized to the probe in the Pst I digested DNAs. The double digested DNAs presented the same pattern of hybridization as with Hind III alone. These results suggest that the F3'5'H gene is single copy in Glycine max.


Figure 6
View larger version (81K):
[in this window]
[in a new window]

  		Figure 6. Glycine max F3'5'H gene polymorphisms and copy number. A DNA gel blot showing the hybridization of a genomic F3'5'H probe (3.5 kb) to DNAs isolated from three pairs of W1 isolines digested with HindIII and/or PstI. The smaller size HindIII fragment shows the 53-bp size difference between the W1 and w1 lines both in the single (HindIII) and double (HindIII and PstI) digested samples. HindIII cuts the F3'5'H gene in exon 3; the hybridization pattern shows the resulting two fragments in samples digested with this enzyme. PstI does not cut the F3'5'H gene.

 
Expression of the Glycine max Flavonoid 3', 5'-Hydroxylase Gene in Soybean Tissues
The W1 allele is a genetic marker for flower color but not for seed coat color in soybean; therefore, it was possible that the F3'5'H gene could manifest tissue-specific expression. We have previously shown that both the F3'H (T) and the F3H1 (Wp) genes are strongly expressed in seed coats and to a much lesser extent in other tissues examined, including flowers and flower buds (Zabala and Vodkin, 2003, 2005). To determine the level of expression of the F3'5'H gene in different tissues, we analyzed the intensity of hybridization of several F3'5'H cDNA probes to three different RNA blots containing RNAs from multiple tissues (roots, hypocotyls, unifoliates, meristems, and flower buds) as well as various stages of seed coat and cotyledon development. The tissues were dissected from plants of the L79-908 purple flower line (W1,ii,R,T,Wp) as well as seed coats, cotyledons, and flower buds from the Williams (w1,ii,R,T,Wp) and L76-2023 (w1,ii,R,t,Wp) white flower lines. Surprisingly, the level of hybridization was so low that it was practically undetectable in most tissues analyzed even after exposing films for as long as 13 d (data not shown).

Nonetheless, we determined that the GmF3'5'H gene was expressed at very low levels in flowers of all three lines, the purple flower L79-908 (W1,ii,R,T,Wp) and white flower Williams (w1,ii,R,T,Wp) and L76-2023 (w1,ii,R,t,Wp), by amplification of the F3'5'H mRNA via RT–PCR with the 81F and 1731R primers (see "Materials and Methods" for details). Figure 7 shows cDNAs amplified from the three samples with sizes resembling the predicted 1654 (W1) and 1707 (w1) nucleotide mRNAs. The amplified cDNA products are low abundance considering that the entire 50 µL PCR reactions were loaded in the gel, indicating that the initial F3'5'H mRNA pool in the three samples consisted of very few molecules. In summary, the F3'5H gene appears to be expressed at very low levels from both the purple flower W1 and white flower w1 alleles, and it does not seem to follow the same pattern of high expression in the seed coats as was shown for the F3'H (T) and F3H1 (Wp) genes (Zabala and Vodkin, 2003, 2005).


Figure 7
View larger version (38K):
[in this window]
[in a new window]

  		Figure 7. Expression of F3'5'H by reverse transcriptase–polymerase chain reaction (RT–PCR) in purple- and white-flowered plants. Ethidium bromide stained gel showing the RT–PCR amplified cDNAs from flower buds of three W1 locus isolines, L79-908 (W1,ii,R,T,Wp), ‘Williams’ (w1,ii,R,T,Wp), and L76-2023 (w1,ii,R,t,Wp). The ~1.6-kb amplified cDNAs shown in the (+) lanes are evidence that the two F3'5'H alleles in W1 and w1 lines are expressed at low levels in the flowers of these soybean lines. The (-) lanes are the result of parallel PCR reactions that were allowed in the absence of superscript to assess the extent of DNA contamination (negative controls).

