2022 Volume 72 Issue 4 Pages 297-305
‘Kitahonami’ is a soft red winter wheat (Triticum aestivum L.) cultivar that has high yield, good agronomic performance and good quality characteristics. It currently accounts for 73% of the wheat cultivation area of Hokkaido the northern island in Japan and 42% of Japan’s overall wheat cultivation. However, this cultivar is susceptible to Wheat yellow mosaic virus (WYMV). WYMV has become widespread recently, with serious virus damage reported in Tokachi and Ohotsuku districts, which are the main wheat production areas in Hokkaido. Here, we report a new wheat breeding line ‘Kitami-94’, which was developed over four years by repeated backcrossing with ‘Kitahonami’ using DNA markers for WYMV resistance linked to the Qym1 and Qym2 from ‘Madsen’. Basic maps of Qym1 and Qym2 were created and used to confirm that ‘Kitami-94’ reliably carried the two resistance genes. ‘Kitami-94’ demonstrated WYMV resistance, and had agronomic traits and quality equivalent to ‘Kitahonami’ except for higher polyphenol oxidase activity and lower thousand grain weight. ‘Kitami-94’ may be useful for elucidating the mechanism of WYMV resistance in the background of ‘Kitahonami’, and for developing new cultivars.
‘Kitahonami’ is a leading wheat cultivar in Japan, with high yield and good agronomic performance. This cultivar is a soft red winter wheat with good milling and noodle-making quality, low ash content, and excellent flour color (Yanagisawa et al. 2007). Since 2006, this cultivar has been recommended by the government of Hokkaido, the northern island of Japan where 66% of Japanese wheat is produced. ‘Kitahonami’ is highly valued by farmers, millers, and processors, and was grown on about 73% of the wheat cultivation area in Hokkaido in 2019 (Anonymous 2021).
Unfortunately, ‘Kitahonami’ is susceptible to Wheat yellow mosaic virus (WYMV). WYMV was first described in Japan by Sawada (1927). Initially, damage was reported primarily in western Japan (Ikata and Kawai 1940), but WYMV appeared in Hokkaido in 1991 (Kusume et al. 1997). Since that time, it has become widespread in Hokkaido. The number of Hokkaido municipalities with WYMV-infested fields increased from 6 in 1994 to 57 in 2010 (Horita et al. 2011). In recent years the damage caused by WYMV has been serious in Tokachi and Ohotsuku districts, which are the main wheat production areas in Hokkaido and are widely planted with ‘Kitahonami’. New wheat cultivars are needed to resist this disease.
WYMV-resistant germplasm, and molecular markers linked to their resistance genes, have been developed. For example, ‘Madsen’ is a WYMV resistant cultivar developed in the USA (Allan et al. 1989). By combining two ‘Madsen’ alleles at the Qym1 and Qym2 QTLs, WYMV was completely controlled in virus nursery fields in Hokkaido (Liu et al. 2016, Suzuki et al. 2015). Although this research is still at the experimental level, the data has encouraged use of a DNA marker-assisted repeated backcrossing approach to breed WYMV resistant cultivars within a short timeframe. Qym1 has been linked to Xwmc041 in chromosome 2DL, and Qym2 has been linked to Xwmc754 in chromosome 3BS (Suzuki et al. 2015).
The initial aim of our research project was to generate a new wheat cultivar that is resistant to both WYMV and eye spot (Pseudocercosparella herpotrichoides). The BC1F4 line 4196-228 was originally selected for this dual purpose. However, due to the urgent need for a new cultivar with WYMV resistance, the focus of the work changed to WYMV resistance alone. The goals of the work reported here include creation of basic maps of Qym1 and Qym2 as single factor genes and development of near isogenic lines (NILs) of ‘Kitahonami’ using molecular markers closely linked to these QTLs, and evaluation of the WYMV resistance, grain yield, agronomic traits, and quality of grain and flour of the NILs, to develop a new line with WYMV resistance in addition to performance and quality equivalent to that of ‘Kitahonami’.
Prior to the work reported here, back crossing had been performed with cross number 4196, using the recurrent parent ‘Kitami-81’ (subsequently named ‘Kitahonami’) and the WYMV resistance donor cultivar ‘Madsen’. Cross 4196 (BC1F2) was harvested without selection in 2007. A total of 690 plants in the BC1F3 generation were evaluated for stem length and maturity date, and 141 plants were selected by their spike and seed appearance to create breeding lines. The population was also advanced by selection of agronomic traits in the BC1F2 and BC1F3 generations. Ninety-two BC1F4 lines were genotyped using markers at wmc041, wmc754 (Supplemental Table 1; Suzuki et al. 2015), and XustSSR2001-7DL (Groenewald et al. 2003). Line 4196-228, with ‘Madsen’ genotype at three markers was crossed with ‘Kitahonami’ in 2009, repeatedly backcrossed, and NILs were selected (Fig. 1).
