Single nucleotide polymorphisms in a regulatory site of VRN-A1 first intron are associated with differences in vernalization requirement in winter wheat

Single nucleotide polymorphisms in a regulatory site of VRN-A1 first intron are associated with... Winter wheats require a long exposure to cold temperatures (vernalization) to accelerate flowering. However, varieties dif- fer in the length of the period of cold required to saturate the vernalization response. Here we show that single nucleotide polymorphisms (SNP) at the binding site of the GRP2 protein in the VRN-A1 first intron (henceforth, RIP3) are associated with significant differences in heading time after a partial vernalization treatment. The ancestral winter VRN-A1 allele in ‘Triple Dirk C’ has one SNP in the RIP3 region (1_SNP) relative to the canonical RIP3 sequence, whereas the derived ‘Jagger’ allele has three SNPs (3_SNPs). Both varieties have a single VRN-A1 copy encoding identical proteins. In an F population generated from a cross between these two varieties, plants with the 3_SNPs haplotype headed significantly earlier (P < 0.001) than those with the 1_SNP haplotype, both in the absence of vernalization (17 days difference) and after 3-weeks of vernalization (11 days difference). Plants with the 3_SNPs haplotype showed higher VRN-A1 transcript levels than those with the 1_SNP haplotype. The 3_SNPs haplotype was also associated with early heading in a panel of 127 winter wheat varieties grown in three separate controlled-environment experiments under partial vernalization (36 to 54 days, P < 0.001) and one experiment under field conditions (21 d, P < 0.0001). The RIP3 polymorphisms can be used by wheat breeders to develop winter wheat varieties adapted to regions with different duration or intensity of the cold season. Keywords Wheat · Flowering · Vernalization · VRN1 · RIP3 · GRP2 Abbreviations FT1 FLOWERING LOCUS T1 GRP2 GLYCINE RICH RNA-BINDING PROTEIN 2 RIP3 RNA immune precipitation fragment 3 Communicated by S. Hohmann. SNP Single nucleotide polymorphism VRN1 VERNALIZATION1 gene Electronic supplementary material The online version of this article (https ://doi.org/10.1007/s0043 8-018-1455-0) contains VRN2 VERNALIZATION2 gene supplementary material, which is available to authorized users. * Jorge Dubcovsky Department of Plant Sciences, University of California, jdubcovsky@ucdavis.edu Davis, CA 95616-8515, USA Nestor Kippes Present Address: Department of Plant Biology and Genome nfkippes@ucdavis.edu Center, University of California, Davis, CA 95616, USA Mohammed Guedira Department of Crop Science, North Carolina State mguedir@ncsu.edu University, Raleigh, NC 27695, USA Lijuan Lin USDA-ARS Plant Science Research Unit, Raleigh, lljlin@ucdavis.edu NC 27695, USA Maria A. Alvarez Howard Hughes Medical Institute, Chevy Chase, MD 20815, malvarez@ucdavis.edu USA Gina L. Brown-Guedira gbrowng@ncsu.edu Vol.:(0123456789) 1 3 Molecular Genetics and Genomics VRN3 VERNALIZATION3 gene tetraploid and hexaploid wheat (e.g., by marker assisted VRN-D4 VERNALIZATION4 gene D genome selection) can generate spring wheats (Distelfeld et  al. 2009b; Kippes et al. 2016). During the fall, VRN2 represses the expression of VRN3 Introduction (= FT1), a homolog of the Arabidopsis flowering promoter FLOWERING LOCUS T (FT) (Yan et al. 2006). FT encodes Wheat is one of the most widely cultivated cereals and an a mobile protein that travels from the leaves to the apex important staple food worldwide. Almost 70% of 740M (Corbesier et al. 2007; Tamaki et al. 2007) where it forms tonnes of annual production are used for direct human con- a complex with FD-like and 14-3-3 proteins. This complex sumption making wheat yields crucial for human nutrition binds to the VRN1 promoter and induces flowering (Taoka and global food security (FAOSTAT 2015). Recent esti- et al. 2011; Li et al. 2015). During the winter, a slight induc- mates of the impact of climate change on crops yield have tion of VRN1 is sufficient to repress VRN2, which favors raised the interest in understanding how plant development FT1 up-regulation when days get longer during spring (Chen is modulated by environmental cues (Cang et al. 2016; and Dubcovsky 2012). These interactions result in a positive Cook et al. 2012; Franks et al. 2007; Liu et al. 2016). feedback-loop that irreversibly promotes wheat flowering in Wheat varieties are divided into two major categories the spring (Loukoianov et al. 2005). (winter and spring) based on their growth habit. Winter The natural allelic variation in VRN-A1 responsible for wheats require a long exposure to low temperatures (vernali- differences between spring and winter wheats is well char - zation) to accelerate the transition from the vegetative to the acterized. However, less is known about the role of VRN- reproductive phase. This requirement prevents exposure of A1 on the differences among winter wheats in the duration delicate floral meristems to damaging freezing temperatures of the cold period required for saturating the vernalization (Distelfeld et al. 2009a; Woods et al. 2016). By contrast, response. Three independent studies found that a large pro- spring wheats have reduced or no vernalization requirement. portion of the variation in vernalization requirement among Winter wheats are sown in fall in regions where wheat can winter wheat varieties is linked to the VRN-A1 locus. How- tolerate winter freezing temperatures. However, where win- ever, the three studies propose alternative explanations for ter temperatures are too low, spring wheats are sown in the these differences. Diaz et al. (2012) suggested that the dif- spring to avoid freezing. In Mediterranean regions, with ferences were caused by the presence of a single copy of mild and rainy winters, spring wheats are planted in the fall VRN-A1 in ‘Claire’ and three in ‘Hereward’. Li et al. (2013) to take advantage of water availability during the winter. argued that the differences in heading time were caused Wheat varieties sown in these contrasting regions show by an amino acid polymorphism at position 180 between different allele profiles at the main vernalization gene VER- ‘Jagger’ (alanine) and ‘2174’ (valine). More recently, Kip- NALIZATION1 (VRN1), a MADS-box transcription factor pes et al. (2015) found that both ‘Claire’ and ‘Jagger’ also homologous to the meristem identity gene APETALA 1 differed from ‘Hereward’ and ‘2174’ by single nucleotide (AP1) in Arabidopsis (Yan et  al. 2003; Trevaskis et  al. polymorphisms (SNPs) at the binding site of a GLYCINE 2003). Winter wheats carry the ancestral VRN1 alleles, RICH RNA-BINDING PROTEIN 2 (GRP2) in a region of whereas spring wheats show deletions or mutations in the VRN-A1 first intron designated as RNA Immune Precipi- regulatory regions located in the promoter or first intron. tation fragment 3 (RIP3, Xiao et al. 2014 and; Kippes et al. Changes in these regions have been observed in all three 2015). RIP3 natural polymorphisms were shown to disrupt VRN1 homologs (VRN-A1, VRN-B1 and VRN-D1), as well the binding of GRP2 to the RIP3 site (Kippes et al. 2015), as in the relatively recent duplicated paralog VRN-D4 (Fu but their effects on VRN1 expression and heading time were et al. 2005; Konopatskaia et al. 2016; Muterko et al. 2015, not characterized. No variation in the RIP3 site was detected 2016; Yan et al. 2004a, 2003; Kippes et al. 2015; Chu et al. in the VRN-B1 or VRN-D1 genes (Kippes et al. 2015). 