Abstract
Flowering time is an important agronomic trait for Brassica rapa crops, and previous breeding work in Brassica has successfully transmitted other important agronomic traits from donor species. However, there has been no previous attempts to produce hybrids replacing the original Brassica FLC alleles with alien FLC alleles. In this paper, we introduce the creation of a chromosome substitution line (CSSL) containing a homozygous introgression of Flowering Locus C from Brassica oleracea (BoFLC2) into a B. rapa genomic background, and characterize the CSSL line with respect to the parental cultivars. The preferential transmission of alien chromosome inheritance and the pattern of transmission observed during the production of the CSSLs are also discussed.
Introduction
The Brassica rapa and B. oleracea species each contain several agronomically important crops. Varieties of B. rapa includes turnip, mizuna, napa cabbage, bok choy, and leafy vegetables grown as both food and feed crops; and B. oleracea contains many common food and feed crops, such as cabbage, broccoli, cauliflower, kale, savoy, and kohlrabi. B. rapa, B. nigra, and B. oleracea are diploid organisms capable of interspecific hybridization. This relationship was first identified by Nagaharu U in 1935 and is known as the triangle of U (U 1935). The respective genomes are referred to as the A, B, and C genomes. The three species are closely related, sharing a common ancestor that experienced a whole genome triplication (WGT) event approximately 15.9 million years ago (MYA) after its divergence from Arabidopsis thaliana (Cheng et al. 2014). The speciation events that led to the A, B, and C genomes is estimated to have occurred approximately 4.6 MYA, identified by interspecific genome comparisons (Liu et al. 2014). The B. rapa (AA) and B. oleracea (CC) species serve as the parental genomes for the agronomically important allopolyploid Brassica napus (AACC); a seed oil-crop species that produces rapeseed and canola. Thus, B. napus is an amphidiploid species that is derived from the interspecific hybridization of B. rapa and B. oleracea.
Both B. rapa and B. oleracea possess a broad range of vernalization requirements amongst their respective varieties, ranging from near-obligate to a complete lack of the vernalization requirement. However, both species contain biennial varieties possessing near-obligate vernalization requirements, meaning exposure to a prolonged cold period is necessary for the transition from the vegetative growth state to inflorescence (Friend 1985). Vernalization maximizes reproductive success by preventing premature flowering prior to the winter season (Andrés and Coupland 2012). B. rapa varieties with a vernalization requirement generally require four to eight weeks of low temperatures (5°C) to flower (Kim et al. 2007). B. oleracea generally requires a longer cold period of six or more weeks of low temperatures (5°C) in order to flower (Friend 1985). In contrast to B. rapa, the onset of cold must take place after the plant has reached the mature vegetative growth stage of development for successful vernalization to occur in B. oleracea (Ito et al. 1966).
Vernalization in both species is controlled by the MADS domain protein FLOWERING LOCUS C (FLC), first identified in the model organism for plant genetic research Arabidopsis thaliana (Michaels and Amasino 1999). The expression of the floral integrators FLOWERING LOCUS T (FT) and SOC1 are both regulated by FLC in a dose-dependent manner (Lee et al. 2000, Onouchi et al. 2000). Thus, FLC functions as the central suppressor of flowering. Subsequently, QTL studies conducted on flowering time in B. napus, B. rapa, and B. oleracea have identified various paralogs of FLC as strong contributors (Axelsson et al. 2001, Kakizaki et al. 2011, Okazaki et al. 2007, Razi et al. 2008, Schranz et al. 2002, Tadege et al. 2001, Zhao et al. 2010). Further examination of the genetic sequences of FLC alleles has revealed that allelic differences in the non-coding region of intron one are responsible for differences in flowering times between varieties in all three species (Kitamoto et al. 2013, Lin et al. 2005, Razi et al. 2008, Schiessl et al. 2014, Wu et al. 2012, Yuan et al. 2009). Other studies have identified the number of the functional copies of FLC paralogs (Golicz et al. 2016). While geographic climatic variation may partially account for the differences seen in flowering times due to the differing length and severity of the winter season at different latitudes (Irwin et al. 2016), other studies attribute the FLC sequence variations in Brassica species to the selective pressures exerted by agricultural breeding (Golicz et al. 2016, Martynov and Khavkin 2005).
In B. rapa cultivars, stable year-round production is compulsory because of their importance as a vegetable. To accomplish this, the development of late-bolting cultivars with a winter cropping pattern is required. Because FLC acts as a key regulatory gene in the vernalization pathway of both B. rapa and B. oleracea, and allelic differences in the FLC genes present in Brassica affects flowering time, we reasoned that selection for a cabbage-like strong vernalization requirement would be of agricultural benefit to B. rapa cultivars such as Chinese cabbage. We thus thought it useful to attempt to assign species-specific vernalization-response phenotypes to chromosomes using chromosome segment substitution lines (CSSL) derived from the interspecific crossing between both B. rapa and B. oleracea. Previous breeding work in Brassica using monosomic alien addition lines (MAALs) has successfully transmitted important agronomic traits such as disease resistance (Akaba et al. 2009, Chèvre et al. 1997a, Kaneko et al. 1996, Peterka et al. 2004, Tsunoda et al. 1980), and yellow seed color (Heneen et al. 2012) from donor species possessing the respective desired trait. Additionally, when making MAALs and CSSLs derived from different Brassica species, the use of chromosome specific markers identified each chromosome in the A, B, and C genomes was very effective (Chèvre et al. 1997a, 1997b, Heneen et al. 2012, Li et al. 2013). However, the production and application of MAALs and CSSLs has been scarcely reported in Brassica, and to our knowledge there have been no previous attempts reported on the production of MAALs or CSSLs replacing the original Brassica FLC alleles with alien FLC alleles.
The objectives of this study were two-fold: 1) to analyze additional alien chromosome inheritance patterns during the process of the construction of a CSSL of B. rapa possessing an introgression of the FLC paralog BoFLC2, located on the upper-arm of Chromosome 2 in the B. oleracea genome; and 2) to characterize the vernalization-response phenotype and flowering time in a winter cropping pattern in relation to the parental lines. BoFLC2 was selected as our target FLC allele for introgression, because it was shown to possess a strong contribution to the flowering time and reproductive development of cabbage (Okazaki et al. 2007, Ridge et al. 2014).
