Abstract
The horticulturally valuable traits of faba bean are poorly explored, including the available information on the genetics of flower color and pattern. This lack of understanding has reduced the inclusion of unique flower color into the horticultural-type faba bean market. The modes of inheritance of two flower colors (red petals and yellow spot on wing petals) were examined through the development of multiple F2 segregating populations. The inheritance of red flower was confirmed for two recessive genes and yellow wing spot inheritance was confirmed for a single recessive gene. These populations led to the discovery of combinations of red and yellow flower color that have not been previously reported. The solid wing petal color gene was confirmed as a single recessive gene. Understanding the inheritance of flower color in faba bean can lead to improvement of current vegetable types and opens up possibilities for ornamental markets.
Faba bean (Vicia faba) is a protein-rich, nutritious legume crop (Khazaei and Vandenberg, 2020; Warsame et al., 2018). It can be consumed as both grain (pulse) and vegetable. In its horticultural form, pods are harvested when the seeds are filling, and the seeds are removed for fresh vegetable consumption (Anthony, 2017). The floriculture industry is another unexplored market for faba bean, as it exhibits large variation for many ornamental traits, including flower color. Its flowers are typical papilionaceous, with five petals consisting of one standard, two wings, and two parts fused into a keel (Duc, 1997). The wild-type flower is white petalled with a pronounced black spot on each wing petal and dark striae on the standard petal. The known flower colors in faba bean are wild type, white (absence of any pigments in the wing petals), yellow wing spots, solid brown, pink, diffused yellow, and red (Cabrera, 1988; Duc et al., 2015). Flower color has been used as a morphological indicator of high (wild-type) tannin content as one of the important antinutritional factors in seeds. The seeds of colored-flowered faba beans have high tannin content and white-flowered faba beans are low in tannin (Martín et al., 1991; Zanotto et al., 2020a).
In 1930, the first genetic study of faba bean flower color examined the inheritance of white flower, where it was determined to be a single recessive gene (Erith, 1930). Later reports have shown that white flower color, together with tannin-free seedcoat, is determined by two complementary zero tannin genes, zt1 and zt2 (Crofts et al., 1980; Picard, 1976; Rowlands and Corner, 1962). These genes have pleiotropic effects causing nonexpression of the stipule spot and stem color (Metz et al., 1992). The zt1 and zt2 genes have been genetically mapped in faba bean chromosomes 2 and 3, respectively (Gutierrez and Torres, 2019; Zanotto et al., 2020a). Previous work on the genetics of red and yellow-spotted flowers has been limited to the work of Sjödin (1971) and Cabrera (1988). Sjödin (1971) identified four loci that affect flower color and denoted them as sp-a (refers to zt1), sp-b (refers to zt2), dp-a, and dp-b. Red flower color appears when the locus dp-a is homozygous recessive and the yellow wing spot appears when dp-b is homozygous recessive. The monogenic inheritance of the solid wing color was confirmed by Cabrera (1988) and then determined to be segregating independently of the standard petal color. There has been limited research on the yellow-spotted mutant. However, both Sjödin (1971) and Cabrera (1988) performed a cross between a yellow-spotted type and a wild-type plant that indicated monofactorial inheritance. In addition, Cabrera and Martin (1989) showed that yellow pigment in flowers had pleiotropic effects resulting in yellow testa in faba bean.
The first step for developing successful horticultural-type faba beans is an understanding of the genetics of the potential aesthetic traits, especially flower color. Faba beans have large clusters of fragrant flowers in an array of colors and patterns. By creating combinations of these unique characteristics together with reduced toxic constituents (vicine and convicine, v-c), faba bean has the potential to enter the ornamental and floriculture markets. Therefore, the primary aims of this study were to investigate the genetic control of red flower color and yellow wing spot along with their interactions in faba bean. The motivation of this research was to facilitate the expansion of faba bean into the ornamental market.
Materials and Methods
Plant material.
Eight faba bean inbred lines were used for this study. The flower color characteristics and origin of these genotypes are given in Table 1. The flower color of parental lines are presented in Fig. 1A–D. Aurora/2, IG 114476 (accession DOI: 10.18730/8VJQF), Rinrei (Fukuta et al., 2004), and IG 12658 (accession DOI: 10.18730/60V47) were the sources of wild-type flower color. CDC Snowdrop and Disco/2 were used as sources of white flower (low tannin) carrying zt1 and zt2 genes, respectively. Red-flowered genotype P47-1 was derived from the vegetable faba bean “1778 (Crimson-Flowered),” which was listed by Fearing Burr in 1863 in a horticultural seed catalogue (Burr, 1863; Watson, 1996). The source of yellow wing spot flower was a spontaneous mutation (Sjödin, 1971).
Overview of the color characteristics of floral tissues, tannin content, origin, and percent homozygosity of faba bean inbred lines used in this study.
Flower color diversity of parental lines used in this study. (A) Wild type. (B) Yellow wing spotted. (C) White. (D) Red.
Citation: HortScience horts 55, 10; 10.21273/HORTSCI15238-20
Flower color diversity of parental lines used in this study. (A) Wild type. (B) Yellow wing spotted. (C) White. (D) Red.
Citation: HortScience horts 55, 10; 10.21273/HORTSCI15238-20
Flower color diversity of parental lines used in this study. (A) Wild type. (B) Yellow wing spotted. (C) White. (D) Red.
Citation: HortScience horts 55, 10; 10.21273/HORTSCI15238-20
Hybridization.
To study the genetics of red flower, the crosses P47-1 × Aurora/2, P47-1 × IG 114476, P47-1 × Rinrei, P47-1 × IG 12658, P47-1 × CDC Snowdrop, P47-1 × Disco/2, and P47-1 × Gelber were prepared along with their reciprocals. Crosses Aurora/2 × Gelber, P47-1 × Gelber, CDC Snowdrop × Gelber, and Disco/2 × Gelber were made to study the inheritance of yellow wing spot. Crosses were prepared by hand in the insect-proof growth chambers of the College of Agriculture and Bioresources at the University of Saskatchewan, Canada.
Growing conditions.
Seeds of parental lines were scarified, inoculated with Rhizobium leguminosarum bv. viciae and then planted into 4-L pots of soilless growing mixture No. 3 (Sun-Gro Horticulture, Agawam, MA) and placed in a growth chamber. The light conditions in the phytotron chamber were 16-h days at a photon flux of 300 μmol·m−2·s−1 followed by 8 h of dark. The temperature was set to 21 °C during the light interval, and to 18 °C for the dark interval. Plants were watered as necessary and fertilized with a blend of 15–30–15 (N–P–K) biweekly after the 10th node stage. The parental lines, F1 and F2 seeds were grown into 10-cm pots under similar growing conditions explained previously to score them all at the same growing conditions. Biological control was used using both beneficial mites (Amblyseius cucumeris) and nematodes (Steinernema feltiae) to suppress the thrips population.
Phenotyping.
The flower color was scored visually for each plant by recording the standard color, wing color, and presence of a wing spot. The flower color of the parent lines in this study were evaluated according to the horticultural color chart (Wilson, 1938). The term “solid” is used to describe a flower phenotype where the wing petal is totally pigmented and “spot” refers to the phenotypes of observable wing spots.
Parental line genotyping.
The eight parental lines were genotyped using 875 single nucleotide polymorphism (SNP) markers developed by Webb et al. (2016) to examine their percent homozygosity. DNA was extracted from a seedling leaf of a single plant of each parent line. Ten discs (diameter 5 mm) from healthy, newly expanded leaves of the parental lines were shipped to the LGC Genomics laboratory (LGC Genomics, Beverly, MA) for genomic DNA extraction and SNP genotyping according to the manufacturer’s instructions.
Statistical analysis.