2011; Pidal et al. 2009). In this study, we show that polymorphisms in the VRN-A1 Additional differences in wheat vernalization require- RIP3 region segregating in an F population are associated ment have been detected at the VERNALIZATION2 (VRN2) with differences in VRN-A1 transcript levels and heading and VERNALIZATION3 (VRN3) loci. VRN2 encodes a pro- time independently of differences in VRN-A1 copy number tein with a putative zinc finger and a CCT domain (ZCCT) and protein sequence. We also show that polymorphisms in that acts as a dominant long-day repressor of VRN3 and the RIP3 region in a panel of winter wheat varieties from dif- flowering. VRN2 deletions or loss-of-function mutations ferent geographic regions are associated with differences in result in spring growth habit in diploid wheat and barley heading time. The distribution of RIP3 alleles and VRN-A1 accessions (Yan et al. 2004b), but have not been observed copy number among winter wheat varieties and the utiliza- so far in natural polyploid wheat species. Combination of tion of these variants in winter wheat breeding are discussed. deletions or non-functional vrn2 alleles at all homologs in 1 3 Molecular Genetics and Genomics Allele Specific PCR (KASP) fluorescent assays were used Materials and methods to detect polymorphism among the 127 accessions at wheat vernalization loci. PCR was run according to manufacturer’s Plant material instructions, using a reaction volume of 4 µL, which con- sisted of 2 µL 2× KASPar reaction mix, 0.05 µL 72× assay The parental lines of the F population analyzed in this −1 mix, and 2 µL of template DNA (10 ng µL ). Endpoint study were the winter wheat lines Triple Dirk C (TDC) and genotyping was conducted using the software KlusterCaller CS5402 (Kippes et al. 2014, 2015). TDC is part of a set of (LGC Genomics, Hoddeson, UK). near isogenic lines for vernalization alleles described in Pug- Assays developed from published sequences of causal sley 1971 and 1972. CS5402 is a substitution line of chromo- genes were used to distinguish lines in the core collection some 5D of Chinese Spring (dominant Vrn-D1 allele) by the possessing spring alleles at Vrn-A1, Vrn-B1, Vrn-D1, and Ae. tauschii chromosome 5D (recessive vrn-D1 allele). Both Vrn-B3 loci (Supplemental Table 1). The exception was the lines have a single copy of the VRN-A1 gene homozygous KASP assay wMAS000033 used for detection of the Vrn- for the recessive vrn-A1 allele, which encodes for identical A1a spring allele developed from the contextual sequences VRN-A1 proteins (alanine residue at position 180, Li et al. of iSelect SNP marker IWA0001 determined to be associated 2013). The two accessions differ in the sequence of the RIP3 with Vrn-A1a. Accessions carrying alleles for spring growth region in intron 1. Relative to the RIP3 canonical sequence, habit at any VRN locus were excluded from further analysis. the VRN-A1 haplotype from TDC has one SNP (henceforth The VRN-A1 RIP3 region was sequenced in the 127 win- 1_SNP, GenBank AY747600) and the one from CS5402 ter accessions by Sanger sequencing using primers listed in three SNPs (henceforth 3_SNPs, GenBank KR422423, Kip- Table 1. Genotyping of the RIP3 alleles in the F population pes et al. 2015). was conducted with a KASP assay (Table 1) run on a 7500 For the association study, we used a panel of 127 win- Fast Real-Time PCR system (Applied Biosystems). VRN-A1 ter lines from the USDA National Germplasm Collection copy number was determined using a TaqMan assay from from diverse geographical origin. Sixty-five percent of the four biological replications per genotype as described in lines were collected from 31 different countries in Europe, Kippes et al. (2015). Differences in flowering time between 15% from 9 countries in Asia, 18% from 4 countries in the VRN-A1 copy number classes were analyzed using Tukey’s Americas, and 1% from Australia. These 127 accessions test (P < 0.05). The VRN-A1 promoter from Triple Dirk C were selected from a larger set after filtering out accessions (TDC) was sequenced by Sanger sequencing (GenBank carrying alleles associated with a spring growth habit using MH347747) using four overlapping PCR products ampli- available molecular markers (Yan et al. 2004b, 2006; Fu fied with primers listed in Table S3. et al. 2005). Principal component analysis Molecular markers and VRN1 polymorphisms Accessions from the winter panel were previously genotyped Genomic DNA was isolated from individual plants for the F with the Illumina Infinium 9K iSelect platform as part of study and from bulks of leaves from four seedlings of each the characterization of the NSGC core collection of winter accession for the winter wheat panel (MAG-Bind Plant DNA and facultative common wheat (Bonman et al. 2015). Geno- Plus 96 kit, Omega Bio-Tek, Norcross, GA). Kompetitive types of 4,483 markers for all accessions were obtained from Table 1 List of primers used Use Name Primer sequence (5′–3′) for RIP3 characterization and VRN-A1 transcripts abundance RIP3 A genome sequencing RIP-3_F1AAT CAC ACC TCA GGA TTT CAT by RT-qPCR RIP-3_R1GAT GGG TCA TAA GGT TTT GC RIP3 A genome Kasp assay 1_SNP_FTCT CAC AGT CAT TGT TGT TGG TAT G 3_SNPs_FTCT CAC AGT CAT TGT TGT TGG TAT C ReverseAGC AAT CAA GTT GTA ACA TAA ATA ATTA VRN1 RT-qPCR full tran- q.VRNA1-L-F1TCC ACC GAG TCA TGT ATG GA script q.VRNA1-L-R1GAG AAC CTT TTC TGC ATA AGAA VRN1 RT-qPCR short q.VRNA1-S-F1CAC CAA GGG AAA GCT CTA CG transcript q.VRNA1-S-R1GTT AAC TTG TAA CTG GGA GCTAA KASP assay primers do not include tail sequences 1 3 Molecular Genetics and Genomics The Triticeae Toolbox (https ://triti ceaet oolbo x.org/wheat ). photoperiod. Heading date was noted as the number of days Principal component analysis was conducted in Tassel v5.0 after transplanting when the spike fully emerged from the (http://www.maize genet ics.net/tasse l) using the covariance boot (Zadoks 60, Zadoks et al. 1974). The experiment was matrix for markers having minor allele frequency greater ended 175 days after planting. than or equal to 0.05. Lines missing more than 10% of data were removed. Greenhouse After the vernalization treatment described above, plants RT‑qPCR were grown at greenhouse facilities at UCD (four replica- tions) and NCSU (up to six replications) in a completely The last expanded leaf of four biological replications per randomized design. Supplemental lighting was used to pro- genotype was collected in liquid nitrogen, and RNA was vide 16-h light/8-h dark photoperiod. During the growth extracted using the Spectrum Plant Total RNA Kit (Sigma- period, the minimum temperatures during the night varied Aldrich). RNA samples were treated with DNase I (RQ1 from 10 to 18 °C and the maximum temperatures during the RNase-Free DNase, Promega) and first strand cDNAs were day ranged from 21 to 32 °C. The experiment was ended 210 synthesized from 1 µg of total RNA using the High Capacity days after planting. Heading date is reported as the number Reverse Transcription Kit (Applied Biosystems). Quantita- of days after transplanting when the spike fully emerged tive RT-PCR was performed using SYBR Green and a 7500 from the boot (Zadoks 60, Zadoks et al. 1974). Fast Real-Time PCR system (Applied Biosystems). ACTIN was used as an endogenous control. Primers are detailed Field in Table 1; primers for ACTIN were previously described (Dubcovsky et al. 2006). Transcript levels were expressed Three grams of seed of each entry were sown in a single as linearized fold-ACTIN levels calculated by the formula (ACTIN CT − TARGET CT) 1.5  m row at the NCSU Research Farm at Raleigh, NC 2 ± standard error (SE) of the mean. (35.73°N, 78.68°W; elevation: 116.5  m) on October 17, The resulting number indicates the ratio between the initial 2015. Heading date was recorded as the number days after number of molecules of the target gene and the number of January 1st, when 50% of the plants in a plot were at Zadoks molecules of ACTIN. 