 
In plants of the L83-930 (W1irtWp) line with a recessive t allele (Fig. 2B) pigmentation could be the result of anthocyanidins and proanthocyanidins derived from two branches of the pathway, the delphinidins and pelargonidins. Figure 2B shows the relatively small contribution delphinidins make to seed color pigmentation. The color of the seeds with the W1 allele is very similar to those seeds with the recessive allele w1. In plants of the Williams (w1iiRTWp) line with the T allele, synthesis of the red cyanidins could potentially occur, but to date, there is no evidence of cyanidins being synthesized in flowers. As shown in Fig. 2A, flowers with w1iiRTWp genotype are white. This suggests a compartmentalization of the expression of T and the cyanidin biosynthetic pathway in the seed coat and pubescence hairs (Toda et al., 2002; Zabala and Vodkin, 2003). In contrast, W1 appears to be expressed at very low levels in the flowers and other tissues, including the seed coats.


   Discussion
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
 Discussion
Conclusions
 REFERENCES
 
Soybean geneticists and breeders have used flower, hypocotyl, pubescence, and seed color loci as markers in their breeding programs and restriction fragment length polymorphism (RFLP) mapping because they have clearly visible phenotypes. Since pigmentation in those tissues is mainly due to anthocyanidins and proanthocyanidins, one would expect that the color loci markers encode enzymes of the anthocyanin pathway and that as such they should affect similarly pigmentation of the flowers, hypocotyls, pubescence and seed coats. Interestingly, in soybean the loci that affect flower color (W1, W2, W3, W4, Wm and Wp) are different from those controlling seed pigmentation (I, R, T) which suggests a functional compartmentalization of the different branches of the anthocyanin pathway and/or different regulatory controls of the expression of those genes. It is now known that seed coat pigmentation is controlled in a tissue specific manner by the I locus that encodes a family of CHS genes and in its dominant form is silenced via siRNA (Todd and Vodkin, 1996; Tuteja et al., 2004: Senda et al., 2004). The T locus encodes a flavonoid 3'-hydroxylase (F3'H) (Toda et al., 2002; Zabala and Vodkin, 2003) that directs the synthesis of the cyanidin branch of the anthocyanin pathway. The fact that the T locus is not considered a flower color marker suggests that cyanidins and proanthocyanidins may not contribute to flower color or protection of this organ from UV radiation or predators. The function of the R locus has not been identified, but it has been speculated that it acts after the formation of leucoanthocyanidin but before the formation of anthocyanins (Todd and Vodkin, 1993).

In contrast to the three loci controlling seed coat pigmentation, flower coloration (purple, white, purple-throat, near white, magenta, and pink) is affected by at least six loci (W1, W2, W3, W4, Wm, and Wp) working epistatically. W3 likely encodes a DFR (Fasoula et al., 1995), an enzyme that is required for the synthesis of all three anthocyanin classes, delphinidins, pelargonidins, and cyanidins (Fig. 1). The W3 and homozygous w4 alleles produce the purple-throat and near-white flower phenotypes, respectively, each of which still displays some pigmentation. Wp encodes an F3H required earlier than DFR in the pathway for the synthesis of all three anthocyanin classes, as shown in Fig. 1 (Zabala and Vodkin, 2005). Flowers of plants with the wp allele are also pigmented and have been referred to as very light pink in color (Stephens and Nickell, 1991, 1992; Johnson et al., 1998). In this instance, we know that the wp mutation, a transposon insertion in the second intron of the F3H1 gene, permits a low level of F3H expression that affects the color of the flowers more noticeably than that of the seed coats (Zabala and Vodkin, 2005; Zabala and Vodkin, unpublished data, 2007). Wm encodes an FLS that reduces dihydroflavonols to flavonols (Takahashi et al., 2007). Plants with the wm allele also have pigmented flowers, known as magenta. The product of the Wm locus is not required for the synthesis of anthocyanins but the lesser amount of the yellow pigmented flavonols, most likely myricetin and possibly kaempferol alter flower anthocyanin copigmentation from the purple (Wm) to the lesser blue (magenta) (wm) phenotype (Takahashi et al., 2007).