Breeding of ‘Kitahonami’ near-isogenic lines (NILs). NILs are shown in boxes.
Eight genetic markers linked to the resistance genes (Supplemental Table 1) were used to select plants. The marker, ym115, was converted from an AFLP fragment according to the methods of Suzuki et al. (2013), using resistant and susceptible recombinant inbred lines from the cross between the cultivar ‘Hokushin’ (WYMV susceptible; Yanagisawa et al. 2000) and ‘Madsen’. Analysis of ym115 and cfp059 was performed using Hotstar taq (QIAGEN, Hilden, Germany) and annealing temperature was 55°C. SSR markers were identified using Taq Gold DNA polymerase (Applied Biosystems, Foster City, CA, USA) and annealing temperature was 56°C. PCR products were analyzed with an ABI Prism 3500 Genetic Analyzer (Applied Biosystems) with GeneMapper software, as described by Suzuki et al. (2015) except that 2% agarose gel was used for ym115 and cfp059. Individual plants that were heterozygous at all eight markers were selected in the BCnF1 generations, and individuals homozygous at four markers near the peak of the QTL were selected in the BCnF2 generations, to fix a QTL genotype. Selection in BCnF3 was based on agronomic trait scores being similar to those of ‘Kitahonami’.
WYMV disease severity and infection rateDisease severity was evaluated visually in the Date nursery and rated on a 0 to 4 scale (not infected to severely infected, respectively), as described by Takeuchi et al. (2010). Double-antibody sandwich enzyme-linked immunosorbent assay (ELISA; Clark 1981) was performed using polyclonal antibodies to WYMV (Ueda et al. 1998). Bulked samples of five plants per line and one leaf segment per plant were analyzed. Lines that were visually scored 0 and were WYMV negative in the ELISA test were classed as resistant; all other lines were classed as susceptible.
The WYMV resistance of ‘Kitami-94’ was further evaluated in WYMV-infested fields located in Date, Chitose, Sarabetsu, Tanno, and Kitami, Japan. Samples comprised of one leaf segment from each of 30 individual plants in a plot were collected in April of each year, and were analyzed by ELISA, as above.
Agronomic traitsAgronomic performance was assessed by growing the breeding lines in multiple field locations in Japan, i.e., at the Chuo Agricultural Experiment Station (AES) in Naganuma, at Kamikawa AES in Pippu, at Tokachi AES in Memuro, at Kitami AES in Kunneppu, and in the WYMV-infested fields. All entries were sown in the middle of September. The plot design was randomized block with two replications, but ‘Kitami-94’ was tested in four replications at three AES and in six replications at Kitami AES. Cultivation details for the experimental field plots are listed in Supplemental Table 2. Agronomic traits including heading date, maturity date, stem length, number of spikes, grain yield, thousand grain weight, and test weight were assessed, either in the field or in the lab after harvest. Protein content of the harvested grain was measured using a near-infrared reflectance instrument (Infratec 1241, Foss, Denmark).
Quality of flour and ‘udon’ noodlesGrains were milled using a Buhler test mill after conditioning the moisture level to 14.5% (w/w) overnight. Straight flour yield was calculated as the total recovered products. Flours were adjusted to 60% extraction in all subsequent tests. The ash content was determined using the rapid magnesium acetate method (American Association of Cereal Chemists 2000). The pasting color of the flours (6 g flour + 10 ml distilled water) was evaluated with a colorimeter (ZE 6000, Nippon Denshoku Industries, Tokyo, Japan). Amylose content was measured by starch-iodine reaction. Briefly, the absorbance at 600 nm of wheat flour samples (100 mg) was measured in an automated flow analyzer (AutoAnalyzer II, Bran Luebbe, Hamburg, Germany), using the method described by Juliano (1971). Udon (Japanese white salted noodles) were prepared and evaluated according to the method described by Toyokawa et al. (1989a, 1989b). Boiled noodles were scored by five trained panelists. The maximum score of 25 was given for color, 20 for surface appearance, 10 for appropriate firmness, 25 for elasticity, 10 for smoothness in the mouth, and 10 for a natural flavor.
Polyphenol oxidase activity and genotyping of PPO-D1Polyphenol oxidase (PPO) activity of the grain and the flour was measured using the dihydroxy-L-phenylalanine (L-DOPA) method of Anderson and Morris (2001), for grain, the L-DOPA method modified by Ito et al. (2008) for flour. Five grains were incubated with shaking at 20°C for 1 hour in 1.5 ml L-DOPA solution (10 mM L-DOPA, 50 mM 3-morpholinopropanesulfonic acid, pH 6.5) in a tube and PPO activities were determined by measuring absorbance of the supernatant at 475 nm with a spectrophotometer (U-2900, HITACHI Ltd., Tokyo, Japan). For flour, 0.2 g of flour and 4 ml of L-DOPA solution were placed in a flat-bottom test tube and shaken for 1 hour at room temperature, incubated at 20°C for 18 hours, stirred well, and measured with the colorimeter (ZE 6000, Nippon Denshoku Industries, Tokyo, Japan). For the genotyping of PPO-D1, we used PPO29 marker described by He et al. (2007). Analysis of PPO29 was performed using Hotstar taq (QIAGEN, Hilden, Germany) and annealing temperature was 55°C.
Mapping of Qym1 and Qym2QTL analysis has located Qym1 in an interval between Xgwm539 and Xgwm349 on the long arm of 2D, and Qym2 in an interval between Xbarc147 and Xwmc623 on the short arm of chromosome 3B (Suzuki et al. 2015). In the current study, populations of F2:F3 progeny and four types of ‘Kitahonami’ and ‘Hokushin’ NILs were developed, in which Qym1 and Qym2 were homozygous for either the ‘Madsen’ or the recurrent parent genotype.
The Qym1 locus was mapped using a segregating population derived from a ‘Kitahonami’ NIL. Thirty-six BC6F2 seeds derived from ‘Kitahonami’/4832F1-19 (Fig. 1) were sown in 2012, and three plants that were heterozygous at Qym1 and homozygous for the ‘Madsen’ genotype at Qym2 were selected to make a mapping population of Qym1 using DNA markers (Supplemental Table 1). These plants, equivalent to the heterozygous F1 generation, were self-pollinated and F2 equivalent seeds were harvested in March 2014. Immediately 200 seeds were sown in the greenhouse to obtain F3 equivalent lines in August 2014. A total of 102 F3 equivalent lines that produced more than 60 seeds were sown in the WYMV-infested nursery in two replicates of 30 seeds each, and the WYMV resistance of the plants was evaluated in 2015. These 102 lines were genotyped by DNA markers flanking Qym1.
The Qym2 locus was mapped using a segregating population derived from a ‘Hokushin’ NIL. Eighty BC6F2 seeds derived from a cross between Takikeimugi-3 (BC5 line generated by using ‘Hokushin’ as recurrent parent and ‘Madsen’ as resistance donor parent; Takeuchi et al. 2010) and ‘Hokushin’ were sown in 2010, and one plant that was homozygous at Qym1 and heterozygous at Qym2 was selected as the mapping population of Qym2 using the markers described by Suzuki et al. (2015). This plant, equivalent to the heterozygous F1 generation, was self-pollinated and 262 F2 equivalent seeds were sown in 2012. One spike of each individual was bagged at heading date to prevent outcrossing, and the bagged spikes were harvested in 2013. A total of 104 F3 equivalent lines that yielded more than 40 seeds each were sown in two replicates of 20 seeds in the WYMV nursery, and the virus resistance of the plants was evaluated in 2014. These F3 lines were genotyped by DNA markers around Qym2. The polymorphic markers used for mapping were compared to web page data (https://wheat.pw.usda.Gov/ggpages/SSRclub/GeneticPhysical/) and to cfp059 (Paux et al. 2008).
Three replications of the ‘Kitahonami’- and ‘Hokushin’-NILs were sown in 2013 and the virus infection rate of 20 individual plants was scored in 2014. Basic linkage maps were constructed using MAPMAKER/Exp v3.0b (Lander et al. 1987). Recombination frequencies were converted into map distances using Kosambi’s mapping function (Kosambi 1943).
In the BC1F4 generation only a single line, 4196-228, was homozygous for the ‘Madsen’ genotype at wmc041 (linked to Qym1) and wmc754 (linked to Qym2) and heterozygous at XustSSR2001-7D. Among the 35 NILs that were grown in the WYMV nursery in 2010 and their WYMV infection rates scored, 4196-228 was judged to be resistant to infection by the virus. This line was homozygous for the ‘Madsen’ genotype at markers wmc601 through gwm349 flanking the Qym1 locus, and gwm389 (located between barc147 and gwm493) through wmc623 flanking the Qym2 locus (marker locations as reported by Suzuki et al. 2015). Therefore line 4196-228 was selected for further backcrossing (Fig. 1).
Agronomic traits and quality of NILsSix NILs were evaluated for agronomic traits and quality from 2015 to 2017. The stem length in all NILs was 2 to 4-cm shorter (P < 0.05) than that of ‘Kitahonami’ grown in the same year, except for KK1935 which was not significantly different from ‘Kitahonami’ (Table 1). Thousand grain weight (TGW) of KK1947, KK1960, and KK1963 was 0.9, 3 and 2.4 g lighter, respectively, (P < 0.05) than that of ‘Kitahonami’. Heading date, maturity date, spike length, number of spikes, grain yield, test weight, and protein content of seeds of these NILs were not significantly different from the values for ‘Kitahonami’ (Table 1). The flour quality of these lines were very similar to each other (Supplemental Table 3). Line KK1947 had agronomic trait values and quality very similar to ‘Kitahonami’ and its TGW was heavier than that of KK1948 (Table 1). It was named ‘Kitami-94’, and was tested for certification as a recommended cultivar for Hokkaido. The new promising line ‘Kitami-94’ was studied extensively in the subsequent phases of the research reported here.
| Year | Name | WYMV resistance (in Date) | Genreration | Heading date | Maturity date | Stem length (cm) | Spike length (cm) | Number of spikes (per m2) | Grain yield (kg ha–1) | 1000-grain weight (g) | Test weight (g) | Protein content of seeds (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2015 | KK1934 | Resistant | BC2F6 | 6.01 | 7.22 | 81** | 8.6 | 724 | 9374 | 41.3 | 828 | 9.3 |
| KK1935 | Resistant | BC2F6 | 6.01 | 7.22 | 86 | 8.5 | 759 | 9591 | 41.9 | 833 | 9.3 | |
| Kitahonami | Susceptible | 6.01 | 7.22 | 85 | 8.7 | 627 | 9253 | 42.3 | 848 | 9.3 | ||
| 2016 | KK1947 | Resistant | BC6F4 | 6.01 | 7.25 | 84* | 8.8 | 815 | 7494 | 40.7* | 819 | 10.9 |
| KK1948 | Resistant | BC6F4 | 6.02 | 7.25 | 83* | 8.9 | 732 | 7518 | 39.8 | 823 | 11.2 | |
| Kitahonami | Susceptible | 6.01 | 7.25 | 86 | 8.6 | 811 | 7255 | 41.6 | 828 | 10.8 | ||
| 2017 | KK1960 | Resistant | BC6F5 | 6.05 | 7.20 | 81* | 8.8 | 582 | 7478 | 37.9* | 832 | 10.3 |
| KK1963 | Resistant | BC6F5 | 6.05 | 7.20 | 81* | 8.8 | 603 | 7415 | 38.5* | 834 | 10.4 | |
| Kitahonami | Susceptible | 6.04 | 7.20 | 84 | 8.8 | 583 | 7248 | 40.9 | 845 | 10.3 | ||
| Mean | NILs | 6.02 | 7.22 | 83 | 8.7 | 703 | 8145 | 40.0 | 828.3 | 10.2 | ||
| Kitahonami | 6.02 | 7.22 | 85 | 8.7 | 674 | 7918 | 41.6 | 840.3 | 10.1 |
*,** Significantly different from ‘Kitahonami’ at the 5% and 1% levels, respectively (Student’s t-test).
Mean stem length of ‘Kitami-94’ was 3- and 2-cm shorter than that of ‘Kitahonami’ (P < 0.05) at the Tokachi AES and Kitami AES, respectively (Table 2). The TGW of ‘Kitami-94’ was lighter (P < 0.05) than that of ‘Kitahonami’ at all AES locations, and the test weight of ‘Kitami-94’ was lighter (P < 0.01) than that of ‘Kitahonami’ at three of the four AES locations. Mean seed protein content in ‘Kitami-94’ was 2 points lower (P < 0.05) than that of ‘Kitahonami’ at the Kitami AES only (Table 2). There were no significant differences between ‘Kitami-94’ and ‘Kitahonami’ in the other agronomic traits.
| Test field | Line and cultivar name | Heading date | Maturity date | Stem length (cm) | Spike length (cm) | Number of spikes (per m2) | Grain yield (kg ha–1) | 1000-grain weight (g) | Test weight (g) | Seeds protein (%) | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Chuo AES | Kitami-94 | 6.01 | 7.18 | 85 | 9.2 | 806 | 8336 | 36.6 |  ** |
823 | ** |
9.4 | ||
| Kitahonami | 6.01 | 7.17 | 86 | 9.2 | 793 | 8183 | 38.8 | 830 | 9.6 | |||||
| Kamikawa AES | Kitami-94 | 6.04 | 7.18 | 74 | 8.6 | 458 | 5545 | 39.3 | * |
829 | ** |
9.5 | ||
| Kitahonami | 6.04 | 7.18 | 74 | 8.7 | 460 | 5568 | 41.3 | 834 | 9.5 | |||||
| Tokachi AES | Kitami-94 | 6.01 | 7.23 | 78 | ** |
8.9 | 713 | 6435 | 40.0 | ** |
849 | ** |
11.1 | |
| Kitahonami | 6.02 | 7.23 | 81 | 8.9 | 665 | 6484 | 42.2 | 859 | 11.2 | |||||
| Kitami AES | Kitami-94 | 6.06 | 7.26 | 78 | ** |
8.6 | 717 | 8866 | 40.4 | * |
839 | 9.8 | * |
|
| Kitahonami | 6.05 | 7.26 | 80 | 8.6 | 666 | 8691 | 42.3 | 844 | 10.0 | |||||
| Mean of 4 AES | Kitami-94 | 6.02 | 7.21 | 79 | 8.8 | 674 | 7295 | 39.1 | 835 | 9.9 | ||||
| Kitahonami | 6.03 | 7.21 | 80 | 8.9 | 646 | 7232 | 41.1 | 842 | 10.1 | |||||
Data are means of the results from 2017–2020.
*,** Significantly different at the 5% and 1% levels, respectively (Student’s t-test).
There were no significant differences between ‘Kitami-94’ and ‘Kitahonami’ in flour yield, grain ash content, flour protein content, flour color, or amylose content (Table 3). PPO activity was higher in ‘Kitami-94’ than in ‘Kitahonami’ in both grain (P < 0.05) and flour (P < 0.01) (Table 4). The PPO-D1 genotype in ‘Kitami-94’ was the active type (PPO-D1b), which was the same as in ‘Madsen’. Although more gray spots were observed on the uncooked noodle sheet in ‘Kitami-94’ than in ‘Kitahonami’ (data not presented), the cooked noodle scores were not significantly different from each other (Table 3).
| Year | Name | Flour yield (%) | Grain ash (%) | Flour protein (%) | Pasting color of the flour | Amylose content (%) | Noodle scoring | ||
|---|---|---|---|---|---|---|---|---|---|
| L* | a* | b* | |||||||
| 2017 | Kitami-94 | 73.6 | 1.25 | 8.7 | 87.92 | –0.35 | 15.79 | 22.4 | 69.9 |
| Kitahonami | 74.8 | 1.24 | 8.7 | 87.70 | –0.28 | 15.65 | 21.9 | 70.0 | |
| 2018 | Kitami-94 | 74.6 | 1.27 | 9.5 | 87.37 | –0.04 | 16.89 | 21.8 | 69.8 |
| Kitahonami | 73.8 | 1.25 | 9.7 | 87.79 | –0.19 | 16.46 | 21.9 | 70.0 | |
| 2019 | Kitami-94 | 73.2 | 1.23 | 8.4 | 87.67 | –0.07 | 16.89 | 20.5 | 70.2 |
| Kitahonami | 73.2 | 1.27 | 8.8 | 87.62 | –0.01 | 16.69 | 20.7 | 70.0 | |
| 2020 | Kitami-94 | 73.1 | 1.26 | 8.3 | 87.75 | –0.15 | 16.63 | 20.9 | 69.9 |
| Kitahonami | 72.4 | 1.22 | 8.4 | 87.89 | –0.27 | 16.63 | 21.3 | 70.0 | |
| Mean | Kitami-94 | 73.8 | 1.25 | 8.7 | 87.68 | –0.15 | 16.55 | 21.4 | 70.0 |
| Kitahonami | 73.9 | 1.25 | 8.9 | 87.75 | –0.19 | 16.36 | 21.5 | 70.0 | |
| Sample | Flour ash content | Flour protein content | Grain PPOa (475 nm) | Flour PPOb (L*) | Genotype by PPO29 | ||
|---|---|---|---|---|---|---|---|
| Kitahonami | 0.37 | 9.7 | 0.229 | * |
59.26 | ** |
PPO-D1a |
| Kitami-94 | 0.40 | 9.5 | 0.325 | 56.18 | PPO-D1b | ||
| Madsenc | – | – | – | – | PPO-D1b | ||
Flour properties and PPO data are in 2017.