60 growth stage. Phenotyping Results Vernalization The RIP3 region at the first intron of VRN‑A1 Seeds were germinated in 2 × 2 plastic inch pots filled with is associated with differences in flowering time vermiculite in a greenhouse with 16-h light and 8-h dark photoperiod. One week after planting, seedlings were moved To determine the effect of the two VRN-A1 RIP3 natural to a growth chamber for 3 weeks of vernalization at 4 °C alleles on heading time, we intercrossed TDC (1_SNP allele) under 16-h light/8-h dark photoperiod. Similar conditions with CS5402 (3_SNPs allele). These two lines have a single were used in the facilities at the University of California, VRN-A1 copy encoding identical proteins. We generated an Davis (UCD) and North Carolina State University (NCSU). F population of 142 plants segregating for these haplotypes, After vernalization, up to 10 seedlings were transplanted genotyped them for the RIP3 alleles and recorded heading into 0.7 L black plastic containers (Stuewe and Sons, Tan- time under three different vernalization treatments (no ver - gent, OR) with 1:1 Metro Mix 2 and soil. Two grams of nalization, 3 weeks vernalization and 7 weeks vernalization). slow release fertilizer (Multicote 14–14–16, N-P-K) was In the absence of vernalization, the parental line incorporated with the soil in each cone. Cones were placed CS5402 headed 30.2 days earlier than TDC, whereas F in greenhouses or growth chambers under the conditions plants homozygous for the 3_SNPs haplotype flowered described below. 16.9 days earlier than plants homozygous for the 1_SNP haplotype (P < 0.0001, Fig. 1a). When plants were vernal- Growth chamber ized for 3 weeks, the difference in heading time between parental lines was reduced to 12 days (TDC later than After vernalization, four replications of each entry were CS5402) and those between F homozygous plants to grown in a controlled-environment chamber at NCSU in a 11.3 days (1_SNP haplotype later than 3_SNPs haplo- completely randomized design. Plants were grown under type, P < 0.0001, Fig.  1b). After 7 weeks of vernaliza- 20/18  °C day/night temperature and 16-h light/8-h dark tion, the difference in heading time between the parental 1 3 Molecular Genetics and Genomics Flowering time differences are linked to differences in VRN‑A1 expression levels To study the effects of RIP3 polymorphisms on VRN-A1 transcript levels, we sampled plants from the F popula- tion homozygous for the 1_SNP and 3_SNPs haplotypes at different time points of the partial vernalization treat- ment. Leaf samples were collected from 3-week-old plants immediately before vernalization (3w), 48  h after the plants were transfer to 4 °C (48 h), after 1 and 3 weeks of vernalization (1wV and 3wV) and 3 weeks after the plants were returned to room temperature (3wR). Since two alternative splice variants of VRN-A1 were detected in the study of Xiao et al. (2014), we designed specific primers to amplify each variant separately. The long VRN-A1 transcript variant (henceforth, long vari- ant) encodes the complete gene, whereas the short VRN- A1 transcript variant (600 bp, henceforth, short variant) includes the complete first exon (185 bp) and a small por - tion of the first intron. The short transcript ends a few base pairs downstream of the RIP3 region, located 2,767 bp downstream of the VRN-A1 start codon in TDC (Fig. 2a). Figure 2b presents the transcript levels of the long variant, Fig. 2c the short variant, and Fig. 2d the ratio between the two splice variants. For the VRN-A1 long variant, the transcript levels of the F plants carrying the 3_SNPs haplotype were 2- to 13-fold larger than those of plants carrying the 1_SNP haplotype. However, the differences were not significant at any of the time points or in the combined repeated meas- urements ANOVA (P = 0.18, Fig. 2b). For the VRN-A1 short variant, transcript levels of the F plants carrying the 3_SNPs haplotype were significantly higher than those of plants carrying the 1_SNP haplotype in the combined repeated measures ANOVA (P = 0.0003). However, for the individual time points the difference between haplotypes was significant only at 3 weeks after vernalization (P = 0.0177, Fig. 2c 3wV). The ratio between the short and long VRN-A1 variants showed a decrease during and after vernalization (Fig. 2d). This was the result of faster increases of the long variant Fig. 1 Difference in heading time between TDC (1_SNP) and relative to the short variant (Fig. 2d). After vernalization, CS5402 (3_SNPs) and between F plants homozygous for the RIP3 when the plants were returned to room temperature, the haplotypes. Data represents heading time means of at least eight plants per genotype under three different treatments (no vernalization short/long variant ratio was from 40- to 60-fold smaller and 3 or 7 weeks of vernalization). Different letters indicate signifi- than the same ratio before vernalization (Fig.  2d), sug- cant differences (Tukey’s test P < 0.05) gesting a decreasing importance of the short variant at this time point. Before vernalization, the ratio between lines was reduced to only 2 days (TDC later than CS5402, the short and long VRN-A1 variants was three times higher P = 0.0128) and those between homozygous F plants for in the plants carrying the 1_SNP RIP3 haplotype than in the two RIP3 homozygous plants were no longer signifi- those carrying the 3_SNPs haplotype (Fig. 2d, P = 0.0462). cant (0.9 d, P = 0.0578). These results show that there are No significant differences between haplotypes were significant differences in heading time linked to polymor - detected for the other time points (Fig. 2d). phisms at the RIP3 intronic region. 1 3 Molecular Genetics and Genomics Fig. 2 Transcript levels of VRN- A1 alternative splice variants during vernalization. Transcript levels of VRN-A1 alternative splice variants were studied in F plants homozygous for the two RIP3 haplotypes. Leaf samples were collected from 3-week-old plants immediately before vernalization (3w), 48 h after the plants were transfer to 4 °C (48 h), after 1 and 3 weeks of vernalization (1 and 3wV) and 3 weeks after the plants were returned to room temperature (3wR). (A) Schematic representation of the different VRNA1 transcripts studied. Arrowheads indicate regions complementary to the qRT-PCR primers utilized. The VRN-A1 alternative splice variant includes the RIP3 region (red). VRNA1-long correspond to the complete gene (B) and VRNA1-short to the alterna- tive splice variant (C). Average ratios of short/long transcript versions are presented in D. Bars represent means of four biological replications and error bars correspond to SEM (* P < 0.05) Fig. 3 Principal component analysis (PCA) of the winter panel used cles (1_SNP) or open diamonds (3_SNPs). a First (PC1) and second in this study. The PCA was based on 4483 SNP markers. Colors indi- (PC2) principal components. b First (PC1) and third (PC3) principal cate geographical origin and RIP3 genotypes are indicated by cir- components their effect in a panel of 127 winter lines from diverse Frequency and effect of the RIP3 haplotypes on heading time in a winter wheat panel geographical origins. The 1_SNP haplotype was found in 90.5% of the accessions whereas the 3_SNPs haplotype Based on the observed effect of the RIP3 haplotypes on was found in only 9.5% of the