In contrast to those three flower color loci described above, a recessive allele of the W1 locus abolishes flower pigmentation in plants with the genotype w1,ii,R,T,Wp. This suggests that the gene at the W1 locus is required for the synthesis of the anthocyanins that paint the soybean flowers. We report here the cloning and sequence of a F3'5'H gene from the L79-908 (W1,ii,R,T,Wp) and a recessive allele from the Williams (w1,ii,R,T,Wp) isoline. The F3'5'H gene (4657 bp) contained two introns (754 bp and 2102 bp) and three exons predicting a 1801-bp cDNA. High degree of similarity (80–74% identical aa) of the predicted amino acid sequence with those of other plant species F3'5'H sequences supports the annotation of the isolated gene as a F3'5'H (Supplemental Fig. 2). The recessive allele (4713 bp) had two introns (755 bp and 2104 bp) and the three exons predicted a 1854-bp cDNA. Alignment of the two genomic sequences revealed the nature of the mutation to be the insertion of 53 extra bases and the substitution of 12 nucleotides in the w1 allele relative to the W1 allele. These 65 bases clearly have a tandem repeat nature at a minimum of 18 bases that extends even further (to 37 bases) if the borders are considered. Thus, the w1 allele contains a tandem 37 bp repeat unit separated by 10 nucleotides. Its origin is likely a repetition and rearrangement of existing nucleotides within the protein coding region of the normal F3'5'H allele as some of both the 5' and 3' nucleotides flanking the insertion are included in the extended tandem repeat unit (Fig. 3B and Supplemental Fig. 1). The first 21 of the 37 nucleotides in the tandem repeat are identical to a region (genomic sequence nucleotides 4189 to 4209) found in both the W1 and w1 alleles just 5' of the tandem repeats (Fig. 3B and Supplemental Fig. 1). This region perfectly matches the 12 substituted nucleotides of insertion repeat at w1 position 4242, indicating that the coding region at 4189 is likely one segment of the origin of the tandem repeats in the w1 allele. The 65-bp insertion introduced a stop codon that terminates the translation product prematurely (Fig. 4). The heme-binding domain characteristic of the F3'5'H P450 enzymes is retained in the resulting truncated peptide, but the lack of function in the w1 plants suggests that an additional portion of the C terminus is required for the stability and function of the F3'5'H mRNA or protein product.

Correlation between the 65-bp insertion and the w1 allele introduced independently in multiple white flower lines (Table 1; Fig. 5) and in the F2 population segregating from a cross between the purple flower L66-14 (W1,ii,R,T,Wp) and white flower L68-2056 (w1,I,r,t,Wp) parents (Table 2) are evidence that the F3'5'H gene isolated is likely encoded by the W1 locus. The GmF3'5'H gene (W1) appears to be single copy based on the simple hybridization pattern obtained with a 3.5-kb genomic probe to digested genomic DNAs (Fig. 6). In addition, it is expressed at very low levels in most soybean tissues judging from the almost undetectable hybridization to blots with RNAs from multiple tissues. Using the more sensitive RT–PCR technique, it was found that both the dominant and the recessive alleles are expressed at low levels in flower buds (Fig. 7). Thus, it appears that low amount of the F3'5'H enzyme that directs and is required for the synthesis of delphinidins, the blue-purple anthocyanidins, is sufficient to account for most of the coloration in soybean flowers. Based on the low-level expression of the GmF3'5'H gene, the contribution of the delphinidins to seed coat pigmentation is likely lower compared with that of the cyanidins. As mentioned, F3'5'H RNAs could not be detected in RNA blots, while the F3'H gene (T) that drives the synthesis of the cyanidin branch of the anthocyanin pathway is strongly expressed in seed coats as well as the F3H gene (Wp) (Zabala and Vodkin, 2003, 2005). The above observations are supported by the anthocyanin measurements reported by Buzzell et al. (1987). They found that in seed coats of TTW1W1 and TTw1w1 genotypes, the cyanidin-3-glycoside was the main pigment. They also detected delphinidin-3-glycoside in seed coats with TTW1W1 and ttW1W1 genotypes but much lower amounts compared with the cyanidin-3-glycosides in TTW1W1 or TTw1w1. In addition, Todd and Vodkin (1993) measured proanthocyanidins in pigmented seed coats during seed development, and they were able to detect procyanidin and propelargonidin but were unable to detect prodelphinidin. Because Buzzell et al. (1987) had reported measuring anthocyanin delphinidins in pigmented seed coats with the W1 allele, Todd and Vodkin speculated that the action of the W1 gene is specific to the anthocyanin pathway, and there was no parallel function in the proanthocyanidin pathway.