Statistical goodness of fit of observed offspring segregation ratios to their expected counterparts for flower color characteristics were determined by the standard χ2 test using R Statistical Computing (R Development Core Team, 2019). The expected number of offspring in each cross was obtained from the hypothesis of either one or two unlinked recessive genes inherited in a simple Mendelian fashion. Cross P47-1 × Gelber was evaluated as a trihybrid gene model.
Results
All inbred lines used as parents were highly homozygous except Aurora/2 (Table 1). The genotyping call on parental lines is presented in Supplemental Table 1. Counts for red and non-red flowers for each cross are shown in Table 2. F1 plants from all “red × non-red” and their reciprocal crosses produced wild-type flowers. The F2 results suggested that the red flower phenotype was controlled by two recessive genes (15 non-red:1 red). However, when the red-flowered genotype was crossed with the wild-type genotypes or white-flowered carrying zt1 gene, a better fit to the expected segregation ratio was observed compared with crosses with the yellow-spotted and zt2 genotypes (Table 2). The F2 progenies of cross IG 114476 × P47-1 segregated 85 wild type, 18 solid brown, 17 pink, and 7 solid red flowers. This ratio fit to 9:3:3:1 Mendelian ratio of the double recessive gene model (χ2 = 6.053, P = 0.109); however, no similar ratio was found for other “red × non-red” cross combinations or their reciprocal crosses.
Observed offspring segregation ratios, value of χ2 test and corresponding P value for non-red and red-flowered (standard, wings, and keel) for F1 and F2 generations of faba bean crosses.
In “wild-type × yellow spotted” crosses, two distinct wing spot color phenotypes were observed: black and yellow. The F2 segregation ratio confirmed a monogenic Mendelian inheritance (three black spotted:one yellow spotted), with black spotted being dominant to yellow-spotted wing petals. The F2 populations involving zt1 (CDC Snowdrop) and zt2 (Disco/2) fit a 9:3:4 recessive epistasis where white flower was epistatic over the yellow spot (Table 3). When both white and yellow genes were present as homozygous recessive alleles, the solid white wing was expressed.
Observed offspring segregation ratios, value of χ2 test and corresponding P value for flower yellow wing spot for F1 and F2 generations involving yellow wing spotted faba bean (Gelber).
The χ2 test for the solid and spotted wing phenotype is presented in Table 4. All 11 populations fit the segregation ratio of 3:1 for solid wing:spotted wing. This confirms that the solid wing phenotype is controlled by a single recessive gene. This gene controls the spot pattern on the wing petal, rather than petal color, like the other genes discussed previously.
Observed offspring segregation ratios, value of χ2 test and corresponding P value for spotted and solid wing petals for F1 and F2 generations of faba bean crosses.
The genotype and phenotype of cross P47-1 × Gelber are given in Table 5. This cross was evaluated as a trihybrid. Eight distinct phenotypic classes were observed. The F1 plants produced wild-type flowers. In F2 phenotypes, solid brown, solid yellow, pink/brown (standard/wing), pink/yellow, and red/yellow were observed.
Genotypes proposed for cross P47-1 × Gelber along with the F1 and segregating trihybrid F2 faba bean plants.
Discussion
Here we present results of multiple F1 crosses and F2 populations derived from contrasting parental lines from different origins and flower colors. Our results confirmed the previous results on the genetics of red and yellow-spotted wing flowers (Cabrera, 1988; Sjödin, 1971). Our crosses also led to the discovery of combinations of red and yellow flower colors that have not been previously reported. As most faba bean pollen and flowers contain the pyrimidine glycosides v-c (Khazaei et al., 2019), the incorporation of low v-c into new flower phenotypes is critical.
Gregor Mendel’s laws for inheritance studies are still relevant today. There are several examples of plants that follow Mendelian inheritance patterns for flower color. Aside from pea (Pisum sativum), Mendel’s model plant (Ellis et al., 2011), the flower colors of many other ornamental plants are controlled by qualitative genes that follow Mendelian genetics (Anderson, 2005). The inheritance of anthocyanin in petunia (Petunia hybrida) flowers is controlled by two independent genes (Griesbach, 1996, 2005). In Stokes aster (Stokesia laevis), flower color is controlled by at least three loci and is consistent with simple Mendelian genetics (Barb et al., 2008). A three-gene model was also proposed for the inheritance of flower color in Anagallis monelli (Freyre and Griesbach, 2004). Due to these consistencies over many species, the hypothesis for flower color in faba bean was developed based on simple Mendelian genetic models. They have also been used to describe many other traits in faba bean such as hilum color, seedcoat color, flower color, pigmentation and dwarfism (Khazaei et al., 2014; O’Sullivan and Angra, 2016; Zanotto et al., 2020a).
The genetics of red flower color had limited published work in this species. A publicly available red-flowered faba bean called “1778” was the initial source of the red flower used to develop P47-1, the red-flowered line used in this study. In pea, two genes influenced the regulation of purple flower color (Moreau et al., 2012), which is in agreement with our results on faba bean. Our finding that the yellow wing spot is controlled by a single recessive gene, is consistent with the findings of Cabrera (1988); however, the relationship between yellow wing spot and white wing has not been previously reported. Our results showed an epistatic relationship for wing spot color with both zt1 and zt2. The yellow pigment appears to be a mutation of the black pigment.
Flower color in the P47-1 × Gelber population and its reciprocal segregated for three genes, making them trihybrids. Trihybrids are much more complex to classify due to the large population size required. In our study, the population sizes were not large enough to accurately test trihybrid segregation ratios of 27:9:9:9:3:3:3:1. These crosses had similar phenotypes to dihybrid crosses with red flower (red × non-red), but there was another set of phenotypes in which the brown pigmentation was substituted for yellow. We performed test crosses to determine the genotypes of the rare phenotypes. A cross between a solid yellow flower phenotype and a red flower phenotype resulted in an F1 plant with dark brown flowers. Another cross between a pink/yellow flower and red flower resulted in a 1:1 ratio of red:pink/brown flowered F1 hybrids. These test cross phenotypes were consistent with the proposed genotypes in Table 5, and indicate the logical assumption that the phenotype expressing both yellow and red is the most recessive genotype, expressing both the double recessive red genes and the single recessive yellow gene (Table 5) for the trihybrid cross of Gelber × P47-1 and its reciprocal.
Throughout the crosses, many colors and patterns were recorded, although additional rare phenotypes were discovered (e.g., pink standard/yellow wing and red standard/yellow wing, Table 5) by combining the yellow spot mutation with the double recessive red flower. Previous reports indicate the existence of a solid yellow flower, and the pink and yellow flower (Cabrera, 1988), but there are no reports of the existence of a red and yellow flower. This phenotype arose from a trihybrid cross between red-flowered and yellow spot flowered (P47-1 × Gelber). The red and yellow flower phenotype occurs only in the complete homozygous recessive genotype from the trihybrid, which is expected at a frequency of 63:1 genotype in a trihybrid.
Faba bean flowers exhibit color variation within the same color classes, and varying color patterns. The standardized color of the parent source of crimson red-flowered in this study was Rhodamine purple 29/1 according to the horticultural color chart (Wilson, 1938). Very few of the red flowers in the F2 populations fit this same color standard. There were gradients of colors within the same color classes, with clear variation in intensity and hue. Within the red phenotype, for example, the pigmentation exhibited hues from burgundy to purple. For example, the red flower phenotype derived from crosses with IG 114476 and Rinrei was a cooler tone and appeared purple, whereas red flower types derived from an Aurora/2 background had warmer red tones. Environmental conditions such as temperature, light intensity, pH, or even insect pressure may influence the flower color (Oh et al., 2014). Another unique aspect of this work was the presence of reciprocal crosses to study the genetics of red flower color. When P47-1 was used as a male parent, a better fit to 15:1 ratio was observed than a female parent (Table 2). Faba bean flower color varies in intensity, hue, and pattern, properties that are influenced by the environmental stresses that can cause anthocyanin upregulation, the maternal effects, and genotype-specific gene interactions with color.
Our preliminary results of biochemical analysis revealed that red flower color in faba bean was produced by anthocyanins. For brown, pink, and yellow flowers, no anthocyanins were found in flower tissue (manuscript in preparation). Anthocyanins are red-purple pigments that can be found in many plant species (reviewed in Ng and Smith, 2016). A recent study also showed that flavonols were abundant in flowers of white-flowered faba beans (Zanotto et al., 2020b).
Conclusions
In faba bean, red flower and yellow wing spot are controlled by double recessive and single recessive genes, respectively. Although this explanation seems straightforward, faba bean flower color has many other environmental and genetic influences. Faba bean flower color also had epistatic genetic relationships between colors and within specific populations. This study was not only able to determine the inheritance of two unique faba bean flower colors, but a new flower color phenotype was reported for the first time.
Literature Cited
Anderson, N.O. 2005 Breeding flower seed crops, p. 53–86. In: M.B. McDonald and F.Y. Kwong (eds.). Flower seeds: biology and technology. CABI International, Wallingford, UK
Anthony, J.B. 2017 Peas and beans, p. 66–93. In: J. Atherton (ed.). Crop production science in horticulture (series). CABI, Oxfordshire, UK
Barb, J.G., Werner, D.J. & Griesbach, R.J. 2008 Genetics and biochemistry of flower color in stokes aster J. Amer. Soc. Hort. Sci. 133 569 578
Burr, F. 1863 Field and Garden Vegetables of America. Crosby and Nichols, Boston
Cabrera, A. 1988 Inheritance of flower color in Vicia faba L. FABIS Newslett. 22:3–7
Cabrera, A. & Martin, A. 1989 Analysis of genetic linkage in faba bean (Vicia faba L.). FABIS Newslett. 24:3–5
Crofts, H.J., Evans, L.E. & McVetty, P.B.E. 1980 Inheritance, characterization and selection of tannin-free faba beans (Vicia faba L.) Can. J. Plant Sci. 60 1135 1140
Duc, G. 1997 Faba bean (Vicia faba L.) Field Crops Res. 53 99 109
Duc, G., Aleksić, J.M., Marget, P., Mikic, A., Paull, J., Redden, R.J., Sass, O., Stoddard, F.L., Vandenberg, A., Vishnyakova, M. & Torres, A.M. 2015 Faba bean, p. 141–178. In: A. De Ron (ed.). Grain legumes. Handbook of plant breeding, vol. 10. Springer, New York, NY
Ellis, T.N., Hofer, J.M., Timmerman-Vaughan, G.M., Coyne, C.J. & Hellens, R.P. 2011 Mendel, 150 years on Trends Plant Sci. 16 590 596
Erith, A.G. 1930 The inheritance of colour, size form of seeds, and of flower colour in Vicia faba L Genetica 12 477 510
Freyre, R. & Griesbach, R.J. 2004 Inheritance of flower color in Anagallis monelli L HortScience 39 1220 1223
Fukuta, N., Fujiok, S., Takatsuto, S., Yoshida, S., Fukuta, Y. & Nakayama, M. 2004 ‘Rinrei’, a brassinosteroid-deficient dwarf mutant of faba bean (Vicia faba L.) Physiol. Plant. 121 506 512
Griesbach, R.J. 1996 The inheritance of flower colour in Petunia hybrida Vilm J. Hered. 87 241 245
Griesbach, R.J. 2005 Biochemistry and genetics of flower color, p. 89–114. In: J. Janick (ed.). Plant Breeding Reviews. vol. 25. John Wiley & Sons, Inc., Hoboken, NJ
Gutierrez, N. & Torres, A.M. 2019 Characterization and diagnostic marker for TTG1 regulating tannin and anthocyanin biosynthesis in faba bean Sci. Rep. 9 16174