accessions. A principal component analysis showed that lines with the 3_SNPs heading time in the F population, we decided to study 1 3 Molecular Genetics and Genomics haplotype tend to cluster together in two separate groups Table 2 ANOVA for heading time of winter wheat lines carrying dif- ferent RIP3 haplotypes of Asian or European origin (Fig.  3, Supplemental a a b c Table S2). GH NC GH UCDGrowth chamber Field To study the effect of the RIP3 haplotypes on heading 3 SNPs 106.1 ± 2.3 104.3 ± 3.2 105.8 ± 2.7 104.8 ± 1.1 time the winter wheat panel was grown in three independent 1 SNP 161.0 ± 6.5 150.1 ± 4.4 142.6 ± 7.6 126.4 ± 4.5 experiments with partial vernalization conditions and one Dif. (days) 54.9 45.8 36.8 21.6 field experiment. The first experiment was conducted in a P value 1.11E-11 3.31E-06 1.01E-05 1.95E-08 grown chamber, where plants were first exposed to 3 weeks of vernalization and then moved to room temperature condi- GH = Greenhouse after 3-weeks vernalization tions. In two additional experiments performed at the Uni- Growth chamber after 3-week vernalization versity of California Davis (UCD) and North Carolina State Field conditions after natural vernalization (Raleigh, NC) University (NCSU), plants were transferred to greenhouses after 3 weeks of vernalization. The final experiment was experiment, plants with the 3_SNPs haplotype flowered conducted under field conditions in Raleigh, North Caro- lina. In the three experiments grown under controlled-envi- 21 days earlier than the plants with the 1_SNP haplotype (Fig. 4d; Table 2). ronments, plants carrying the 3_SNPs haplotype flowered significantly earlier (36–54 days, P < 0.0001) than plants with the 1_SNP haplotype (Fig. 4a–c; Table 2). In the field Fig. 4 Heading time differences between RIP3 haplotypes in a win- ferred to greenhouses at UCD and NCSU after 3-weeks vernalization. ter wheat panel. Independent experiments were conducted under a–c c Plants were vernalized in a growth chamber at 4  °C for 3 weeks control environmental conditions (3-week vernalization, or d) under and then temperature settings were switched to warm conditions natural vernalization in the field (Raleigh, NC). a, b Plants trans- (20/18 °C day/night) until heading 1 3 Molecular Genetics and Genomics other linked SNP or genes affecting the trait. This is particu- Variation in VRN‑A1 copy number larly critical for the VRN-A1 locus that is tightly linked with the PHYC gene (0.02 cM, Yan et al. 2003). PHYC mutants To study the effect of VRN-A1 copy number variation on heading time we characterized all lines in the winter wheat affect heading time in wheat (Chen et al. 2014) and natural variation in this gene is associated with variation in flower - panel with a TaqMan assay developed by Diaz et al. (2012). VRN-A1 copy number varied from one (10.2% of the lines) ing time in Arabidopsis (Balasubramanian et al. 2006) and pearl millet (Saïdou et al. 2014). Fortunately, Li et al. (2013) to four copies (3.9% of the lines), with most lines carrying two (35.4%) or three copies (50.4%). Most of the winter were able to find recombination events between VRN-A1 and PHYC that demonstrated that the differences in vernalization lines with a single VRN-A1 copy have the RIP3 3_SNPs haplotype. The only exception was PI 627798, which has requirement between their parental lines ‘Jagger’ and ‘2174’ were linked to VRN-A1 and not to PHYC. a single VRN-A1 copy and the 1_SNP haplotype (same as TDC). PI 627798 heading time was more similar to plants with the 1_SNP haplotype and multiple VRN-A1 copies, than to plants with the 3_SNPs haplotype and a single VRN-A1 Different polymorphisms in VRN‑A1 may contribute to flowering time differences copy. Compared with the plants in the last class, PI 627798 was the latest flowering in two experiments (Fig.  4a, d) and Although there is agreement on the contribution of VRN-A1 the second latest in the remaining one (Fig. 4c, not tested in b). These results are consistent with the late flowering of the to the differences in vernalization requirement among winter wheats, there is no agreement on the current interpretation plants carrying the 1_SNP haplotype in the F population. We detected no significant differences in heading time of the causal polymorphisms. Diaz et al. (2012) suggested that the differences in heading time linked to VRN-A1 were among varieties carrying two, three or four copies of VRN- A1 in any of the three pairwise comparisons. Taken together, caused by die ff rences in VRN-A1 copy number among Claire (1 copy), Malacca (2 copies) and Hereward (3 copies). these results suggest that VRN-A1 copy number variation has limited effect on heading time under the conditions used in However, the F population they generated from the cross between Claire and Hereward segregated also for the Ala/Val our four experiments. polymorphism at position 180 of the VRN-A1 protein and for the 3_SNPs/1_SNP polymorphism in the RIP3 site of the Discussion first intron, complicating the interpretation of these results. Diaz et al. (2012) also analyzed a double haploid popu- VRN‑A1 is linked to differences in vernalization lation from the cross between Malacca and Hereward, and found that the plants carrying three VRN-A1 copies tended to requirement in winter wheat head later than the plants carrying two VRN-A1 copies after 4 weeks of vernalization. In this population, both parental The length of the vernalization period required to saturate the acceleration of flowering varies widely among winter lines have the valine residue at position 180 and the 1_SNP haplotype at the RIP3 site, increasing the chances that the wheat varieties. Some varieties reach the saturation point after only 3 weeks of vernalization (sometimes called ‘fac- observed differences in heading time were caused by the dif- ferences in VRN-A1 copy number. Similarly, Guedira et al. ultative’ types), but most varieties require approximately 6 weeks of vernalization to reach this point. In some excep- (2016) observed in a RIL population from a cross between cultivars 26R61 and AGS 2000, both having the Val 180, tional cases, vernalization treatments of up to 8 weeks are necessary to saturate the acceleration of flowering (Brooking that after 2 or 4 weeks of vernalization lines having three VRN-A1 copies flowered later than those having two VRN- 1996; Kosner and Pankova 2002). The genetic factors controlling differences in vernali- A1 copies. However, it is not possible to rule out completely the effect of a linked gene given the small size of these seg- zation requirement among winter varieties are not as well understood as those controlling differences between winter regating populations. Li et  al. (2013) proposed that the early heading time and spring varieties (Distelfeld et al. 2009a). However, stud- ies that genetically mapped genes controlling heading time observed after partial vernalization of the plants carrying the Jagger VRN-A1 allele (Ala 180) relative to 2174 (Val in winter wheat using partial vernalization treatments (3–4 weeks of cold treatment) found that at least part of these dif- 180) were caused by different amino acid residues at posi- tion 180. However, these two winter wheat varieties differed ferences were linked to the VRN-A1 locus (Diaz et al. 2012; Li et al. 2013). A similar conclusion is supported by this also in VRN-A1 copy number (Jagger one VRNA1 copy vs. 2174 two VRNA1 copies) and the RIP3 haplotype in the first study. One limitation of linkage studies using small segregating intron (Jagger 3_SNPs vs. 2174 1_SNP), complicating the