Materials and Methods
Plant materials and crossing scheme
A commercial cultivar, CR Kanki (Nihon Norin Seed Co. Ltd., Japan) of Chinese cabbage (B. rapa var. pekinensis) and a DH line (P01) of B. oleracea var. capitata cv. Reiho (Ishi seed company, Japan) were used as female and male parents, respectively. Flower buds were emasculated and immediately pollinated with fresh pollen grains collected from the male parent. Pollinated flowers were covered with thin paper bags. The ovaries were collected 4–20 days after pollination (DAP) for F1 embryo rescue. Ovary culture was carried out according to the method reported by Inomata (1977). The surface-sterilized ovaries were placed on MS (Murashige and Skoog 1962) medium aseptically supplied with 3% sucrose and 0.8% agar, adjusted to pH 5.8. Plastic petri dishes 90 × 15 mm were used for the cultures and placed in a growth chamber maintained at 24°C with a 16 h photoperiod. Ovaries were kept on the medium until embryos were fully germinated and rooted. The seedlings were transplanted into 3.5-inch pots containing vermiculite soil:regular soil (1:1) and covered with transparent polyethylene sheet. The plants were then placed in a growth chamber maintained at 22°C with a 16 h photoperiod at light intensity of 300 μmol·m−2·s−1. After proper hardening, the plants were transferred to a greenhouse. Twenty-Five F1 hybrids were obtained from the culture of ovules derived from the crossing between B. rapa cv. CR Kanki and B. oleracea DH line Reiho P01. After adequate vernalization of F1 plants, in order to restore seed fertility by doubling the chromosome number, colchicine solution 0.05% was applied on leaf axils of each F1 plant as per the previous report (Chen et al. 1988). The flowers with pollen in the treated F1 plants were self-pollinated to produce amphidiploids. Simultaneously, in order to produce BC1 plants, the fertile pollen of the F1 hybrids were backcrossed to B. rapa. The embryos obtained from the backcross were excised from the ovule at 20–30 days after crossing and cultured on MS medium. As a result, more than 60 BC1 plants were obtained. Additionally, amphidiploids were used as the pollen parent and backcrossed to CR Kanki to produce BC1F1 plants. Ploidy level of BC1F1 plants was measured by flow cytometry, as described previously (Okazaki et al. 2005).
Then the BC1F1 plants were used as pollen parents and backcrossed to B. rapa resulting in 264 BC2F1 plants, from which 160 plants were genotyped by the C chromosome specific DNA markers and days-to-bolting after vernalization was examined (Fig. 1). To shorten the breeding cycle from the BC1 to BC3 generations, embryos obtained in the backcrossing were excised from the ovules 20–30 days after pollination and cultured on MS medium. The resultant plants were flowered by adequate vernalization (later mentioned). Among the BC2F1 plants, #81 and #174 plants, plants containing the BoFLC2 genomic region were identified by genotyping, selected as female parents, and backcrossed to CR Kanki. In the resulting BC3F1 population, the #174-12 and #174-22 plants were selected from the #174 line and self-pollinated to produce BC3F2 lines. In the #174-12 line of BC3F2, three plants (#174-12-18, -26, and -28) homozygous for the BoFLC2 allele were selected and self-pollinated. The obtained seedlings were characterized for vernalization requirement and genotyped by DNA marker analysis.
Vernalization was given to the plants cultured in the test tube for the BC2F1 and BC3F1, seedlings in Petri dish culture for BC3F2, and seedlings in soil for BC3F3, as follows. When the BC2F1 and BC3F1 seedlings formed roots in ovule culture, they were transferred to MS medium in test tubes, and treated in a growth chamber at 4°C for 8 weeks. After vernalization, they were planted in commercial soil suitable for growth. For the BC3F2 lines (#174-12 and #174-22), the seeds were sown on MS medium containing 1% sucrose and then grown at 4°C for 8 weeks. After vernalization, they were then planted in commercial soil.
For the BC3F3 generation, two lines, #174-12-18 and #174-12-28 were characterized to assess bolting and flowering under three different cropping patterns in field trials conducted in Ogata Ami-machi Inashiki-gun, Ibaraki-ken, Japan (35.9819516 Latitude, 140.2668901 Longitude). Comparison to the parental early-heading B. rapa cv. CR Kanki No. 100 and three late-heading commercial cultivars, B. rapa cv. Haruhinata, B. rapa cv. Kigaku No. 70, and B. rapa cv. Harurisou was tested using five to ten plants from each lineage (n = 5 to 10). For cropping pattern one, seeds were sown in commercial soil in a greenhouse environment on January 14th, 2017 using a hotbed for one month and then the seedlings were grown for 2 weeks without heating for acclimatization to the cold. After that, the seedlings were transplanted outdoors at the end of February and grown under field conditions (Supplemental Fig. 1). For cropping pattern two, seeds were sown in commercial soil in a greenhouse environment on January 14th, 2017, using a hotbed for one month. Then the seedlings were transplanted outdoors in the middle of February and grown under field conditions. For cropping pattern three, seeds were sown in commercial soil in a greenhouse environment on February 1st, 2017 using a hotbed for one month and then seedlings were transplanted outdoors at the beginning of March and grown under field conditions. For all three cropping patterns, plants were grown in the field under cover using vinyl tunnels (Type 9245) and 20 kg/a of nitrogen based fertilizer was applied to the soil. Flowering was assessed by the presence or absence of inflorescence (flower buds or flowers) and bolting stem length was measured, in centimeters, from the crown to the tip of the apical stem. For plants that had already flowered at the time of measurement, a stem bolting length of 100.0 cm was recorded. The flowering and bolting data was collected for cropping patterns one, two, and three on May 5th, 2017; May 16th, 2017; and May 19th, 2017; respectively. The data was then imported into R (Ihaka and Gentleman 1996) and statistical analysis performed for both flowering and bolting stem length in response to cropping pattern and lineage by linear regression, using factorial analysis of variance (ANOVA) with an interaction term for cropping pattern and lineage. For a significant interaction term, post-hoc pairwise comparison of least-square means with Tukey-method correction was used (Lenth 2016).