*,** Significantly different from ‘Kitahonami’ at the 5% and 1% levels, respectively (Student’s t-test).
a Grain PPO activity was determined by measuring absorbance of the supernatant at 475 nm with a spectrophotometer.
b Flour PPO was measured by colorimeter using the dihydroxy-L-phenylalanine method.
c For Madsen, only PPO-D1 allele was tested.
The WYMV infection rates in ‘Kitahonami’ were much higher than in ‘Kitami-94’ in all WYMV-infested fields (Table 5). The heading dates and maturity dates of ‘Kitahonami’ were almost always 3 days later than those of ‘Kitami-94’. The mean stem length of ‘Kitahonami’ was shorter than that of ‘Kitami-94’ in four of the five fields (Table 5). Although these differences were not significant, the trend likely reflects the effects of the WYMV virus infection.
| Test field | Line and cultiver name | WYMV infection % | Heading date | Maturity date | Stem length (cm) | Spike length (cm) | Number of spikes (per m2) | Grain yield (kg ha–1) | 1000-grain weight (g) | Test weight (g) | Seeds protein (%) | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Date | Kitami-94 | 0.9 | 5/29 | 7/19 | 78 | 9.2 | 839 | 7880 | 37.5 | 802 | 9.9 | ||
| Kitahonami | 99.2 | 6/1 | 7/20 | 73 | 8.4 | 694 | 6320 | 36.3 | 804 | 9.7 | |||
| Chitose | Kitami-94 | 0.0 | – | – | 79 | 9.1 | 745 | 8010 | 37.7 | 812 | 10.6 | ||
| Kitahonami | 34.5 | – | – | 80 | 9.5 | 714 | 5385 | 37.4 | 802 | 10.9 | |||
| Sarabetsu | Kitami-94 | 3.1 | 6/3 | 7/25 | 74 | 8.4 | 750 | 6820 | 39.5 | 855 | 11.7 | ||
| Kitahonami | 65.0 | 6/7 | 7/28 | 73 | 8.3 | 723 | 4810 | 35.7 | 846 | 12.4 | |||
| Tanno | Kitami-94 | 5.9 | 6/4 | 7/23 | 74 | 8.8 | 680 | 8525 | 39.1 | 826 | 10.4 | ||
| Kitahonami | 90.0 | 6/8 | 7/26 | 68 | 8.4 | 673 | 7185 | 38.7 | 819 | 10.9 | |||
| Kitami | Kitami-94 | 1.7 | 5/31 | 7/21 | 74 | 8.9 | 718 | 8330 | 39.0 | 825 | 10.8 | ||
| Kitahonami | 93.2 | 6/4 | 7/25 | 72 | 8.7 | 683 | 7195 | 39.8 | 834 | 11.8 | |||
| Mean of 5 area | Kitami-94 | 2.3 | 6/2 | 7/22 | 76 | 8.8 | 746 | 7913 | ** |
38.5 | 824 | 10.7 | * |
| Kitahonami | 76.4 | 6/5 | 7/25 | 73 | 8.6 | 697 | 6179 | 37.6 | 821 | 11.1 | |||
Data are means of results from 2018–2019 for the Date, Sarabetsu, Tanno, and Kitami fields, and from 2019–2020 for Chitose.
*,** Significantly different at the 5% and 1% levels, respectively (Student’s t-test).
Test weight and TGW of ‘Kitami-94’ were not significantly different from those of ‘Kitahonami’ (Table 5). Importantly, however, the mean grain yield of ‘Kitahonami’ was significantly lower (P < 0.01) than that of ‘Kitami-94’ when both were grown in WYMV-infested fields, with differences ranging from 14% to 33% (Table 5). Also, the mean protein content of ‘Kitahonami’ was 0.4 points higher (P < 0.05) than that of ‘Kitami-94’, likely due to the reduced yields.
Effect of single resistance genes on WYMV infectionAll tested NILs, including ‘Kitami-94’, that inherited both Qym1 and Qym2 QTLs from ‘Madsen’ had almost complete resistance to WYMV (Tables 1, 5). With regard to the individual effect of each QTL, the WYMV infection rates were very similar regardless of the broader genetic background of the NIL. The infection rates of NILs with ‘Madsen’ alleles at both the Qym1 and Qym2 loci (Qym1.m and Qym2.m, respectively) were 0% (Table 6). Infection rates of NILs with only Qym1.m were ca. 45%, whereas in NILs with only Qym2.m the infection rates were ca. 30%. In NILs with no ‘Madsen’ alleles the infection rates were ca. 100% (Table 6).