interpretation of the results. populations is that they cannot rule out the possibility of 1 3 Molecular Genetics and Genomics The VRN-A1 allele in CS5402 is almost identical to the transcript (Xiao et al. 2014). During vernalization, GRP2 is alleles present in Jagger and Claire, all carrying a single O-GlcNAc modified and its levels in the nucleus decrease, VRN-A1 copy, the 3_SNPs RIP3 haplotype and Ala 180. allowing higher VRN1 mRNA accumulation (Xiao et al. Therefore, it is likely that the contrasting RIP3 haplotypes 2014). segregating in the Jagger (3_SNPs) × 2174 (1_SNP) (Li Differences in the speed and processivness of tran- et al. 2013) and Claire (3_SNPs) × Hereward (1_SNP) (Diaz scription affects the selection of alternative splicing sites et al. 2012) populations could have contributed to the differ - (de la Mata et al. 2003). Therefore, the stronger binding ences in heading time observed after partial vernalization in of the GRP2 protein to the translated RIP3 sites with the these studies. This conclusion, does not rule out the possibil- 1_SNP haplotype may favor the VRN-A1 short alterna- ity that the polymorphisms at position 180 or the differences tive splice variant, first described by Xiao et  al. (2014). in copy number could have also contributed to the observed We detected in silico a short alternative splice variant differences in heading time in the previous studies. for VRN-B1, which showed the same structure as the one The plants from the F population segregating for the described for VRN-A1 (http://plants. ensemb l.org, TGACv1, RIP3 haplotypes have a single VRN-A1 copy encoding iden- Traes_5BL_89636D032.1). However, we did not find this tical proteins, but differ in VRN-A1 transcript levels. There- short variant in the D genome of Chinese Spring, a result fore, polymorphisms at the VRN-A1 regulatory regions are that is consistent with a deletion encompassing the RIP3 site good candidates to explain the differences in heading time in the first intron of VRN-D1. linked to this gene. A comparison of the VRN-A1 promoter The binding of the GRP2 protein to the pre-mRNA RIP3 regions (2254 bp upstream from the start codon) from the site may explain the significantly higher short/long variant 1_SNP and 3_SNPs haplotypes revealed no-polymorphisms ratio observed before vernalization in the F plants carrying in the first 436 bp (Supplemental Figure S1). The rest of the 1_SNP haplotype relative to those carrying the 3_SNPs the promoter region (437–2254 bp) showed seven SNPs and haplotype in this study (Fig. 2d). The reduced GRP2 levels four indels (1–2 bp), but none of them were located within during and after vernalization would favor the long tran- known regulatory elements (Pidal et al. 2009; Kane et al. script variant and explain the decrease in the short/long 2007; Li and Dubcovsky 2008; Li et al. 2015), predicted variant ratio in these later time points. The higher relative binding sites of transcription factors, or evolutionary con- abundance of the short splice variant before vernalization served regions (Supplemental Figure S1). By contrast, the may contribute to maintain low levels of functional VRN-A1 RIP3 polymorphisms have been shown to affect the binding until the vernalization requirement is satisfied. Alternatively, of GRP2 proteins to the pre-mVRN-A1 transcripts (Kippes the 155 amino acids encoded by the short variant (includ- et al. 2015), a result consistent with the differences in rela- ing the MADS-box domain MADS_MEF2_like, cd00265) tive abundance of alternative splice variants described in the may interact with other MADS-Box proteins altering their following section. function. Transgenic experiments overexpressing this short variant will be required to test this hypothesis. Once the Differences in VRN‑A1 expression are consistent role of these alternative splice variants is better understood, with the proposed RIP3/GRP2 molecular it may provide an additional entry point to modulate the mechanisms vernalization requirement in wheat. Based on the previous results and discussion, we favor the Diaz et al. (2012) observed faster and higher VRN-A1 tran- hypothesis that the polymorphisms in the RIP3 region are script levels in Claire (3_SNPs haplotype) than in the two responsible for the differences in VRN-A1 transcript levels varieties carrying the 1_SNP haplotype. Re-analysis of and heading time between the lines carrying the 1_SNP and the expression data from Li et al. (2013) using a two-way 3_SNPs haplotypes. However, we cannot rule the possibility ANOVA with time (3 weeks and 6 weeks) and genotypes of effects caused by linked polymorphisms in the promoter (Jagger and 2174) as factors, revealed higher VRN-A1 region or outside the sequenced region. transcript levels in Jagger (3_SNPs) than in 2174 (1_SNP, P < 0.0001, Supplemental Table S2). These results are simi- lar to the ones presented here, and suggest that at least part Winter wheat varieties carrying the 3_SNPs of the differences in heading time between these two VRN- haplotype were detected at low frequencies A1 alleles are regulated at the transcriptional level. We have recently shown that the natural polymorphisms A recent study of 1,100 winter wheat lines from different found in the 3_SNPs haplotype result in reduced binding of regions of the world (with emphasis in European varieties) the GRP2 protein to the RIP3 site in the VRN-A1 first intron found a single VRN-A1 copy in only 7% of the varieties (Kippes et al. 2015). GRP2 has been previously shown to (Würschum et al. 2015). This percentage is similar to the be a repressor of flowering that binds the pre-mVRN-A1 10.2% of varieties with a single VRN-A1 copy found in this 1 3 Molecular Genetics and Genomics study. Unfortunately, Würschum et al. (2015) did not have and within the range of the accession with multiple VRN- information about the RIP3 haplotypes. A1 copies and the 1_SNP haplotype. The late flowering of Since 92.3% of the accessions with a single VRN-A1 PI 627798 is also consistent with the late flowering of the copy in our winter panel have the 3_SNPs haplotype, we plants carrying the 1_SNP haplotype in our F population, will assume for the following discussion that in the win- which suggests that the RIP3 haplotypes correlate better ter panel from Würschum et al. (2015) most of the acces- with heading time in winter wheat than the differences in sions carrying a single VRN-A1 copy also carry the 3_SNPs VRN-A1 copy number. allele. Würschum et al. (2015) found that the frequency of the varieties with a single VRN-A1 copy was larger in South- VRN‑A1 copy number variants within the 1_ ern Europe and the UK where the winters are milder (Wür- SNP haplotype showed limited association schum et al. 2015). This distribution is consistent with the with differences in heading time milder vernalization requirement of the varieties carrying the 3_SNPs haplotype. We present the geographical origins Once we removed the effect of the 3_SNPs haplotype, we of 12 accessions with 3_SNPs haplotypes detected in this did not detect differences in heading time among the win - study in Supplemental Table S4, but due to the small sample ter wheat varieties with two, three or four VRN-A1 copies size, it is difficult to draw any solid conclusion. in any of the field or controlled environment experiments The only accession in this survey with one VRN-A1 copy with partial vernalization (Fig. 5). Even the single variety and the 1_SNP haplotype (PI 627798) was the latest or sec- found with one VRN-A1 copy with the 1_SNP haplotype ond latest flowering line when