Genotyping and chromosome observation of the obtained progenies
For collection of genotype data, total genomic DNA was extracted from the young leaves of backcrossed progenies using the CTAB (cetyltrimethyl-ammonium bromide) method (Murray and Thompson 1980). The primer pairs identified in each C genome chromosome were selected from the B. oleracea linkage map, and their map positions were confirmed by mapping to a B. oleracea linkage map (Nagaoka et al. 2010) and by the Bolbase genome database. Markers were amplified by PCR in 5 μl reaction solution mixture (EmeraldAmp Max PCR Master Mix, Takara Bio. Inc., Japan) with the following parameters: 1 cycle of 94°C for 5 min, 35 cycle of 94°C for 1 min, optimized annealing temperature (50 to 56°C) for 30 s, and 72°C for 90 s followed by a final extension 72°C for 7 min. Electrophoresis was conducted using an 8–13% polyacrylamide gel (Kikuchi et al. 1999) for separation of amplified products. The gel was stained with a Gelstar solution (0.1 μl/10 ml; Takara Biomedicals, Japan). The chromosome number of seedlings of BC3F3 was determined from root tips according to the method reported by Karim et al. (2016).
Results
Characterization of BC2F1 population
The random assortment of C genome chromosomes during the meiosis of BC1F1 is thought to cause 50% of the transmission rate for each chromosome in the BC2F1 plant. To assess this theoretical ratio, two DNA loci whose positions are already determined in the linkage map (Nagaoka et al. 2010) were assigned to each C chromosome (Supplemental Fig. 2). Since the A and C genomes share nucleotide sequence similarity, we compared PCR profiles among CR Kanki, Reiho P01, and their hybrids to examine whether our designed DNA markers can distinguish between homologous sequences of the A and C genomes; consequently, 19 DNA markers were chosen (Supplemental Fig. 2, Supplemental Table 1). The screening of 169 BC2F1 plants using the chromosome specific DNA markers revealed that in most cases both loci were simultaneously transmitted, while in some plants a single locus of each chromosome was separately transmitted to a plant (Fig. 2). The transmission rate of C chromosome specific DNA loci ranged from 14.2% to 69.2%, depending on the chromosome. Each C chromosome was transmitted as a whole chromosome from 11.8% to 47.3% of the plants (32.5% on average), as indicated by the horizontal lines between the measurements of each loci on a given chromosome, and as an incomplete chromosome from 0% to 27.2% of the plants (9.24% on average).
The expected number of extra C genome chromosomes transmitted to the BC2F1 population ranges from 0 to 9. To assess chromosome inheritance in the BC2F1 plants, we used two C-chromosome specific DNA loci per chromosome. When at least one of the two C-chromosome specific loci is detectable in a BC2 plant, the plant is regarded as an aneuploid possessing the corresponding C-chromosome segment (Fig. 3). The marker assay revealed that the number of extra C genome fragments/chromosomes of the BC2F1 population ranged from 0 to 9 (4.59 on average), and was skewed towards both the smaller (n = 10–12) and the larger numbers (n = 16–19), as compared to the binominal theoretical distribution (Fig. 3). The number of days to bolting in the BC2F1 plants was observed along with CR Kanki and Reiho P01 after 8 weeks of vernalization, as mentioned in Materials and Methods. CR Kanki flowered up to 29–49 days after 8 weeks of vernalization, but Reiho P01 had no flowering. In the BC2F1 population (n = 169), 28% of the plants bloomed within 60 days after 8 weeks of vernalization and the remaining 72% of the plants did not. The groups containing more C genome fragments/chromosomes tended not to bloom after 8 weeks of vernalization, indicating that C genome elements added to A genome made the plants resistant to vernalization.
To examine the effect of each C genome chromosome on the flowering response of the BC2 plants, the plants were categorized with or without the respective chromosome, regardless of the number of extra C genome chromosomes added in each plant (Fig. 4). As a result, the plants with the C2A chromosome region, detected by the C2A DNA marker, significantly had no flowering after 8 weeks of vernalization, while the plants with the C6 chromosome had a higher flowering rate at the significant level (p < 0.01) as detected by binomial analysis (Fig. 4a). In contrast, the plants without the C2A and C6 chromosome regions revealed the promotion and the delay of flowering after 8 weeks of vernalization, respectively (Fig. 4b). It was previously reported that the BoFLC2 allele that delays flowering in B. oleracea was found on the C2A chromosome region (Okazaki et al. 2007). Based upon these results, thereafter, we focused on C2A chromosome region.
Characterization of BC3F1 population
Among BC2F1 population, #81 and #174 plants, which were confirmed to contain the C2A genomic region by the C2A DNA marker (designated as the C2A chromosome region), were selected as female parents and backcrossed with CR Kanki. #81 had only C2A and C2B chromosome regions, and #174 had C2A and C9A chromosome regions (data not shown).
The screening of 76 plants of the BC3F1 #81 line, using the C2 chromosome specific DNA markers, revealed that the transmission rates were 0.38, 0.43, and 0.45 in BoFLC2, C2A, and C2B markers, respectively (Fig. 5a). The DNA marker analysis indicated that #174 line contained a more limited region of the C2 chromosome (BoFLC2 to C2A marker position), compared to the #81 line (BoFLC2 to C2B marker position). The transmission rate of the #174 line was the same as that of the #81 line. We found a segregation of flowering time in both the #81 and #174 BC3F1 lines after cold treatment at 4–9°C for 8 weeks (Fig. 5b, 5c). The plants with a C2A chromosome region, detected by the BoFLC2 marker, significantly tended to have no flowering after vernalization in both lines. Counterintuitively, some plants not possessing the BoFLC2 region revealed no flowering or delayed flowering after vernalization, while all of the control plants (CR Kanki) flowered within 28 days after vernalization. The #174 line contained a C9A chromosome region, however we did not check the segregation of the C9A chromosome region in this generation, but evidence of its transmission and segregation was present in the #174-12-26 fixed line (as discussed below).
From the BC3F1 population, #81-4, #81-54, #174-12, and #174-22 plants were selected from the #81 and #174 lines and self-pollinated to produce BC3F2 lines. #81-4 was selected from the non-flowered plants having the BrFLC2 homozygotes (lacking BoFLC2) and the other 3 lines from the heterozygote (BrFLC2/BoFLC2) BC2 plants. We found segregation of the BoFLC2 allele, including heterozygotes containing BrFLC2 and BoFLC2 alleles and homozygotes of both the BoFLC2 and BrFLC2 alleles, respectively (Fig. 6b–6d). The plants with a BoFLC2 allele tended to show no flowering after vernalization. All of the control plants (CR Kanki) flowered within 28 days after vernalization. We observed plants homozygous for the BoFLC2 allele in the #174-12 line, but not in the #81-54 and #174-22 lines. Then, several plants (including #174-12-18, -26, and -28) homozygous for the BoFLC2 allele were self-pollinated to develop the BC3F3 lines. We then examined the chromosome number in the root tips of the #174-12-18, #174-12-26, and #174-12-28 seedlings, and confirmed the euploidy (2n = 20; Data not shown).
On the other hand, we have the unexpected results that the #81-4 plants where the BoFLC2 gene was not detected exhibited no flowering or delayed flowering (Fig. 6a), while all the control plants (CR Kanki) flowered within 28 days after vernalization. Therefore, we made a subsequent BC3F2 generation of the #81-4 lacking the BoFLC2 gene (Supplemental Fig. 4). These subsequent plants (#81-4-18 and #81-4-19) revealed a flowering response like that of the B. rapa parent, indicating a failure to maintain tolerance against vernalization in the #81-4 line lacking the BoFLC2 region.
The BC3F3 #174-12-18 and #174-12-28 populations show reduced flowering and bolting in a winter cropping pattern
A field trial was conducted on two BC3F3 populations, #174-12-18 and #174-12-28, homozygous for BoFLC2, along with CR Kanki No. 100 (recurrent parent) and three late-heading varieties of Chinese cabbage. Weather data recorded during the course of the experiment shows that daily mean temperatures were adequate for vernalization to occur (Supplemental Fig. 5). Results of the field trial showed that the C2 introgressed populations responded more like the late-heading varieties of B. rapa than that of the parental B. rapa cv. CR Kanki No. 100 (Supplemental Fig. 6, Supplemental Table 2). ANOVA on flowering in response to lineage and cropping pattern revealed that both main effects were statistically significant with respect to flowering, with p-values p < 2.2·10−16 and p = 1.74·10−9 for cropping pattern and lineage, respectively. The interaction between cropping pattern and lineage (cropping pattern × lineage) was also found to be statistically significant, with a p-value of p = 5.30·10−10. Post-hoc analysis of the least-square means of the interaction term revealed that the parental CR Kanki flowers similarly under all cropping patterns (Supplemental Fig. 7, Supplemental Table 3). However, the #174-12-18 and #174-12-28 lines show a marked difference in inflorescence under cropping pattern one, with respect to the early-heading parent CR Kanki, grouping with the late-heading cultivars Hinata, Kigaku No. 70, and Harurisou (Supplemental Fig. 8).
The ANOVA conducted for stem bolting length in response to cropping pattern and lineage revealed significant effects for both cropping pattern and lineage, with p-values of p = 1.01·10−6 and p < 2.2·10−16, respectively. The mean bolting length for CR Kanki No. 100 was recorded as 100.00 cm under all three cropping patterns, as all the plants had already flowered at the time of measurement (Supplemental Fig. 9). The interaction effect was also significant, with a p-value of 8.23·10−5. The post-hoc least-square means analysis of the interaction term revealed that the two introgressed populations grouped together under cropping pattern one, with a least-squares mean bolting lengths of 2.76 cm and 4.94 cm for #174-12-28 and #174-12-18, respectively (Supplemental Fig. 10, Supplemental Table 4).
Discussion
Chromosome transmission in the BC2 population shows preferential rates of transmission
Several chromosome addition lines, including monosomic alien addition lines (MAALs) and disomic addition lines, have been developed from different cross-combinations in the Brassicaceae (Heneen et al. 2012, Kaneko and Bang 2014). The aims of those studies were to assign species-specific characteristics to particular chromosomes, dissect genome homology by monitoring chromosome pairing/recombination through meiosis, and transfer desirable agronomic traits between species. Backcrossing to recurrent parents via interspecific crossing is the essential approach to produce MAALs.
In this study, we found that the transmission rates of the C chromosomes varied. Leflon et al. (2006) reported 23.3–46.8% for the whole C chromosome transmission and 1.3–13% for the fragmented C chromosome transmission in the cross of AAC × AA (CD × C in their designation). Leflon et al. (2006) also found that N12 (DY2 in their designation, corresponding to C2) and N15 (DY18, corresponding to C5) was frequently transmitted via female gametes, and N18 (DY8, corresponding to C8) was less frequently transmitted. In contrast, in this study, an AAC sesquiploid was crossed as the male parent to B. rapa (AA); consequently, the C5, C6, C7, C8, and C9 chromosomes were less frequently transmitted, and two fragments (C1A and C2B) of the C1 and C2 chromosomes were transmitted to the subsequent progeny with a significantly high frequency. The discrepancy between Leflon et al. (2006) and this study may be due to the difference of the crossing direction and/or the parental materials used in the two studies. It has been reported that in the 4x × 2x crosses of tulips, additional chromosomes tend to be transmitted more frequently by the female gametes than by the male gamete (Mizuochi et al. 2009). Chèvre et al. (1997b) reported a very low male transmission rate of chromosome 3 (1.3%) from the B genome in B. napus-B. nigra MAALs, while the remaining chromosomes (1, 2, 4 and 5) in their respective MAALs were frequently transmitted (18.1–28.8%). Thus, the distorted chromosome transmission rates, especially via male gametes, has been generally reported in several species.
We found that the number of extra C genome chromosomes transmitted to the BC2F1 population ranged from 0 to 9 (Fig. 3). This frequency distribution was skewed towards the tails of the binomial distribution, resulting in both smaller (n = 10–12) and larger (n = 16–19) numbers of transmitted chromosomes. The data shown in Fig. 3 is represented from plants with two DNA markers, and either of the two DNA marker of the respective chromosome. We therefore overestimate the number of extra C chromosomes in BC2F1 plants from the point of view of counting fragmented chromosomes. However, since the transmitted rate of the fragmented chromosomes was not high (Fig. 2), it is likely that the unbalanced gametes (n = 13–15) are less frequently transferred to the BC2F1 population from the sesquidploid (AAC); and it is also likely that other relatively balanced gametes, with lower and higher numbers of C genome chromosomes, are frequently transmitted. This is in agreement with previous studies (Lee and Namai 1992, Leflon et al. 2006, Lu et al. 2001).
Transmission of C2 in the BC3 population indicates a homozygous transmission of BoFLC2
In the subsequent BC3F1 progenies of this study, #81 and #174 BC2F1 plants containing the C2A genomic region were subjected to analysis to examine the patterns of alien chromosome inheritance. The transmission rate of C2 chromosome specific loci was around 40%, indicating the transmission rate was sufficient for development of a BoFLC2 introgression line (Fig. 5). McGrath and Quiros (1990) and Heneen et al. (2012) studied alien chromosome inheritance patterns using the backcrossed or selfed progenies of MAALs derived from the cross of B. rapa (AA) × B. napus (AACC). They reported the following transmission rates of C genome chromosome loci: 30.4% for 6Pdg-1 and 38.0% for Pgm1 (McGrath and Quiros 1990), and 30.3–36.7% for C1–C9 chromosome (Heneen et al. 2012). These transmission rates were like that of the BC3F1 population in our study. In the subsequent BC3F2 progenies developed from the selfing of plants heterozygous for the BoFLC2/BrFLC2 markers, we found a segregation of the BoFLC2 allele. Consequently, we identified seven plants homozygous for the BoFLC2 allele in the #174-12 lines; where BrFLC2 homo-, BoFLC2 homo-, and heterozygotes segregated at a ratio of 7:7:15, fitting the theoretical 1:1:2 ratio (p = 0.98); the BoFLC2 homozygous plants lacked the BrFLC2 locus.
Marker based analysis of A2 in the #174-12-26 population indicates that the BrFLC2 allele was homozygously replaced with BoFLC2 in the A genome background, via homologous recombination of the target chromosomal region, with a region of homozygous introgression derived from C2 approximately 6.5 Mbp in size and corresponding to the top of the A2 chromosome (Shea et al. 2017). The size of the introgression was calculated as the distance between the top of C2 and the locus of the BnGMS239 marker using coordinates with respect to the Bolbase reference genome for B. oleracea var. capitata. It was also estimated by the newly developed algorithm (designated as IntroMap, Shea et al. 2017) by scanning a re-sequenced #174-12-26 genome and identifying introgression regions derived from alien species hybridization. The scan using IntroMap confirmed the #174-12-26 line contains the BoFLC2 region and a small region encompassing no flowering related genes of the C9 chromosome.
Flowering rates in the BC3F1 population were possibly affected by genome shock
In the BC3F1 populations of the #81 and #174 lines, the segregation of non-flowering and flowering plants after vernalization was observed, and the plants with the C2A chromosome region detected by the BoFLC2 marker significantly tended to have no flowering after vernalization. In analyses of the BC2 and BC3F1 populations using C2 chromosome specific DNA markers, the #81 lines contained only the C2 chromosome. Similarly, a whole genome scan using IntroMap confirmed the #174-12-26 line contains the BoFLC2 region and a small C9 chromosome region encompassing no flowering related genes (Shea et al. 2017). This suggests that BoFLC2 acts as a repressor of flowering, consistent with previous reports (Okazaki et al. 2007, Ridge et al. 2014). On the other hand, we have the unexpected results that some plants in the BC3F1 and BC3F2 generations, where the BoFLC2 gene was not detected (BrFLC2 homozygous plants), exhibited no flowering or delayed flowering (Figs. 5, 6) but the subsequent progeny of these plants failed to maintain tolerance to vernalization such that they revealed a flowering response like that of the B. rapa parent (Fig. 6, Supplemental Fig. 4). This restoration of the B. rapa vernalization type, in the subsequent generations of the two lines homozygous for the BrFLC2 gene, is not caused by deleting other introgressed C genomic regions from the lines, as the previously mentioned genome-wide analysis revealed there are no additional C genome regions. Therefore, the failure of the BrFLC2 homozygous plants to flower after vernalization in the BC3F1 and the subsequent populations may possibly be explained by genome shock (McClintock 1984). Either via the perturbation in genome-wide expression that results from either monosomy or nullisomy for an A genome chromosome, similar to C genomic results reported in B. napus (Zhu et al. 2015), or through the loss of transgenerational DNA methylation marks at the introgressed locus. Introgression resulting in a disruption of epigenetic regulatory activity via loss of the transgenerational inheritance of DNA methylation was reported in Arabidopsis ddm1 mutants, and recovered by complementation with DDM1 (Ito et al. 2015, Kakutani et al. 1998). This would explain the phenotypic instability observed in the earlier generations, followed by stabilization in subsequent generations (Comai et al. 2003). However, analysis of genome-wide DNA methylation levels and gene expression in the flowering and non-flowering BC3F1 and BC3F2 populations is required to confirm this idea.