| a) ‘Hokushin’ | b) ‘Kitahonami’ | |||||
|---|---|---|---|---|---|---|
| Locusa | WYMV infection % | Locusb | WYMV infection % | |||
| Qym1 | Qym2 | Qym1 | Qym2 | |||
| Madsen | Madsen | 0.0 | Madsen | Madsen | 0.0 | |
| Madsen | Hokushin | 46.7 | Madsen | Kitahonami | 45.0 | |
| Hokushin | Madsen | 30.0 | Kitahonami | Madsen | 28.3 | |
| Hokushin | Hokushin | 98.3 | Kitahonami | Kitahonami | 100.0 | |
a Homozygous for wmc041 and ym115 in Qym1 and gwm493, wmc754, cfd79 and gpw3248 in Qym2 region.
b Homozygous for wmc041, ym115 and gwm349 in Qym1 and gwm389, gwm493, wmc754, cfp1844 and wmc623 in Qym2 region.
In the population of F2:F3 progeny (produced by selfing F2 equivalent plants), developed from the cross between ‘Kitahonami’ and 4832F1-19 (Fig. 1), the plants with more than 35% infection rate were considered to be Qym1.k/Qym1.k (i.e., homozygous ‘Kitahonami’), whereas those with infection rates below 15% were presumed to be either Qym1.m/Qym1.m or Qym1.m/Qym1.k (Fig. 2a). Marker analysis and phenotype analysis of 102 F3 equivalent progeny, indicated that the virus resistance was inherited in a monogenic manner, and that Qym1 mapped at 2.2 cM distal to Xgpw5244 (Fig. 2c).
Frequency distributions of WYMV infection rate, and basic maps of Qym1 and Qym2. a) F2:3 population segregating for Qym1 while fixed for homozygous ‘Madsen’ genotype at Qym2; b) F2:3 population segregating for Qym2 while fixed for homozygous ‘Madsen’ genotype at Qym1; c) linkage map of Qym1; d) linkage map of Qym2. Dark-shaded bars represent families judged to be resistant to WYMV; pale-shaded bars represent families with intermediate infection rates; unshaded bars represent families judged to be susceptible to WYMV. Families with intermediate infection rates were excluded from the gene mapping.
In the population of F2:F3 progeny developed from the cross between ‘Takikeimugi-3’ and ‘Hokushin’, the plants with more than 70% infection rate were considered to be homozygous ‘Hokushin’ (Qym2.h/Qym2.h), and those with less than 35% infection rate were regarded as either homozygous Qym2.m/Qym2.m or heterozygous Qym2.m/Qym2.h (Fig. 2b). Marker analysis and phenotype analysis of 104 F3 equivalent progeny, indicated that the virus resistance was again inherited in a monogenic manner, and that Qym2 was located in an interval of 3.6 cM between Xwmc754 and Xcfd79 and mapped to the same position as cfp059 (Fig. 2d).
‘Kitami-94’ was bred from the original line 4196-228 over a period of four years. Basic mapping indicated that 4196-228 and all the resulting NILs reliably carried the two resistance genes, at Qym1 and Qym2. This study confirmed single resistance effects of both Qym1.m and Qym2.m in the advanced backcross lines. It also found that Qym2.m reduced the WYMV infection rate more strongly than did Qym1.m.
‘Kitami-94’ developed in this study was highly resistant to WYMV but was otherwise similar to ‘Kitahonami’ in agronomic traits. The yield reduction observed in ‘Kitahonami’ grown in the WYMV-infested field was not observed in ‘Kitami-94’. The heading and maturity dates of WYMV-infested ‘Kitahonami’ were both later than those of ‘Kitami-94’. In the general production fields where ‘Kitahonami’ is unevenly infected by WYMV, the disease not only reduces yield but also make harvesting problematic due to the uneven maturity. In the non-infested fields, ‘Kitami-94’ had the same yield as ‘Kitahonami’, but had slightly higher spike number and smaller grain size (lower TGW) than ‘Kitahonami’. A gene increasing the spike number may have been responsible for the smaller grain size. For example, Suzuki et al. (2012) found that cultivar ‘Sumai3’ had a Fusarium head blight resistance QTL near Xgwm539 on 2DL, and that this QTL was significantly correlated with decreased TGW. Interestingly, if the gene responsible for grain size on 2DL in ‘Madsen’ is identical to the ‘Sumai3’ gene reported by Suzuki et al. (2012), this identification of a gene responsible for grain weight near Qym1 may facilitate wheat breeding for resistance to other diseases.
Although ‘Kitahonami’ is a high yielding cultivar, it sometimes produces narrow grains (Sakuma et al. 2019), particularly under unfavorable weather conditions. The TGW of ‘Kitami-94’ was typically lower than that of ‘Kitahonami’, which may be attributable to a grain size gene linked to Qym1. This characteristic might cause a problem that more thin grains are produced under unfavorable weather conditions.
The WYMV infection rates of ‘Kitahonami’ in the Chitose and Sarabetsu WYMV-infested fields were lower than was observed in the other WYMV-infested fields, however the yield reduction of ‘Kitahonami’ grown in these two fields was actually more severe than observed in the other WYMV-infested fields. The reason for this is not clear. Other yield-reducing diseases may have been present in these two fields, the yields of ‘Kitami-94’ grown there did not appear to be reduced. Perhaps there is another resistance gene near Qym1 or Qym2 that prevented yield loss in those fields, but considerable additional study would be needed to investigate this possibility.
The grain quality of ‘Kitami-94’ was very similar to that of ‘Kitahonami’, with the exception of PPO activity. ‘Kitami-94’ had the same active type of PPO-D1b as ‘Madsen’. ‘Kitami-94’, with highly active PPO, had more gray spots on the uncooked noodle sheet than ‘Kitahonami’. However, PPO activity of this sort is not a problem in making Japanese ‘udon’ noodles, because the noodles are boiled immediately after preparation. Thus the boiled ‘udon’ noodles of ‘Kitahonami’ and ‘Kitami-94’ were not visibly different. Importantly, however, wheat flour is also used to make various other products such as Chinese noodles and sweets and so on, gray spots might reduce the value of the product for example ramen and dumpling skins. The NILs and ‘Kitami-94’ had slightly shorter stem length than those of ‘Kitahonami’. Genes with a strong effect on plant height, such as Rht-B1 on 4B and Rht-D1on 4D (Ellis et al. 2002, Gale and Law 1977), Rht8 on 2DS (Korzun et al. 1998) and many others with minor effects have been reported. ‘Madsen’ may have a genetic factor, perhaps on 2DL or 3BS, that slightly shortens stem length. However, the effect on the NILs in this study was small, and did not appear to have a strong impact on other agronomic traits.
In their study of WYMV resistant NILs derived from ‘Madsen’, Kasuya et al. (2017) found that the lines in which the 2DL genotype is ‘Madsen’ have relatively lighter TGW and grain test weight. Zhang et al. (2015) reported that the QTLs related to TGW, grain width, and ratio of grain length to grain width are located near Xwmc041 in a mutant generated by exposure to ethyl methanesulfonate. Based on this information, lines with TGW similar to that of ‘Kitahonami’ were selected in the work reported here, to break the linkage between resistance genes and small grain size. Despite this effort, the TGW of all of the selected lines in the current study was less than that of ‘Kitahonami’. This implies that it is difficult to break the linkage between WYMV resistance and TGW by normal selection, and it may be necessary to search for recombinant individuals using a much larger population. However, if the resistance gene can be isolated using the basic map of Qym1 developed in this study, it may be possible to efficiently produce lines that eliminate the linkage with undesirable traits mentioned above.
It is likely that the ‘Kitami-94’ line will be useful for basic research to elucidate the mechanisms of PPO activity and resistance to WYMV. Kobayashi et al. (2020) produced lines with broken linkage between Ppo-D1b and WYMV resistance, using 960 plants from recurrent back cross progenies. With the help of this report we have begun to break the linkage of undesirable traits with Qym1 on 2DL in ‘Kitami-94’. Future development of new lines without linkage drag could become a powerful breeding tool for pyramiding resistance genes. Recently we started wheat breeding combining Qym1 and Qym2 with a new WYMV resistance gene at Qym4 derived from OW104 (Yamashita et al. 2020) in order to breed varieties with stronger resistance to WYMV. These approaches may help breeders respond to the ongoing changes in viral strains and climate.
The WYMV response alleles at Qym1 and Qym2 present in cvs ‘Madsen’ and ‘Hokushin’ are denoted Ym1/Ym2 and ym1/ym2, respectively, according to recommendation rule for gene symbolization in wheat (McIntosh et al. 2013).
YY, HJ, SO, TS, and TS produced experimental materials. TS, MK, CS, TI, MK, AS, and TS performed the experiments. TS and TK designed the experiments to evaluate resistance genes and basic maps. TS and TK analyzed the data and wrote the manuscript.
This work was supported by a grant from MEXT KAKENHI (22780008) and the Ministry of Agriculture, Forestry, and Fisheries of Japan (Genomics for Agricultural Innovation TRG1003, Genomics-based Technology for Agricultural Improvement, TRS1003, 26097C) and Hokkaido Agricultural Cooperative Association.