compared with the acces- (PI 627798) flowered within the range of the varieties with sions with a single VRN-A1 copy and the 3_SNPs haplotype, multiple VRN-A1 copies. times among groups were compared using a Tukey’s test. Different Fig. 5 Heading time differences among winter wheats with different letters above the box-plots indicate significant differences among VRN-A1 copy number. The relative copy number and VRN-A1 hap- lotypes are described in supplemental Table S4. Lines were grouped groups (P < 0.05) based on the estimated VRN-A1 copy number and average heading 1 3 Molecular Genetics and Genomics The study by Würschum et al. (2015) also failed to detect edu and kswheat.com). Although we do not know how much differences in heading time among varieties with two, three the 3_SNPs VRN-A1 haplotype contributed to Jagger suc- or four VRN-A1 copies. However, their heading time stud- cess, it would be interesting to characterize the presence ies were performed under field conditions in Germany of this allele in the multiple varieties derived from Jagger. (> 48°N), where winter conditions were likely sufficient to Additionally, it will be informative to monitor the changes satisfy completely the vernalization requirement. We can- in the 3_SNPs allele frequency as new varieties are released not rule out a role of VRN-A1 copy number under different in this region. conditions. In fact, in a double haploid population from the In summary, we have shown a significant effect of the cross between Malacca (two VRN-A1 copies) and Hereward RIP3 haplotypes on wheat heading times, both under con- (three VRN-A1 copies) subjected to partial vernalization, trolled environments with partial vernalization and in field Diaz et al. (2012) showed that the lines with three VRN-A1 experiments. Our results and those from Würschum et al. copies flowered later than those with two copies. Guedira (2015) suggest that one VRN-A1 copy with the 3_SNPs et al. (2016) observed that RILs from the AGS 2000 × 26R61 haplotype may have an adaptive value in regions with mild RIL population with two VRN-A1 copies flowered earlier winters. The confirmation of the role of the 3_SNPs allele than the lines with three copies when the population was on heading time and in the modulation of the vernaliza- grown in the field at locations with mild winters in the south- tion requirement can provide winter wheat breeders new eastern United States. It would be interesting to test if the genetic tools to improve wheat adaptation to new or chang- same effect can be detected in other biparental populations ing environments. segregating for VRN-A1 copy number variants. Funding This project was supported by the Agriculture and Food Although the frequencies of varieties with different VRN- Research Initiative Competitive Grants 2016-67013-24617 and 2017- A1 copy number alleles in different geographical regions 67007-25939 (WheatCAP) from the USDA National Institute of Food suggest a possible adaptive role, additional studies using and Agriculture and by the Howard Hughes Medical Institute. isogenic lines or biparental populations will be necessary to quantify better the effect of the number of VRN-A1 copies Compliance with ethical standards on the adaptation to different environments. Adaptation to these environments may depend not only on the effect of the Conflict of interest All authors declare that they have no conflicts of interest. different VRN-A1 copy number variants on heading time, but also in their interactions with FR-A2 alleles for frost toler- Ethical approval This article does not contain any studies with human ance. Zhu et al. (2014) showed that winter wheat varieties participants or animals performed by any of the authors. carrying three VRN-A1 copies were more frost tolerant than varieties with two VRN-A1 copies when the FROST TOLER- Open Access This article is distributed under the terms of the Crea- ANCE 2 allele T (FR-A2-T) was present. tive Commons Attribution 4.0 International License (http://creat iveco mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- tion, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Conclusions and practical applications Creative Commons license, and indicate if changes were made. It is still not clear if the relatively low frequency of the 3_ SNPs allele in the Western breeding programs is a result of its recent introduction or the effect of a narrow adap- References tive value, limited to a small range of environments. If the 3_SNPs allele introgression happened recently, it is possi- Balasubramanian S, Sureshkumar S, Agrawal M et al (2006) The PHY- TOCHROME C photoreceptor gene mediates natural variation in ble that this VRN-A1 allele has not reached yet its optimal flowering and growth responses of Arabidopsis thaliana. Nat frequency. It would be interesting to test the value of this Genet 38:711–715 allele in modern winter wheat breeding programs, particu- Bonman JM, Babiker EM, Cuesta-Marcos A, Esvelt-Klos K, Brown- larly in regions with mild winters. It is also possible, that as Guedira G, Chao S, See D, Chen J, Akhunov E, Zhang J, Bock- elman HE (2015) Genetic diversity among wheat accessions global temperatures increase and winters become milder, the from the USDA National Small Grains Collection. Crop Sci 3_SNPs allele will became more valuable in certain regions. 55:1243–1253 The presence of the 3_SNPs haplotype in the success- Brooking IR (1996) Temperature response of vernalization in wheat: A ful winter wheat variety Jagger may represent an example developmental analysis. Ann Bot London 78:507–512 Cang AF, Wilson AA, Wiens JJ (2016) Climate change is projected to of the potential of this allele. Jagger was grown in more outpace rates of niche change in grasses. Biol Lett 12:20160368 than 25% of the Oklahoma acreage for 11 years and was Chen A, Dubcovsky J (2012) Wheat TILLING mutants show that the the dominant wheat variety in Kansas between 1998 and vernalization gene VRN1 down-regulates the flowering repressor 2010 (USDA/NASS Oklahoma Field Office wheat.okstate. 1 3 Molecular Genetics and Genomics VRN2 in leaves but is not essential for flowering. PLoS Genet Konopatskaia I, Vavilova V, Kondratenko EY, Blinov A, Goncharov NP 8:e1003134 (2016) VRN1 genes variability in tetraploid wheat species with a Chen A, Li C, Hu W, Lau M, Lin H, Rockwell NC, Martin SS, Jernstedt spring growth habit. BMC Plant Biol S3:2 JA, Lagarias JC, Dubcovsky J (2014) PHYTOCHROME C plays a Kosner J, Pankova K (2002) Vernalisation response of some winter major role in the acceleration of wheat flowering under long days. wheat cultivars (Triticum aestivum L.). Czech J Genet Plant Breed Proc Natl Acad Sci USA 111:10037–10044 38:97–103 Chu CG, Tan CT, Yu GT, Zhong S, Xu SS, Yan L (2011) A novel Li C, Dubcovsky J (2008) Wheat FT protein regulates VRN1 transcrip- retrotransposon inserted in the dominant Vrn-B1 allele confers tion through interactions with FDL2. Plant J 55:543–554. https:// spring growth habit in tetraploid wheat (Triticum turgidum L.). doi.org/10.1111/j.1365-313X.2008.03526 .x G3-Genes Genom Genet 1:637–645 Li G, Yu M, Fang T, Cao S, Carver BF, Yan L (2013) Vernalization Cook BI, Wolkovich EM, Davies JT, Ault TR, Betancourt JL, Allen requirement duration in winter wheat is controlled by TaVRN-A1 JM, Bolmgren K, Cleland EE, Crimmins TM, Kraft NJB, Lan- at the protein level. Plant J 76:742–753 caster LT, Mazer SJ, McCabe GJ, McGill BJ, Parmesan C, Pau S, Li C, Lin H, Dubcovsky J (2015) Factorial combinations of protein Regetz J, Salamin N, Schwartz MD, Travers SE (2012) Sensitivity interactions generate a multiplicity of florigen activation com- of spring phenology to warming across temporal and spatial cli- plexes in wheat and barley. Plant J 84:70–82 mate gradients in two independent databases. Ecosystems 15:1283 Liu B, Asseng S, Müller C, Ewert F, Elliott J, Lobell D, Martre P, Corbesier L, Vincent C, Jang SH, Fornara F, Fan QZ, Searle I, Giak- Ruane A, Wallach D, Jones J, Rosenzweig C, Aggarwal P, Alder- ountis A, Farrona S, Gissot L, Turnbull C, Coupland G (2007) FT man P, Anothai J, Basso B, Biernath C, Cammarano D, Chal- protein movement contributes to long-distan ce signaling in floral linor A, Deryng D, Sanctis G, Doltra J, Fereres E, Folberth C, induction of Arabidopsis. Science 316:1030–1033 Garcia-Vila M, Gayler S, Hoogenboom G, Hunt L, Izaurralde R, de la Mata M, Alonso CR, Kadener S, Fededa JP, Blaustein M, Pelisch Jabloun M, Jones C, Kersebaum K, Kimball B, Koehler A, Kumar F, Cramer P, Bentley D, Kornblihtt AR (2003) A slow RNA poly- S, Nendel C, O’Leary G, Olesen J, Ottman M, Palosuo T, Prasad merase II affects alternative splicing in vivo. Mol Cell 12:525–532 P, Priesack E, Pugh T, Reynolds M, Rezaei E, Rötter R, Schmid Diaz A, Zikhali M, Turner AS, Isaac P, Laurie DA (2012) Copy num- E, Semenov M, Shcherbak I, Stehfest E, Stöckle C, Stratonovitch ber variation affecting the Photoperiod-B1 and Vernalization-A1 P, Streck T, Supit I, Tao F, Thorburn P, Waha K, Wall G, Wang genes is associated with altered flowering time in wheat (Triticum E, White J, Wolf J, Zhao Z, Zhu Y (2016) Similar estimates of aestivum). Plos One 7:e33234 temperature impacts on global wheat yield by three independent Distelfeld A, Li C, Dubcovsky J (2009a) Regulation of flowering in methods. Nat Clim Change 6:1130–1136 temperate cereals. Curr Opin Plant Biol 12:178–184 Loukoianov A, Yan L, Blechl A, Sanchez A, Dubcovsky J (2005) Reg- Distelfeld A, Tranquilli G, Li C, Yan L, Dubcovsky J (2009b) Genetic ulation of VRN-1 vernalization genes in normal and transgenic and molecular characterization of the VRN2 loci in tetraploid polyploid wheat. Plant Physiol 138:2364–2373 wheat. Plant Physiol 149:245–257 Muterko A, Balashova I, Cockram J, Kalendar R, Sivolap Y (2015) Dubcovsky J, Loukoianov A, Fu D, Valarik M, Sanchez A, Yan L The new wheat vernalization response allele Vrn-D1s is caused (2006) Effect of photoperiod on the regulation of wheat vernaliza- by DNA transposon insertion in the first intron. Plant Mol Biol tion genes VRN1 and VRN2. Plant Mol Biol 60:469–480 Rep 33:294–303 FAOSTAT (2015) Food and Agriculture Organization (FAO Statistical Muterko A, Kalendar R, Salina E (2016) Novel alleles of the VER- Pocketbook 2015, http://www.fao.org/3/a-i4691 e.pdf) NALIZATION1 genes in wheat are associated with modulation Franks SJ, Sim S, Weis AE (2007) Rapid evolution of flowering time of DNA curvature and flexibility in the promoter region. BMC by an annual plant in response to a climate fluctuation. Proc Natl Plant Biol 16:9 Acad Sci USA 104:1278–1282 Pidal B, Yan L, Fu D, Zhang F, Tranquilli G, Dubcovsky J (2009) The Fu D, Szucs P, Yan L, Helguera M, Skinner JS, von Zitzewitz J, Hayes CArG-box in the promoter region of wheat vernalization gene PM, Dubcovsky J (2005) Large deletions within the first intron VRN1 is not necessary to mediate the vernalization response. J in VRN-1 are associated with spring growth habit in barley and Hered 100:355–364 wheat. Mol Genet Genomics 273:54–65 Pugsley AT (1971) A genetic analysis of the spring-winter habit of Guedira M, Xiong M, Hao YF, Johnson J, Harrison S, Marshall D, growth in wheat. Aust J Agric Res 22:21–31 Brown-Guedira G, Zhang A (2016) Heading Date QTL in Winter Pugsley AT (1972) Additional genes inhibiting winter habit in wheat. Wheat (Triticum aestivum L.) Coincide with Major Developmen- Euphytica 21:547–552 tal Genes VERNALIZATION1 and PHOTOPERIOD1. PLoS One Saïdou AA, Clotault J, Couderc M, Mariac C, Devos KM, Thuillet AC, 11(5):e0154242 Amoukou IA, Vigouroux Y (2014) Association mapping, patterns Kane NA, Agharbaoui Z, Diallo AO, Adam H, Tominaga Y, Ouellet of linkage disequilibrium and selection in the vicinity of the PHY- F, Sarhan F (2007) TaVRT2 represses transcription of the wheat TOCHROME C gene in pearl millet. Theor Appl Genet 127:19–32 vernalization gene TaVRN1. Plant J 51:670–680 Tamaki S, Matsuo S, Wong HL, Yokoi S, Shimamoto K (2007) Kippes N, Zhu J, Chen A, Vanzetti L, Lukaszewski A, Nishida H, Hd3a protein is a mobile flowering signal in rice. Science Kato K, Dvorak J, Dubcovsky J (2014) Fine mapping and epistatic 316:1033–1036 interactions of the vernalization gene VRN-D4 in hexaploid wheat. Taoka K, Ohki I, Tsuji H, Furuita K, Hayashi K, Yanase T, Yamaguchi Mol Genet Genomics 289:47–62 M, Nakashima C, Purwestri YA, Tamaki S, Ogaki Y, Shimada Kippes N, Debernardi J, Vasquez-Gross HA, Akpinar BA, Budak H, C, Nakagawa A, Kojima C, Shimamoto K (2011) 14-3-3 pro- Kato K, Chao S, Akhunov E, Dubcovsky J (2015) Identification of teins act as intracellular receptors for rice Hd3a florigen. Nature the VERNALIZATION 4 gene reveals the origin of spring growth 476:332–335 habit in ancient wheats from South Asia. Proc Natl Acad Sci USA Trevaskis B, Bagnall DJ, Ellis MH, Peacock WJ, Dennis ES (2003) 112:E5401-E5410 MADS box genes control vernalization-induced flowering in cere- Kippes N, Chen A, Zhang X, Lukaszewski AJ, Dubcovsky J (2016) als. Proc Natl Acad Sci USA 100:13099–13104 Development and characterization of a spring hexaploid Woods D, Mckeown M, Dong Y, Preston JC, Amasino RM (2016) wheat line with no functional VRN2 genes. Theor Appl Genet Evolution of VRN2/GhD7-like genes in vernalization-mediated 129:1417–1428 repression of grass flowering. Plant Physiol 170:2124–2135 1 3 Molecular Genetics and Genomics Würschum T, Boeven PHG, Langer SM, Longin CFH, Leiser WL Yan L, Loukoianov A, Blechl A, Tranquilli G, Ramakrishna W, San- (2015) Multiply to conquer: Copy number variations at Ppd-B1 Miguel P, Bennetzen JL, Echenique V, Dubcovsky J (2004b) The and Vrn-A1 facilitate global adaptation in wheat. BMC Genet wheat VRN2 gene is a flowering repressor down-regulated by ver - 16:96 nalization. Science 303:1640–1644 Xiao J, Xu S, Li C, Xu Y, Xing L, Niu Y, Huan Q, Tang Y, Zhao C, Yan L, Fu D, Li C, Blechl A, Tranquilli G, Bonafede M, Sanchez A, Wagner D, Gao C, Chong K (2014) O-GlcNAc-mediated interac- Valarik M, Yasuda S, Dubcovsky J (2006) The wheat and barley tion between VER2 and TaGRP2 elicits TaVRN1 mRNA accumu- vernalization gene VRN3 is an orthologue of FT. Proc Natl Acad lation during vernalization in winter wheat. Nat Commun 5:4572 Sci USA 103:19581–19586 Yan L, Loukoianov A, Tranquilli G, Helguera M, Fahima T, Dubcovsky Zadoks J, Chang T, Konzak C (1974) A decimal code for the growth J (2003) Positional cloning of wheat vernalization gene VRN1. stages of cereals. Weed Res 14:415–421 Proc Natl Acad Sci USA 100:6263–6268 Zhu J, Pearce S, Burke A, See DR, Skinner DZ, Dubcovsky J, Camp- Yan L, Helguera M, Kato K, Fukuyama S, Sherman J, Dubcovsky J bell KG (2014) Copy number and haplotype variation at the VRN- (2004a) Allelic variation at the VRN-1 promoter region in poly- A1 and central FR-A2 loci are associated with frost tolerance in ploid wheat. Theor Appl Genet 109:1677–1686 hexaploid wheat. Theor Appl Genet 127:1183–1197 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Molecular Genetics and Genomics Springer Journals

Single nucleotide polymorphisms in a regulatory site of VRN-A1 first intron are associated with differences in vernalization requirement in winter wheat

Free
13 pages
Loading next page...