The BC3F3 population exhibits several interesting traits that require further investigation
During observation of the growth of the BC3F3 C2 introgression lines, the plants exhibited several interesting traits that merit further investigation. Notably, in most cases, the BoFLC2 introgressed plants exhibited a perennial growth habit (Supplemental Fig. 8). This is in contrast to CR Kanki which after flowering and setting seed rapidly senesces. After vernalization and the flowering of the main stem, the main stem transitioned to a woody stem, whereas the axial shoots remained vegetative and later formed heads. However, the perenniality of this line did not appear to persist, unlike B. oleracea. Plants exhibiting perennial growth characteristics appear to do so because of a failure to silence BoFLC2 production via vernalization, with cold treatment resulting in the shoot apical meristem proceeding to bolt, followed by flowering. Axial shoots that maintained vegetative growth in these plants may be influenced by continued expression of BoFLC2 in the root tissues or in the remaining aerial parts, due to their physical proximity to the crown.
The results obtained from the field trials showed that the proportion of flowering plants increased for all cultivars under cropping patterns two and three, with the exception of CR Kanki which always flowered in all three cropping patterns (Supplemental Table 2). This data shows that the C2 introgressed region alters the flowering response of CR Kanki from an early-heading type, to a late-heading type. This delay in flowering may possibly be attributed to the introduction of the alien FLC allele BoFLC2. Previous research in both B. rapa and B. oleracea identified FLC2 as a key functional regulator of flowering time (Irwin et al. 2016, Kitamoto et al. 2013, Ridge et al. 2014). Allelic variation, derived from nucleotide variations present in the first intron, appears to be at least partially responsible for such flowering time variations in both B. rapa (Kitamoto et al. 2013) and B. oleracea (Irwin et al. 2016). However, the location of the intron one nucleotide variations appears to differ between BrFLC2 and BoFLC2, and furthermore, the underlying molecular mechanism ultimately responsible for the observed variation to flowering time as a result of these intron one polymorphisms is currently unknown. Thus, our developed line may be a useful material for further research into such variations. However, a near isogenic line (NIL) containing only the BoFLC2 gene is required, because other genes within the introgressed C2 segment may also affect flowering time.
In this study, we clarified alien chromosome inheritance patterns from BC2 to BC3 populations during the process of the construction of a CSSL of B. rapa, with phenotypic evaluation of the vernalization requirement. In addition, we show that B. oleracea provides an additional source of germplasm useful for the production of vernalization tolerant B. rapa cultivars of agricultural significance, allowing for the creation of a new winter season cropping-pattern Chinese cabbage.
Acknowledgements
This work was supported by Grant-in-Aid for Scientific Research (B) (15H04433) (JSPS) to K. Okazaki.
Literature Cited
- Akaba, M., Y. Kaneko, K. Hatakeyama, M. Ishida, S.W. Bang and Y. Matsuzawa (2009) Identification and evaluation of clubroot resistance of radish chromosome using a Brassica napus-Raphanus sativus monosomic addition line. Breed. Sci. 59: 203–206.
- Andrés, F. and G. Coupland (2012) The genetic basis of flowering responses to seasonal cues. Nat. Rev. Genet. 13: 627–639.
- Axelsson, T., O. Shavorskaya and U. Lagercrantz (2001) Multiple flowering time QTLs within several Brassica species could be the result of duplicated copies of one ancestral gene. Genome 44: 856–864.
- Berry, S. and C. Dean (2015) Environmental perception and epigenetic memory: mechanistic insight through FLC. Plant J. 83: 133–148.
- Chen, B.Y., W.K. Heneen and R. Jonsson (1988) Independent inheritance of Erucic acid content and flower colour in the C-genome of Brassica napus L. Plant Breed. 100: 147–149.
- Cheng, F., J. Wu and X. Wang (2014) Genome triplication drove the diversification of Brassica plants. Hortic. Res. 1: 14024.
- Chèvre, A.M., P. Barret, F. Eber, P. Dupuy, H. Brun, X. Tanguy and M. Renard (1997a) Selection of stable Brassica napus–B. juncea recombinant lines resistant to blackleg (Leptosphaeria maculans). 1. Identification of molecular markers, chromosomal and genomic origin of the introgression. Theor. Appl. Genet. 95: 1104–1111.
- Chèvre, A.M., F. Eber, P. Barret, P. Dupuy and J. Brace (1997b) Identification of the different Brassica nigra chromosomes from both sets of B. oleracea-B. nigra and B. napus-B. nigra addition lines with a special emphasis on chromosome transmission and self-incompatibility. Theor. Appl. Genet. 94: 603–611.
- Comai, L., A. Madlung, C. Josefsson and A. Tyagi (2003) Do the different parental “heteromes” cause genomic shock in newly formed allopolyploids? Philos. Trans. R. Soc. Lond., B, Biol. Sci. 358: 1149–1155.
- Friend, D.J.C. (1985) Brassica. In: Halevy, A. (ed.) Handbook of Flowering, Vol. II. CRC Press, Boca Raton, Florida, pp. 48–77.
- Golicz, A.A., P.E. Bayer, G.C. Barker, P.P. Edger, H.R. Kim, P.A. Martinez, C.K.K. Chan, A. Severin-Ellis, W.R. McCombie, I.A. Parkin et al. (2016) The pangenome of an agronomically important crop plant Brassica oleracea. Nat. Commun. 7: 13390.
- Helliwell, C.A., R.S. Anderssen, M. Robertson and E.J. Finnegan (2015) How is FLC repression initiated by cold? Trends Plant Sci. 20: 76–82.
- Heneen, W.K., M. Geleta, K. Brismar, Z. Xiong, J.C. Pires, R. Hasterok, A.I. Stoute, R.J. Scott, G.J. King and S. Kurup (2012) Seed colour loci, homoeology and linkage groups of the C genome chromosomes revealed in Brassica rapa-B. oleracea monosomic alien addition lines. Ann. Bot. 109: 1227–1242.
- Ihaka, R. and R. Gentleman (1996) R: A language for data analysis and graphics. J. Comput. Graph. Stat. 5: 299–314.
- Irwin, J.A., E. Soumpourou, C. Lister, J.D. Ligthart, S. Kennedy and C. Dean (2016) Nucleotide polymorphism affecting FLC expression underpins heading date variation in horticultural brassicas. Plant J. 87: 597–605.