The W1 locus was one of the first phenotypic markers in soybean to be defined genetically (Woodworth, 1923) and was often employed in constructing the classical linkage groups of soybean phenotypic traits. Variation in flower color at the W1 locus is common in many of the populations used for molecular marker mapping. Thus, W1 has also been used as a phenotypic marker relative to the molecular RFLP and simple sequence repeat maps that have been constructed in recent years. In these maps, it has been positioned on linkage group F (Shoemaker et al., 2004). In the absence of a transgenic complementation of the w1 allele with a functional copy of the F3'5H gene, the formal possibility exists that W1 could encode another enzyme in the pathway and that the rearrangements in the closely linked F3'5'H gene in the multiple w1 genotypes are coincidental. However, the most likely interpretation of the convergent biochemical, molecular, and genetic data are that W1 in soybean encodes an F3'5'H.


   Conclusions
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
 Discussion
 Conclusions
REFERENCES
 
The isolation and sequencing of two related genomic regions from the genomes of two soybean isolines varying at the W1 flower color locus provided the necessary tools to gather sufficient evidence that the W1 locus probably encodes a flavonoid 3'5'-hydroxylase gene (GmF3'5'H). The white flower allele w1 mutation appears to have been created by a small 65-bp insertion consisting of tandem repeats that results in a translation product that terminates prematurely rendering the recessive w1 allele nonfunctional. The GmF3'5'H gene was found to be single copy, and both the W1 and w1 alleles are expressed at very low levels in all tissues examined. Thus, our molecular data on the structure of the F3'5'H genes in lines with the homozygous W1 versus w1 genotypes agree with earlier biochemical evidence (Buzzell et al., 1987) suggesting that the W1 locus may encode F3'5'H. Thus, the W1 locus encodes the enzyme required for the biosynthesis of the delphinidin branch of the anthocyanin pathway that appears to be the major contributor to soybean flower color. The contribution of the delphinidins to seed coat pigmentation is much less than that of the cyanidins. In contrast, cyanidins are abundant in seed coats, but it appears that they are not synthesized in the flowers. Because soybean flowers are self-pollinated and very small, the plant seems not to invest too much energy into coloring them to attract pollinators or protect them from predators with the more colorful cyanidins and protective proanthocyanidins. The association of the W1 locus with F3'5'H also provides an important molecular marker integrating the classical and molecular maps in this important agronomic crop and increases understanding of the basic genetic and molecular control of the flavonoid pathway that is important to both plant protection and nutrient composition.


   ACKNOWLEDGMENTS
 
We thank Laura Guest and Virginia Lukas who directed the DNA sequencing and synthesis of the oligonucleotide primers. Special thanks to Kay Wallheimer for her assistance with the graphic display of the manuscript's figures. We gratefully acknowledge support from grants of the Illinois Soybean Program Operating Board, USDA, and United Soybean Board.

Received for publication December 31, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Supplemental Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zabala, G.
Right arrow Articles by Vodkin, L. O.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Zabala, G.
Right arrow Articles by Vodkin, L. O.
Agricola
Right arrow Articles by Zabala, G.
Right arrow Articles by Vodkin, L. O.
Related Collections
Right arrow Soybean
Right arrow Cell Biology & Molecular Genetics

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome