Abstract
Red radish (Raphanus sativus L.) pigments are attractive natural colorants, because they resist heat and light and give foods an attractive red hue. The major limitation of using red radish colorants in food is their distinctly undesirable flavor resulting from the degradation of glucoraphasatin, which is a glucosinolate predominantly contained in radishes. In the present study, we developed a novel red radish cultivar, ‘Saint Rouge’, for odorless food coloring. ‘Saint Rouge’ was derived from ‘AD-1’, a red radish cultivar used for colorant and bred by introducing a gene conferring glucoraphasatin-less trait using marker-assisted selection. ‘Saint Rouge’ offers a ca. 1.2–1.9-fold greater index for the total color amount than that of the original ‘AD-1’. Meanwhile, the color tone of the extract was quite similar to that of ‘AD-1’. ‘Saint Rouge’ did not contain glucoraphasatin. Therefore, the amount of methyl mercaptan, dimethyl disulfide and dimethyl trisulfide, substances responsible for off-flavors and the sulfur-like odor of radish pigment extracts, was drastically lower in ‘Saint Rouge’ than in ‘AD-1’. These results demonstrate that ‘Saint Rouge’ is a breakthrough red radish cultivar for odorless food colorants.
Introduction
Radish (Raphanus sativus L.) belongs to the Brassicaceae family and is cultivated and consumed worldwide as an enlarged hypocotyl and root (hereafter referred to as the root). The root color is an economically important trait of radishes, ranging from white, black, pink, red, purple, yellow, and green. The major red pigment that accumulates in red radish roots is a pelargonidin-type anthocyanin, which is used as a natural source of food colorant (Ishikura and Hayashi 1963).
Red radish colorant is resistant to heat and light, and has attractive red hue characteristics (Chen et al. 2016, Giusti and Wrolstad 1996). As consumer demand for natural colorants in food products has increased because of both legislation and health concerns, radish colorants are expected to be a desirable source of such ingredients. The major limitation in using red radish colorants in food is the distinct undesirable flavor associated with its preparation. Many attempts have been made to establish methods for extracting off-flavor-free red radish root anthocyanins without affecting their yield and structure (Chen et al. 2016, Gao et al. 2014, Rodriguez-Saona et al. 2001). However, food colorants from radishes have not yet been utilized in many foods, as adding purification and deodorization processes might decrease anthocyanin yields and increase costs, and off-flavors can reoccur with time, even after deodorization.
Glucoraphasatin, a major glucosinolate (GSL) contained in radish roots (Ishida et al. 2012), is hydrolyzed by myrosinase enzyme to generate a radish-specific isothiocyanate, raphasatin, during processing (Friis and Kjær 1966). Raphasatin is chemically unstable, and its degradation by reaction with water produces sulfurous odors and yellow pigments, which are characteristics of processed radish products (Ishida and Morimitsu 2013, Ozawa et al. 1990a, 1990b). The major contributors to the off-flavors of processed radish products are methyl mercaptan, dimethyl disulfide, dimethyl trisulfide, and raphasatin; these are sulfur-containing compounds, with the former three having fecal, sulfur, and stink odors (Chen et al. 2017).
A spontaneous mutant which contains glucoerucin predominantly instead of glucoraphasatin was identified from the Japanese white radish landrace ‘Nishimachi-riso’ (Ishida et al. 2015). Glucoerucin is almost absent in wild-type radish, and its hydrolysis generates the pungent compound erucin. Unlike raphasatin, the degradation products of erucin do not produce an odor or yellow pigments. Genetic analysis revealed that this glucoraphasatin-less trait is controlled by a single recessive gene, so novel white radish cultivar ‘Daikon parental line No.5 (DPL5)’ (Registration number 22662), which lacks glucoraphasatin, has been developed by selfing and selection of ‘Nishimachi-riso’ (Ishida et al. 2015). Glucoraphasatin is synthesized by the enzyme glucoraphasatin synthase 1 (GRS1) using glucoerucin as a substrate. The mutant GRS1 gene (grs1) is deficient in glucoraphasatin synthesis because of an 8.6-kb insertion which leads to an in-frame stop codon (Kakizaki et al. 2017). Using DNA markers to discriminate between the wild type and mutant genotypes (Endo et al. 2023), the new cultivars ‘Yuhaku’ and ‘Sarah White’ have been developed (Ishida and Morimitsu 2020).
Typical Japanese radish pickles (Takuanzuke), which are produced from wild-type radishes, have a distinct odor and yellow color, whereas those from ‘Yuhaku’ and ‘Sarah White’, have extremely low amounts of sulfurous odors with no yellow pigments (Ishida and Morimitsu 2020). Therefore, introducing the glucoraphasatin-less trait into red radish can solve the problem of off-flavors in red radish colorants products.
The aim of this study was to develop a radish F1 cultivar as an ingredient of food colorants with high pigment content and free from off-flavor. Here, we report the breeding process and characteristics of the new cultivar ‘Saint Rouge’.
Materials and Methods
Breeding
Plant materials
‘DPL5’ and NMR366N, an inbred line bred at the Institute of Vegetable and Floriculture Science, NARO (Tsu City, Japan), have both white skin and internal roots, and were used as the donor of grs1. cmsNMR366N was bred from a cytoplasmic male sterility (CMS) line using four successive backcrossings, with NMR366N as the recurrent parent. ADS is an inbred line obtained by selection of ‘AD-1’. ‘AD-1’ is a red radish cultivar with high red pigment content and used as a raw material for food colorants at San-Ei Gen F.F.I., Inc. (Osaka, Japan). The uniformity of its trait (e.g., pigment content, root shape, and root weight) is relatively low as it is an open-pollinated cultivar. Although the genetic basis underlying the elevated red pigment content in ADS remains unclear, we successfully selected a stable, high-pigment inbred line by selfing ‘AD-1’. Furthermore, analysis of the F2 population derived from a cross between ADS and DPL5 revealed continuous segregation of root color from white to red, suggesting that this trait behaves as a quantitative trait. Ten-an-Koshin2-1-10-5 is a selfed line derived from ‘Ten-an-Koshin’ (Sakata Seed Co., Yokohama, Japan), which is a Chinese radish cultivar with green skin and red internal root color, and used as an inbred line of red radish.
Genotyping of grs1
Genomic DNA was isolated from young leaves of seedlings, as described by Thomson and Henry (1995). Markers designed to discriminate between the wild-type and mutant grs1 (Endo et al. 2023) were used for genotyping. PCR amplification was carried out in a 10-μL volume containing 1.0 μL DNA solution, 1.6 μM forward primer (Rs270), 0.8 μM each reverse primer (Rs271 and Rs272), 2 × Quick Taq HS Dye Mix (Toyobo, Osaka, Japan) in a GeneAmp PCR System 9700 (Applied Biosystems, Foster City, CA, USA). The reaction was performed with the following parameters: 1 cycle of 94°C for 2 min; 30 cycles of 94°C for 30 s, 57°C for 30 s and 68°C for 1 min and final 68°C for 7 min. Polymorphism was detected by separating the whole volume of treated DNA on 1.5% agarose gel (0.5 × TBE [Tris borate EDTA]).
Field trials
‘AD-1’ and a red radish F1 cultivar ‘Benikururi’ (Matsunaga Seed Co., Ltd., Konan, Japan) were used as a control and a standard cultivar, respectively. ‘Sarah White’ which is a white radish F1 cultivar derived from DPL5 and has glucoraphasatin-less trait, was also used as a control cultivar for GSL analysis. The same seed lot of each cultivar was used for all trials.
Field trials for property evaluation were conducted in experimental fields at the NARO Institute of Vegetable and Floriculture Science (Kannondai, Tsukuba, Ibaraki, Japan; 36°2ʹN, 140°6ʹE), San-Ei Gen F.F.I., Inc. (Uguisunomori, Kawanishi, Hyogo, Japan; 34°50ʹN, 135°25ʹE), and in a farmer’s field in Eniwa, Hokkaido, Japan (42°55ʹN, 141°33ʹE), for two years (2015 and 2016). An outline of the trials is presented in Table 1. For each trial, seeds were sown in rows 0.60 m apart, with 0.20 m spacing (8.3 plants/m2), and 20 plants per cultivar were grown with three replications. At 63–90 d after sowing, 10 normal plants per cultivar and replicate were harvested, and agronomic traits were evaluated.
Table 1.Cultivation conditions for property evaluation
| Place |
Year |
Sowing date |
Harvest date |
Fertilizer (kg/10 a) |
| N |
P2O5 |
K2O |
| Tsukuba |
2015 |
9/14 |
12/10 |
6.4 |
4.0 |
5.6 |
| 2016 |
9/9 |
12/8 |
10.7 |
6.7 |
9.3 |
| Kawanishi |
2015 |
9/11 |
12/1 |
13.0 |
10.0 |
12.0 |
| 2016 |
9/9 |
12/6 |
13.0 |
10.0 |
12.0 |
| 2023–2024a |
9/19 |
3/14 |
13.0 |
10.0 |
12.0 |
| Eniwa |
2015 |
7/13 |
9/15 |
5.6 |
4.4 |
5.2 |
| 2016 |
7/14 |
9/15 |
5.6 |
4.4 |
5.2 |
a For analysis of color tone and aroma components.
Among the 10 roots used for agronomic trait evaluation, six roots per cultivar and replicate were used for GSL analysis. Extraction and high-performance liquid chromatography analyses of GSL were performed as described by Kakizaki et al. (2017).
Properties of pigment
Samples and preparation of radish extract
For the evaluation of color value, the same individuals were used as those for agronomic trait evaluation (See Field trials section). For analysis of color tone and aroma components, each five roots of ‘Saint Rouge’ and ‘AD-1’ grown in the experimental fields at San-Ei Gen F.F.I., Inc. (Kawanishi, Hyogo, Japan) in 2023–2024 (Table 1) were used.
Whole roots of each sample were cut into small pieces then 30 g of them were immersed into 120 g of 0.5% (w/w) of sulfuric acid aqueous solution. The mixture was stirred for 1 h and then maintained at room temperature for 18 h. After extraction, the mixture was filtered through No. 2 filter paper (Toyo Roshi Co., Ltd., Tokyo, Japan).
Determination of color value
Radish extract was accurately weighed at 3.5 g, made up to 100 mL with McIlvaine buffer (pH 3.0) and the absorbance was measured. A V-760 UV-visible spectrophotometer (JASCO Corporation, Tokyo) was used to determine the absorbance at 513 nm in a 1-cm quartz cell (1.0 cm light path length). The color values of the extracts were calculated according to Japan’s Specifications and Standards for Food Additives, 10th edition (Ministry of Health, Labour and Welfare; Consumer Affairs Agency, Government of Japan 2024). The color value of the radish was calculated by multiplying the color value of the extract by the weight of the extract and dividing by the weight of the sliced radish pieces. The index of the total amount of color was calculated by multiplying the root weight with the color value of the radish.
Determination of color properties
The extract was diluted with McIlvaine buffer (pH 3.0) to OD513nm = 0.8 and transferred into a quartz cell. CIELAB parameters (L*, a*, and b*) were determined using a V-760 UV-visible spectrophotometer. The color difference ΔE*ab between the color pair was calculated as the Euclidean distance between the two points in the color space.
Sampling of volatile compounds
The volatile compounds in the radish extract were investigated using solid-phase microextraction (SPME) coupled with gas chromatography-mass spectrometry (GC/MS). SPME was performed with 65 μm polydimethylsiloxane/divinylbenzene fiber (Sigma-Aldrich Co., St. Louis, MO, USA). 1 g of radish extracts was transferred into 20 mL glass vials, and 1 μL of 100 ppm 3-heptanol was added as an internal standard. The vials were sealed tightly and statically equilibrated at 40°C for 30 min, followed by 30 min of SPME exposure in the headspace at 40°C. After sampling, the fiber was placed into the injection port of the GC/MS and thermally desorbed for 10 min at 250°C.
GC/MS analysis
The GC/MS was performed three times with an Agilent 7890A gas chromatograph (Agilent Technologies, Santa Clara, CA, USA) equipped with a 5975C inert XL mass spectrometric detector (Agilent Technologies). An injector in the splitless mode was maintained at 250°C. The separation was achieved on a DB-WAX UI (60 m × 0.25 mm i.d., 0.25 μm film thickness, Agilent Technologies). Helium was used as a carrier gas at a flow rate of 1.6 mL/min. The inlet temperature was set at 250°C. The column temperature was initially maintained at 50°C for 2 min, increased at a rate of 3°C/min to 220°C, and maintained at 220°C. A mass range of m/z 30–300 was recorded in the full-scan mode.
Methyl mercaptan, dimethyl disulfide, dimethyl trisulfide, and raphasatin, the dominant aromatic compounds in the radish extracts, were identified by comparing the mass spectra and linear retention indices of the samples and standards. Standards were purchased from Tokyo Chemical Industry (Tokyo, Japan) and Sigma-Aldrich (Tokyo, Japan).
Results and Discussion
Breeding process and application for registration
In this study, we aimed to develop a novel radish F1 cultivar characterized by high pigment content and the absence of off-flavor. To introduce the high-pigment trait, we employed recurrent backcrossing using the high-pigment line ADS as the donor parent. To eliminate the radish odor, we introduced the glucoraphasatin-less phenotype controlled by the recessive grs1 gene through marker-assisted selection (MAS). Because grs1 is recessive, both the seed parent (CMS line) and the pollen parent were required to be homozygous for the allele to ensure stable expression of the odor-reduced phenotype in the resulting F1 hybrid. In addition, the selfing process is necessary to select heterozygous individuals from backcrossed progenies by evaluating the phenotype. The use of a co-dominant DNA marker to discriminate the GRS1 genotype (Endo et al. 2023) enables the immediate selection of individuals heterozygous for GRS1 from the first generation of backcrossed progenies, leading to the breeding of the F1 cultivar ‘Saint Rouge’ within six years of beginning breeding.
Through this integrated breeding strategy, we successfully developed a new F1 cultivar combining high pigment content with an off-flavor-free phenotype. The process is summarized in Fig. 1. The seed (cmsRSP14) and pollen (RPP8) parents were developed using the following procedure:
cmsNMR366N was crossed with ADS in 2011 (sowing year) and backcrossed twice with ADS in 2012–2013. The BC2 plant was crossed in 2014 with another BC2 plant that originated from the development of RPP8 (hereafter referred to as TBC2), and the resulting F1 was then crossed in 2015 with the selfed TBC2 plant (TBC2S1). The resulting line was named cmsRSP14. During crossing or backcrossing, promising individuals or lines were selected by genotyping grs1, evaluating the red color value of the root, and assessing the stability of the CMS in field tests.
Ten-an-Koshin2-1-10-5 was crossed with ‘DPL5’ in 2010. In 2012, a single F2 plant was selected and crossed with ADS. Subsequently, two successive backcrosses were performed using ADS as the recurrent parent during 2013–2014. The BC2 plant was selfed twice, and the resultant line with a high pigment content and lacking glucoraphasatin was named RRP8. As described above, the BC2 plant (TBC2) was used as the recurrent parent for the development of cmsRSP14. During backcrossing or selfing, promising individuals or lines were selected by genotyping grs1 and the red color strength of the roots in field tests. F1 derived from a cross between cmsRSP14 and RPP8 was named ‘Daikon Yaei-Kou 1’, and subsequently used for evaluating the performance.
‘Daikon Yaei-Kou 1’ was submitted for registration in 2018 under the name ‘Saint Rouge’. In 2022, ‘Saint Rouge’ was officially registered as a radish cultivar by the Ministry of Agriculture, Forestry, and Fisheries of Japan (registration number 28955).
Agronomic traits
Agronomic traits of ‘Saint Rouge’ were evaluated over two years at three sites (Tsukuba, Kawanishi, Eniwa). Mostly, root weight and diameter of ‘Saint Rouge’ were significantly smaller than those of ‘AD-1’ and ‘Benikururi’ (Fig. 2A, Table 2). Root length was similar among the three cultivars. Leaf weight and leaf length of ‘Saint Rouge’ and ‘AD-1’ were significantly bigger than those of ‘Benikururi’. As such, top/root ratio of ‘Saint Rouge’ was significantly bigger than ‘Benikururi’ and mostly similar to or bigger than ‘AD-1’. Compared trials at Eniwa with those at Tsukuba and Kawanishi, root characteristics of each cultivar tended to be smaller; and top/root ratio of each cultivar was the biggest at Eniwa in both years.
Table 2.Agronomic traits of ‘Saint Rouge’
| Place |
Year |
Cultivar |
Root weight
(g) |
Root lengtha (cm) |
Maximum root diameter (cm) |
Leaf weight
(g) |
Leaf length (cm) |
Top/Root
ratiob |
| Tsukuba |
2015 |
Saint Rouge |
263c |
14.4a |
6.0c |
239b |
51.7b |
0.9a |
| AD-1 |
345b |
14.1a |
7.0b |
345a |
57.3a |
1.0a |
| Benikururi |
481a |
16.9a |
8.2a |
132c |
31.3c |
0.3b |
| 2016 |
Saint Rouge |
297b |
13.0ab |
6.3b |
194a |
44.4b |
0.7a |
| AD-1 |
431a |
13.9a |
7.6a |
223a |
48.4a |
0.5b |
| Benikururi |
468a |
12.4b |
8.0a |
96b |
27.3c |
0.2c |
| Kawanishi |
2015 |
Saint Rouge |
371b |
18.6a |
6.2c |
303b |
57.1b |
0.9b |
| AD-1 |
488b |
19.0a |
7.1b |
504a |
70.0a |
1.2a |
| Benikururi |
924a |
18.3a |
9.3a |
195c |
34.8c |
0.2c |
| 2016 |
Saint Rouge |
217b |
11.1a |
6.0b |
127ab |
39.9a |
0.6a |
| AD-1 |
243b |
11.2a |
6.5b |
146a |
39.6a |
0.6a |
| Benikururi |
461a |
11.3a |
8.2a |
98b |
25.6b |
0.2b |
| Eniwa |
2015 |
Saint Rouge |
206c |
11.0a |
5.9c |
297a |
57.1a |
1.5a |
| AD-1 |
275b |
12.6a |
6.6b |
326a |
58.8a |
1.3b |
| Benikururi |
447a |
12.5a |
8.5a |
114b |
30.8b |
0.3c |
| 2016 |
Saint Rouge |
100c |
11.2a |
4.2c |
154b |
48.6b |
1.6a |
| AD-1 |
166b |
12.1a |
5.4b |
216a |
53.2a |
1.3b |
| Benikururi |
283a |
12.9a |
7.2a |
102c |
28.1c |
0.4c |
a Length of thickened root (measured from the stem base to the point where the root becomes non-thickened).
b Leaf weight/Root weight.
Values followed by the same letter are not significantly different at the 5% level, as determined using the Tukey–Kramer HSD test.
Although ‘Saint Rouge’ is an F1 cultivar, heterosis is not evident. This is likely because both parental lines were developed through backcrossing with ADS, which resulted in a high degree of genetic similarity.
Composition and content of GSLs
GSL composition of ‘Saint Rouge’ was similar to ‘Sarah White’, and completely different from those of ‘AD-1’ and ‘Benikururi’ (Table 3). In all trials, glucoraphasatin, which is a major GSL in radish, was not detected in ‘Saint Rouge’ and ‘Sarah White’. Instead of glucoraphasatin, glucoerucin was the major GSL in both cultivars. Whereas in ‘AD-1’ and ‘Benikururi’, glucoraphasatin was a major GSL and glucoerucin was detected in a small amount or not detected. In all trials, total GSL content of ‘Saint Rouge’ was smaller than ‘AD-1’, and at 65.2–85.3% of ‘AD-1’.
Table 3.Glucosinolate contents in ‘Saint Rouge’ roots
| Place |
Year |
Cultivar or line |
Glucosinolates (μmol/g DW) |
|
Percent of |
| GERa |
GRSb |
Total |
GER/Total
(%) |
GRS/Total
(%) |
| Tsukuba |
2015 |
Saint Rouge |
25.8a |
0.0b |
30.8b |
|
83.8 |
0.0 |
| Sarah White |
18.4b |
0.0b |
19.5c |
|
94.4 |
0.0 |
| AD-1 |
0.6c |
30.1a |
36.1a |
|
1.7 |
83.4 |
| Benikururi |
0.0c |
28.1a |
32.0ab |
|
0.0 |
87.8 |
| 2016 |
Saint Rouge |
21.9a |
0.0c |
27.9b |
|
78.0 |
0.0 |
| Sarah White |
15.3b |
0.0c |
15.8c |
|
96.8 |
0.0 |
| AD-1 |
0.4c |
32.5a |
42.8a |
|
0.9 |
83.4 |
| Benikururi |
0.0d |
23.9b |
30.4b |
|
0.0 |
83.4 |
| Kawanishi |
2015 |
Saint Rouge |
26.7a |
0.0c |
30.5ab |
|
87.5 |
0.0 |
| Sarah White |
16.9b |
0.0c |
17.2c |
|
98.3 |
0.0 |
| AD-1 |
0.7c |
34.4a |
39.8a |
|
1.8 |
86.4 |
| Benikururi |
0.0c |
22.4b |
26.8bc |
|
0.0 |
83.6 |
| 2016 |
Saint Rouge |
22.5a |
0.0c |
25.9c |
|
86.9 |
0.0 |
| AD-1 |
0.4b |
32.6a |
37.8a |
|
1.1 |
86.2 |
| Benikururi |
1.0b |
26.2b |
32.1b |
|
3.1 |
81.4 |
| Eniwa |
2015 |
Saint Rouge |
35.2a |
0.0c |
40.3b |
|
87.7 |
0.0 |
| AD-1 |
1.1b |
53.4a |
59.5a |
|
1.8 |
89.7 |
| Benikururi |
0.3b |
36.3b |
40.5b |
|
0.7 |
89.6 |
| 2016 |
Saint Rouge |
45.3a |
0.0c |
52.7b |
|
86.0 |
0.0 |
| AD-1 |
1.3b |
59.2a |
65.5a |
|
2.0 |
90.4 |
| Benikururi |
0.6b |
52.8b |
56.8ab |
|
1.1 |
93.0 |
a Glucoerucin.
b Glucoraphasatin.
Values followed by the same letter are not significantly different at the 5% level, as determined using the Tukey–Kramer HSD test.
‘Saint Rouge’ has a thick root periderm, and pigments are particularly accumulated in this part (Fig. 2B).
Both the color value and index for the total color amount varied greatly from year to year and place; those of ‘Saint Rouge’ were the highest at Kawanishi in 2015 and the lowest at Eniwa in 2016. However, in all years and across all experimental sites, the color value of ‘Saint Rouge’ ranged from 10.2–12.5, which was 2.0–2.2 times and 3.9–6.9 times higher than ‘AD-1’ and ‘Benikururi’, respectively (Table 4). Similarly, the index for the total color amount was 1.2–1.9 times larger in ‘Saint Rouge’ than in ‘AD-1’. This outcome is achieved through successive backcrossing and selection using the highly pigmented line ADS. Therefore, as a colorant ingredient, ‘Saint Rouge’ performs better than ‘AD-1’ and is suitable for fall-sowing for winter harvest in mild climates, and for summer-sowing for fall harvest in cold areas, with the former sowing strategy potentially being high-yielding.
Table 4.Color value and index for the total color amount of ‘Saint Rouge’ root extract
| Place |
Year |
Cultivar |
Color value |
Index for the total color amounta |
| Tsukuba |
2015 |
Saint Rouge |
12.1a |
2814.3a |
| AD-1 |
5.6b |
1845.0b |
| Benikururi |
3.1c |
1456.8c |
| 2016 |
Saint Rouge |
10.2a |
2750.6a |
| AD-1 |
5.2b |
2058.8b |
| Benikururi |
1.7c |
807.7c |
| Kawanishi |
2015 |
Saint Rouge |
12.5a |
4084.4a |
| AD-1 |
5.8b |
2599.9b |
| Benikururi |
2.5c |
2248.4b |
| 2016 |
Saint Rouge |
11.8a |
2334.7a |
| AD-1 |
5.3b |
1208.9b |
| Benikururi |
2.0c |
899.3b |
| Eniwa |
2015 |
Saint Rouge |
12.4a |
2218.3a |
| AD-1 |
5.9b |
1442.5b |
| Benikururi |
2.1c |
913c |
| 2016 |
Saint Rouge |
10.3a |
917.5a |
| AD-1 |
5.0b |
755.2a |
| Benikururi |
1.5c |
403.7b |
a Color value × Root weight.
Values followed by the same letter are not significantly different at the 5% level, as determined using the Tukey–Kramer HSD test.
CIELAB (L*, a* and b*) parameters for ‘Saint Rouge’ and ‘AD-1’ were highly similar, and the color difference index (ΔE*ab) was very small (0.71) (Table 5), indicating little difference in color tones of ‘Saint Rouge’ and ‘AD-1’.
Table 5.Color tone of ‘Saint Rouge’ root extract
| Cultivar |
L* |
a* |
b* |
ΔE*aba |
| Saint Rouge |
81.1 |
43.3 |
8.65 |
– |
| AD-1 |
80.7 |
43.8 |
8.38 |
0.71 |
aΔE*ab = ((ΔL*)2 + (Δa*)2 + (Δb*)2)1/2.
The four volatile compounds responsible for the off-flavor and sulfur-like odor of the radish pigment extracts were determined using SPME GC-MS (Table 6).
Table 6.Relative peak area of volatile sulfur compounds in ‘Saint Rouge’ extract
| No. |
Compounds |
Liner retention indices (LRI) |
Quantitation ion
(m/z) |
Second confirmation ion (m/z) |
Third confirmation ion (m/z) |
Saint Rouge |
|
AD-1 |
t testd |
| Meana |
RSDb (%) |
Mean |
RSD (%) |
| 1 |
Metyl mercaptan |
700 |
47 |
48 |
45 |
0.001 |
33.0 |
|
0.010 |
21.7 |
** |
| 2 |
Dimethyl disulfide |
1072 |
94 |
45 |
79 |
0.315 |
11.2 |
|
2.012 |
22.2 |
** |
| 3 |
Dimethyl trisulfide |
1387 |
126 |
79 |
45 |
0.052 |
12.5 |
|
0.450 |
13.4 |
** |
| 4 |
Raphasatin |
2147 |
87 |
159 |
45 |
n.d.c |
– |
|
n.d. |
– |
|
| Sum |
|
|
|
|
0.369 |
|
|
2.472 |
|
|
| Intenal Standard |
|
|
|
|
|
|
|
|
|
|
| 3-Heptanol |
1300 |
59 |
69 |
87 |
– |
– |
|
– |
– |
|
a Mean of the peak area/internal standard peak area ratio (n = 3).
b Relative standard deviation.
c Not detected.
d ** means significantly different at the 1% level by t test.
Among them, methyl mercaptan, dimethyl disulfide, and dimethyl trisulfide were drastically lower (10.0–15.7%) in ‘Saint Rouge’ than in ‘AD-1’. Raphasatin was not detected in any of the cultivars.
Because the odor thresholds of most sulfur compounds are extremely low (Takeoka 2001), the types of foods for which red radish colorants can be used are limited. Acid protease processing and adsorption treatments have been developed to deodorize pigments in red radishes (Aoki et al. 2009). Moreover, a unique deodorizing technology has been reported that reduces dimethyl disulfide and dimethyl trisulfide to 15% (Kido 2008). In this study, the quantities of odorous compounds in ‘Saint Rouge’, which are responsible for radish off-flavors in radish pigment extracts (Chen et al. 2017), declined to 10.0–15.7% of those in ‘AD-1’ (Table 6). It has been suggested that radish cultivars with glucoraphasatin-less traits may solve the problem of off-flavor of colorants in radishes that haven’t been subjected to a deodorizing process. In fact, food products containing colorant derived from ‘Saint Rouge’ have been confirmed to be free of distinct odors that could cause practical problems (data not shown).
To summarize, the glucoraphasatin-less red radish cultivar ‘Saint Rouge’ was successfully developed as a source for food colorant. ‘Saint Rouge’ had compact root characteristics compared to the original ‘AD-1’ in most trials (Table 2). Meanwhile, the color value of ‘Saint Rouge’ was consistently and significantly higher than that of ‘AD-1’, regardless of year or location tested; consequently, the index for the total color amount of ‘Saint Rouge’ was 1.2–1.9 times that of ‘AD-1’ (Table 4). The color tone of ‘Saint Rouge’ root extract was quite similar to that of ‘AD-1’ (Table 5). Notably, levels of volatile compounds associated with off-flavors were markedly reduced in ‘Saint Rouge’ (Table 6). These findings indicate that anthocyanin pigments from ‘Saint Rouge’ show great promise as fundamentally new natural colorants, which can be utilized in a wide range of foods including beverages, processed foods, and sweets.
Author Contribution Statement
MI, TO, and KH designed this study. NF, MI, TO, TK and EI evaluated agronomic trait and bred ‘Saint Rouge’. NF performed MAS and analyzed GSLs, and TY and MI evaluated the properties of the pigment. TO and KH directed and supported the evaluation of pigment properties. In addition, TO conducted cultivation trials in Kawanishi. NF and MI drafted the manuscript.
Acknowledgments
The research funding for this study was provided in part by San-Ei Gen F.F.I., Inc. We are grateful to R. Nishioka and M. Kono at San-Ei Gen F.F.I., Inc. as well as Y. Kawamoto, S. Morimoto, and Y. Niina at NARO for their technical assistance.
Literature Cited
- Aoki, H., Y. Shibahara and M. Kadowaki (2009) JP4420594.
- Chen, W., E. Karangwa, J. Yu, S. Xia, B. Feng and X. Zhang (2016) Improving red radish anthocyanin yield and off flavor removal by acidified aqueous organic based medium. RSC Adv 6: 97532–97545.
- Chen, W., E. Karangwa, J. Yu, S. Xia, B. Feng, X. Zhang and C. Jia (2017) Characterizing red radish pigment off-odor and aroma-active compounds by sensory evaluation, gas chromatography-mass spectrometry/olfactometry and partial least square regression. Food Bioproc Tech 10: 1337–1353.
- Endo, R., H. Chikano, E. Itabashi, M. Kawasaki, T. Ohara and T. Kakizaki (2023) Large insertion in radish GRS1 enhances glucoraphanin content in intergeneric hybrids, Raphanobrassica (Raphanus sativus L. × Brassica oleracea var. acephala). Front Plant Sci 14: 1132302.
- Friis, P. and A. Kjær (1966) 4-methylthio-3-butenyl isothiocyanate, the pungent principle of radish root. Acta Chem Scand 20: 698–705.
- Gao, R., P. Jing, S. Ruan, Y. Zhang, S. Zhao, Z. Cai and B. Qian (2014) Removal of off-flavours from radish (Raphanus sativus L.) anthocyanin-rich pigments using chitosan and its mechanism(s). Food Chem 146: 423–428.
- Giusti, M.M. and R.E. Wrolstad (1996) Radish anthocyanin extract as a natural red colorant for maraschino cherries. J Food Sci 61: 688–694.
- Ishida, M., M. Nagata, T. Ohara, T. Kakizaki, K. Hatakeyama and T. Nishio (2012) Small variation of glucosinolate composition in Japanese cultivars of radish (Raphanus sativus L.) requires simple quantitative analysis for breeding of glucosinolate component. Breed Sci 62: 63–70.
- Ishida, M. and Y. Morimitsu (2013) Chemical changes of the breakdown compounds of glucosinolate in processed food of the daikon without containing 4-methylthio-3-butenyl glucosinolate. Journal of Japan Association on Odor Environment 44: 307–314 (in Japanese).
- Ishida, M., T. Kakizaki, Y. Morimitsu, T. Ohara, K. Hatakeyama, H. Yoshiaki, J. Kohori and T. Nishio (2015) Novel glucosinolate composition lacking 4-methylthio-3-butenyl glucosinolate in Japanese white radish (Raphanus sativus L.). Theor Appl Genet 128: 2037–2046.
- Ishida, M. and Y. Morimitsu (2020) Development of the novel Japanese white radish cultivars without containing 4-methylthio-3-butenyl glucosinolate which is causes of sulfurous odor and yellow pigments of manufacturing products. Regulation of Plant Growth & Development 55: 46–51 (in Japanese with English summary).
- Ishikura, N. and K. Hayashi (1963) Chromatographic separation and characterization of the component anthocyanins in radish root. Studies on anthocyanins, XXXVIII. Shokubutsugaku Zasshi 76: 6–13.
- Kakizaki, T., H. Kitashiba, Z. Zou, F. Li, N. Fukino, T. Ohara, T. Nishio and M. Ishida (2017) A 2-oxoglutarate-dependent dioxygenase mediates the biosynthesis of glucoraphasatin in radish. Plant Physiol 173: 1583–1593.
- Kido, H. (2008) New technology on natural colorants: New deodorized technology of red beat color and its application for foods. Food Chemicals 10: 51–56 (in Japanese).
- Ministry of Health, Labour and Welfare; Consumer Affairs Agency, Government of Japan (2024) Japan’s standards and specifications for food additives (10th ed.) https://jp-foodadd.net/110_Koteisyo_en.html.
- Ozawa, Y., S. Kawakishi, Y. Uda and Y. Maeda (1990a) Isolation and identification of a novel β-carboline derivative in salted radish roots, Raphanus sativus L. Agric Biol Chem 54: 1241–1245.
- Ozawa, Y., Y. Uda and S. Kawakishi (1990b) Generation of a β-carboline derivative, the yellowish precursor of processed radish roots, from 4-methylthio-3-butenyl isothiocyanate and l-tryptophan. Agric Biol Chem 54: 1849–1851.
- Rodriguez-Saona, L.E., M.M. Giusti, R.W. Durst and R.E. Wrolstad (2001) Development and process optimization of red radish concentrate extract as potential natural red colorant. J Food Process Preserv 25: 165–182.
- Takeoka, G. (2001) Volatile constituents of asafoetida. In: Takeoka, G.R., M. Güntert and K.-H. Engel (eds.) Aroma active compounds in foods, American Chemical Society, Washington, DC, pp. 33–44.
- Thomson, D. and R. Henry (1995) Single-step protocol for preparation of plant tissue for anaysis by PCR. Biotechniques 19: 394–397, 400.