- Ito, H., T. Saito and T. Hatayama (1966) Time and temperature factors for the flower formation in cabbage. Tohoku J. Agric. Res. 17: 1–15.
- Ito, T., Y. Tarutani, T.K. To, M. Kassam, E. Duvernois-Berthet, S. Cortijo, K. Takashima, H. Saze, A. Toyoda, A. Fujiyama et al. (2015) Genome-wide negative feedback drives transgenerational DNA methylation dynamics in Arabidopsis. PLoS Genet. 11: e1005154.
- Kakizaki, T., T. Kato, N. Fukino, M. Ishida, K. Hatakeyama and S. Matsumoto (2011) Identification of quantitative trait loci controlling late bolting in Chinese cabbage (Brassica rapa L.) parental line Nou 6 gou. Breed. Sci. 61: 151–159.
- Kakutani, T., K. Munakata, E.J. Richards and H. Hirochika (1998) Meiotically and mitotically stable inheritance of DNA hypomethylation induced by ddm1 mutation in Arabidopsis thaliana. Genetics 151: 831–838.
- Kaneko, Y., T. Natsuaki, S.W. Bang and Y. Matsuzawa (1996) Identification and evaluation of Turnip Mosaic Virus (TuMV) resistance gene in kale monosomic addition lines of radish. Breed. Sci. 46: 117–124.
- Kaneko, Y. and S.W. Bang (2014) Interspecific and intergeneric hybridization and chromosomal engineering of Brassicaceae crops. Breed. Sci. 64: 14–22.
- Karim, M.M., N.N. Tonu, M.S. Hossain, T. Funaki, M.B. Meah, D.M. Hossain, M.A.U. Doullah, E. Fukai and K. Okazaki (2016) Marker-assisted selection of low erucic acid quantity in short duration Brassica rapa. Euphytica 208: 535–544.
- Keller, T., J. Abbott, T. Moritz and P. Doerner (2006) Arabidopsis REGULATOR OF AXILLARY MERISTEMS1 controls a leaf axil stem cell niche and modulates vegetative development. Plant Cell 18: 598–611.
- Kikuchi, M., K. Ogawa, T. Yamazaki, S. Kajiwara, A. Sugio, S. Nakamura and K. Shishido (1999) Secretional expression of a Bacillus subtilis xylanase gene in the Basidiomycete Coprinus cinereus. FEMS Microbiol. Lett. 178: 277–282.
- Kim, S.Y., B.S. Park, S.J. Kwon, J. Kim, M.H. Lim, Y.D. Park, D.Y. Kim, S.C. Suh, Y.M. Jin, J.H. Ahn et al. (2007) Delayed flowering time in Arabidopsis and Brassica rapa by the overexpression of FLOWERING LOCUS C (FLC) homologs isolated from Chinese cabbage (Brassica rapa L. ssp. pekinesis). Plant Cell Rep. 26: 327–336.
- Kitamoto, N., S. Yui, K. Nishikawa, Y. Takahata and S. Yokoi (2013) A naturally occurring long insertion in the first intron in the Brassica rapa FLC2 gene causes delayed bolting. Euphytica 196: 213–223.
- Lee, H., S.S. Suh, E. Park, E. Cho, J.H. Ahn, S.G. Kim, J.S. Lee, Y.M. Kwon and I. Lee (2000) The AGAMOUS-LIKE 20 MADS domain protein integrates floral inductive pathways in Arabidopsis. Genes Dev. 14: 2366–2376.
- Lee, K.H. and H. Namai (1992) Stabilization of new types of diploids (2n = 22, 24) through selfing of aneuploids (2n = 21, 22) derived from crossing of sesquidiploids (2n = 29, AAC) and Brassica campestris (2n = 20 AA). Euphytica 60: 1–13.
- Leflon, M., F. Eber, J.C. Letanneur, L. Chelysheva, O. Coriton, V. Huteau, C.D. Ryder, G. Barker, E. Jenczewski and A.M. Chèvre (2006) Pairing and recombination at meiosis of Brassica rapa (AA) × Brassica napus (AACC) hybrids. Theor. Appl. Genet. 113: 1467–1480.
- Lenth, R.V. (2016) Least-squares means: The R package lsmeans. J. Stat. Softw. 69: 1–33.
- Li, P., D. Filiault, M.S. Box, E. Kerdaffrec, C. van Oosterhout, A.M. Wilczek, J. Schmitt, M. McMullhan, J. Bergelson, M. Nordborg et al. (2014) Multiple FLC haplotypes defined by independent cis-regulatory variation underpin life history diversity in Arabidopsis thaliana. Genes Dev. 4: 1635–1640.
- Li, X.F., S.X. Xuan, J.L. Wang, S.L. Zhang, Y.H. Wang, C.H. Zhang, S.X. Shen and E.Q. Shen (2013) Generation and identification of Brassica alboglabra–Brassica campestris monosomic alien addition lines. Genome 56: 171–177.
- Lin, S., J. Wang, S. Poon, C. Su, S. Wang and T. Chiou (2005) Differential regulation of FLOWERING LOCUS C expression by vernalization in cabbage and Arabidopsis. Plant Physiol. 137: 1037–1048.
- Liu, S., Y. Liu, X. Yang, C. Tong, D. Edwards, I.A. Parkin, M. Zhao, J. Ma, J. Yu, S. Huang et al. (2014) The Brassica oleracea genome reveals the asymmetrical evolution of polyploid genomes. Nat. Commun. 5: 3930.
- Lu, C.M., B. Zhang, F. Kakihara and M. Kato (2001) Introgression of genes into cultivated Brassica napus through resynthesis of B. napus via ovule culture and the accompanying change in fatty acid composition. Plant Breed. 120: 405–410.
- Martynov, V.V. and E.E. Khavkin (2005) RFLP analysis of the FLOWERING LOCUS C gene in six Brassica species. Russ. J. Plant Physiol. 52: 352–358.
- McClintock, B. (1984) The significance of responses of the genome to challenge. Science 226: 792–801.
- McGrath, J.M., C. Quiros, J. Harada and B. Landry (1990) Identification of Brassica oleracea monosomic alien chromosome addition lines with molecular markers reveals extensive gene duplication. Mol. Gen. Genet. 223: 198–204.
- Michaels, S.D. and R.M. Amasino (1999) FLOWERING LOCUS C encodes a novel MADS domain protein that acts as a repressor of flowering. Plant Cell 11: 949–956.
- Mizuochi, H., H. Matsuzaki, T. Moue and K. Okazaki (2009) Diploid endosperm formation in Tulipa spp. and identification of a 1:1 maternal-to-paternal genome ratio in endosperms of T. gesneriana L. Sex. Plant Reprod. 22: 27–36.
- Murashige, T. and F. Skoog (1962) A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiol. Plant. 15: 473–497.
- Murray, M.G. and W.F. Thompson (1980) Rapid isolation of high molecular weight plant DNA. Nucleic Acids Res. 8: 4321–4325.
- Nagaoka, T., M.A.U. Doullah, S. Matsumoto, S. Kawasaki, T. Ishikawa, H. Hori and K. Okazaki (2010) Identification of QTLs that control clubroot resistance in Brassica oleracea and comparative analysis of clubroot resistance genes between B. rapa and B. oleracea. Theor. Appl. Genet. 120: 1335–1346.
- Okazaki, K., K. Kurimoto, I. Miyajima, A. Enami, H. Mizuochi, Y. Matsumoto and H. Ohya (2005) Induction of 2n pollen in tulips by arresting the meiotic process with nitrous oxide gas. Euphytica 143: 101–114.
- Okazaki, K., K. Sakamoto, R. Kikuchi, A. Saito, E. Togashi, Y. Kuginuki, S. Matsumoto and M. Hirai (2007) Mapping and characterization of FLC homologs and QTL analysis of flowering time in Brassica oleracea. Theor. Appl. Genet. 114: 595–608.
- Onouchi, H., M.I. Igeno, C. Perilleux, K. Graves and G. Coupland (2000) Mutagenesis of plants overexpressing CONSTANS demonstrates novel interaction among Arabidopsis flowering-time genes. Plant Cell 12: 885–900.
- Peterka, H., H. Budahn, O. Schrader, R. Ahne and W. Schütze (2004) Transfer of resistance against the beet cyst nematode from radish (Raphanus sativus) to rape (Brassica napus) by monosomic chromosome addition. Theor. Appl. Genet. 109: 30–41.
- Poethig, R.S. (2010) The past, present, and future of vegetative phase change. Plant Physiol. 154: 541–544.
- Razi, H., E.C. Howell, H.J. Newbury and M.J. Kearsey (2008) Does sequence polymorphism of FLC paralogues underlie flowering time QTL in Brassica oleracea? Theor. Appl. Genet. 116: 179–192.
- Ridge, S., P.H. Brown, V. Hecht, R.G. Driessen and J.L. Weller (2014) The role of BoFLC2 in cauliflower (Brassica oleracea var. botrytis L.) reproductive development. J. Exp. Bot. 66: 125–135.
- Schiessl, S., B. Samans, B. Hüttel, R. Reinhard and R.J. Snowdon (2014) Capturing sequence variation among flowering-time regulatory gene homologs in the allopolyploid crop species Brassica napus. Front. Plant Sci. 5: 404.
- Schranz, M.E., P. Quijada, S.B. Sung, L. Lukens, R. Amasino and T.C. Osborn (2002) Characterization and effects of the replicated flowering time gene FLC in Brassica rapa. Genetics 162: 1457–1468.
- Shea, D.J., M. Shimizu, N. Nishida, E. Fukai, T. Abe, R. Fujimoto and K. Okazaki (2017) IntroMap: a signal analysis based method for the detection of genomic introgressions. BMC Genet. 18: 101.
- Tadege, M., C.C. Sheldon, C.A. Helliwell, P. Stoutjesdijk, E.S. Dennis and W.J. Peacock (2001) Control of flowering time by FLC orthologues in Brassica napus. Plant J. 28: 545–553.
- Tsunoda, S., K. Hinata and C. Gómez-Campo (1980) Brassica crops and wild allies: Biology and breeding. Japan Scientific Societies Press, Tokyo, Japan, pp. 191–201.
- U, N. (1935) Genome analysis in Brassica with special reference to the experimental formation of B. napus and peculiar mode of fertilization. Japan. J. Bot. 7: 389–452.
- Wu, J., K. Wei, F. Cheng, S. Li, Q. Wang, J. Zhao, G. Bonnema and X. Wang (2012) A naturally occurring InDel variation in BraA.FLC.b (BrFLC2) associated with flowering time variation in Brassica rapa. BMC Plant Biol. 12: 151.
- Yang, H., M. Howard and C. Dean (2014) Antagonistic roles for H3K36me3 and H3K27me3 in the cold-induced epigenetic switch at Arabidopsis FLC. Curr. Biol. 24: 1793–1797.
- Yano, R., Y. Takebayashi, E. Nambara, Y. Kamiya and M. Seo (2013) Combining association mapping and transcriptomics identify HD2B histone deacetylase as a genetic factor associated with seed dormancy in Arabidopsis thaliana. Plant J. 74: 815–828.
- Yuan, Y.X., J. Wu, R.F. Sun, X.W. Zhang, D.H. Xu, G. Bonnema and X.W. Wang (2009) A naturally occurring splicing site mutation in the Brassica rapa FLC1 gene is associated with variation in flowering time. J. Exp. Bot. 60: 1299–1308.
- Zhao, J., V. Kulkarni, N. Liu, D. Pino del Carpio, J. Bucher and G. Bonnema (2010) BrFLC2 (FLOWERING LOCUS C) as a candidate gene for a vernalization response QTL in Brassica rapa. J. Exp. Bot. 61: 1817–1825.
- Zhu, B., Y. Shao, Q. Pan, X. Ge and Z. Li (2015) Genome-wide gene expression perturbation induced by loss of C2 chromosome in allotetraploid Brassica napus L. Front. Plant Sci. 6: 763.