Khazaei, H., O’Sullivan, D.M., Sillanpää, M.J. & Stoddard, F.L. 2014 Genetic analysis reveals a novel locus in Vicia faba decoupling pigmentation in the flower from that in the extra-floral nectaries Mol. Breed. 34 1507 1513
Khazaei, H., Purves, R.W., Hughes, J., Link, W., O’Sullivan, D.M., Schulman, A.H., Björnsdotter, E., Geu-Flores, F., Nadzieja, M., Andersen, S.U., Stougaard, J., Vandenberg, A. & Stoddard, F.L. 2019 Eliminating vicine and convicine, the main anti-nutritional factors restricting faba bean usage Trends Food Sci. Technol. 91 549 556
Khazaei, H. & Vandenberg, A. 2020 Seed mineral composition and protein content of faba beans (Vicia faba L.) with contrasting tannin contents Agronomy 10 511
Martín, A., Cabrera, A. & López Medina, J. 1991 Antinutritional factors in faba bean. Tannin content in Vicia faba: Possibilities for plant breeding CIHEAM - Options Méditerr. 10 105 110
Metz, P.L.J., van Norel, A., Buiel, A.A.M. & Helsper, J.P.F.G. 1992 Inheritance of seedling colour in faba bean (Vicia faba L.) Euphytica 59 231 234
Moreau, C., Ambrose, M.J., Turner, L., Hill, L., Ellis, T.N. & Hofer, J.M. 2012 The b gene of pea encodes a defective flavonoid 3′, 5′-hydroxylase, and confers pink flower colour Plant Physiol. 159 759 768
Ng, J. & Smith, S.D. 2016 How to make a red flower: The combinatorial effect of pigments AoB Plants 8 plw013
Oh, S., Warnasooriya, S.N. & Montgomery, B.L. 2014 Mesophyll-localized phytochromes gate stress- and light-inducible anthocyanin accumulation in Arabidopsis thaliana Plant Signal. Behav. 9 e28013
O’Sullivan, D.M. & Angra, D. 2016 Advances in faba bean genetics and genomics Front. Genet. 7 150
Picard, J. 1976 Aperçu sur l’héredité du caractère absence de tannins dans les graines de féverole (Vicia faba L.). Ann. Amélio Plantes 26:101–106
R Development Core Team 2019 R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. <http://www.R-project.org>
Rowlands, D.G. & Corner, J.J. 1962 Genetics of pigmentation in broad beans (Vicia faba L.). Proc. of 3rd Congress of Eucarpia, Paris, p. 229–234
Sjödin, J. 1971 Induced morphological variation in Vicia faba L Hereditas 67 155 180
Warsame, A.O., O’Sullivan, D.M. & Tosi, P. 2018 Seed storage proteins of faba bean (Vicia faba L): Current status and prospects for genetic improvement J. Agr. Food Chem. 66 12617 12626
Watson, B. 1996 Taylor’s guide to heirloom vegetables. Houghton Mifflin, Boston, NY
Webb, A., Cottage, A., Wood, T., Khamassi, K., Hobbs, D., Gostkiewicz, K., White, M., Khazaei, H., Ali, M., Street, D., Stoddard, F.L., Maalouf, F., Ogbonnaya, F., Link, W., Thomas, T. & O’Sullivan, D.M. 2016 A SNP-based consensus genetic map for synteny-based trait targeting in faba bean (Vicia faba L.) Plant Biotechnol. J. 14 177 185
Wilson, R.F. 1938 Horticultural colour chart. Henry Stone & Son Ltd., Banbury, UK
Zanotto, S., Vandenberg, A. & Khazaei, H. 2020a Development and validation of a robust KASP marker for zt2 locus in faba bean (Vicia faba) Plant Breed. 139 375 380
Zanotto, S., Khazaei, H., Elessawy, F.M., Vandenberg, A. & Purves, R.W. 2020b Do faba bean genotypes carrying different zero tannin genes (zt1 and zt2) differ in phenolic profiles? J. Agr. Food Chem. 68 7530 7540