 
/lp/springer_journal/single-nucleotide-polymorphisms-in-a-regulatory-site-of-vrn-a1-first-VPypt0Pno0
Publisher
Springer Berlin Heidelberg
Copyright
Copyright © 2018 by The Author(s)
Subject
Life Sciences; Plant Genetics and Genomics; Human Genetics; Microbial Genetics and Genomics; Animal Genetics and Genomics; Biochemistry, general
ISSN
1617-4615
eISSN
1617-4623
D.O.I.
10.1007/s00438-018-1455-0
Publisher site
See Article on Publisher Site

Abstract

Winter wheats require a long exposure to cold temperatures (vernalization) to accelerate flowering. However, varieties dif- fer in the length of the period of cold required to saturate the vernalization response. Here we show that single nucleotide polymorphisms (SNP) at the binding site of the GRP2 protein in the VRN-A1 first intron (henceforth, RIP3) are associated with significant differences in heading time after a partial vernalization treatment. The ancestral winter VRN-A1 allele in ‘Triple Dirk C’ has one SNP in the RIP3 region (1_SNP) relative to the canonical RIP3 sequence, whereas the derived ‘Jagger’ allele has three SNPs (3_SNPs). Both varieties have a single VRN-A1 copy encoding identical proteins. In an F population generated from a cross between these two varieties, plants with the 3_SNPs haplotype headed significantly earlier (P < 0.001) than those with the 1_SNP haplotype, both in the absence of vernalization (17 days difference) and after 3-weeks of vernalization (11 days difference). Plants with the 3_SNPs haplotype showed higher VRN-A1 transcript levels than those with the 1_SNP haplotype. The 3_SNPs haplotype was also associated with early heading in a panel of 127 winter wheat varieties grown in three separate controlled-environment experiments under partial vernalization (36 to 54 days, P < 0.001) and one experiment under field conditions (21 d, P < 0.0001). The RIP3 polymorphisms can be used by wheat breeders to develop winter wheat varieties adapted to regions with different duration or intensity of the cold season. Keywords Wheat · Flowering · Vernalization · VRN1 · RIP3 · GRP2 Abbreviations FT1 FLOWERING LOCUS T1 GRP2 GLYCINE RICH RNA-BINDING PROTEIN 2 RIP3 RNA immune precipitation fragment 3 Communicated by S. Hohmann. SNP Single nucleotide polymorphism VRN1 VERNALIZATION1 gene Electronic supplementary material The online version of this article (https ://doi.org/10.1007/s0043 8-018-1455-0) contains VRN2 VERNALIZATION2 gene supplementary material, which is available to authorized users. * Jorge Dubcovsky Department of Plant Sciences, University of California, jdubcovsky@ucdavis.edu Davis, CA 95616-8515, USA Nestor Kippes Present Address: Department of Plant Biology and Genome nfkippes@ucdavis.edu Center, University of California, Davis, CA 95616, USA Mohammed Guedira Department of Crop Science, North Carolina State mguedir@ncsu.edu University, Raleigh, NC 27695, USA Lijuan Lin USDA-ARS Plant Science Research Unit, Raleigh, lljlin@ucdavis.edu NC 27695, USA Maria A. Alvarez Howard Hughes Medical Institute, Chevy Chase, MD 20815, malvarez@ucdavis.edu USA Gina L. Brown-Guedira gbrowng@ncsu.edu Vol.:(0123456789) 1 3 Molecular Genetics and Genomics VRN3 VERNALIZATION3 gene tetraploid and hexaploid wheat (e.g., by marker assisted VRN-D4 VERNALIZATION4 gene D genome selection) can generate spring wheats (Distelfeld et  al. 2009b; Kippes et al. 2016). During the fall, VRN2 represses the expression of VRN3 Introduction (= FT1), a homolog of the Arabidopsis flowering promoter FLOWERING LOCUS T (FT) (Yan et al. 2006). FT encodes Wheat is one of the most widely cultivated cereals and an a mobile protein that travels from the leaves to the apex important staple food worldwide. Almost 70% of 740M (Corbesier et al. 2007; Tamaki et al. 2007) where it forms tonnes of annual production are used for direct human con- a complex with FD-like and 14-3-3 proteins. This complex sumption making wheat yields crucial for human nutrition binds to the VRN1 promoter and induces flowering (Taoka and global food security (FAOSTAT 2015). Recent esti- et al. 2011; Li et al. 2015). During the winter, a slight induc- mates of the impact of climate change on crops yield have tion of VRN1 is sufficient to repress VRN2, which favors raised the interest in understanding how plant development FT1 up-regulation when days get longer during spring (Chen is modulated by environmental cues (Cang et al. 2016; and Dubcovsky 2012). These interactions result in a positive Cook et al. 2012; Franks et al. 2007; Liu et al. 2016). feedback-loop that irreversibly promotes wheat flowering in Wheat varieties are divided into two major categories the spring (Loukoianov et al. 2005). (winter and spring) based on their growth habit. Winter The natural allelic variation in VRN-A1 responsible for wheats require a long exposure to low temperatures (vernali- differences between spring and winter wheats is well char - zation) to accelerate the transition from the vegetative to the acterized. However, less is known about the role of VRN- reproductive phase. This requirement prevents exposure of A1 on the differences among winter wheats in the duration delicate floral meristems to damaging freezing temperatures of the cold period required for saturating the vernalization (Distelfeld et al. 2009a; Woods et al. 2016). By contrast, response. Three independent studies found that a large pro- spring wheats have reduced or no vernalization requirement. portion of the variation in vernalization requirement among Winter wheats are sown in fall in regions where wheat can winter wheat varieties is linked to the VRN-A1 locus. How- tolerate winter freezing temperatures. However, where win- ever, the three studies propose alternative explanations for ter temperatures are too low, spring wheats are sown in the these differences. Diaz et al. (2012) suggested that the dif- spring to avoid freezing. In Mediterranean regions, with ferences were caused by the presence of a single copy of mild and rainy winters, spring wheats are planted in the fall VRN-A1 in ‘Claire’ and three in ‘Hereward’. Li et al. (2013) to take advantage of water availability during the winter. argued that the differences in heading time were caused Wheat varieties sown in these contrasting regions show by an amino acid polymorphism at position 180 between different allele profiles at the main vernalization gene VER- ‘Jagger’ (alanine) and ‘2174’ (valine). More recently, Kip- NALIZATION1 (VRN1), a MADS-box transcription factor pes et al. (2015) found that both ‘Claire’ and ‘Jagger’ also homologous to the meristem identity gene APETALA 1 differed from ‘Hereward’ and ‘2174’ by single nucleotide (AP1) in Arabidopsis (Yan et  al. 2003; Trevaskis et  al. polymorphisms (SNPs) at the binding site of a GLYCINE 2003). Winter wheats carry the ancestral VRN1 alleles, RICH RNA-BINDING PROTEIN 2 (GRP2) in a region of whereas spring wheats show deletions or mutations in the VRN-A1 first intron designated as RNA Immune Precipi- regulatory regions located in the promoter or first intron. tation fragment 3 (RIP3, Xiao et al. 2014 and; Kippes et al. Changes in these regions have been observed in all three 2015). RIP3 natural polymorphisms were shown to disrupt VRN1 homologs (VRN-A1, VRN-B1 and VRN-D1), as well the binding of GRP2 to the RIP3 site (Kippes et al. 2015), as in the relatively recent duplicated paralog VRN-D4 (Fu but their effects on VRN1 expression and heading time were et al. 2005; Konopatskaia et al. 2016; Muterko et al. 2015, not characterized. No variation in the RIP3 site was detected 2016; Yan et al. 2004a, 2003; Kippes et al. 2015; Chu et al. in the VRN-B1 or VRN-D1 genes (Kippes et al. 2015). 2011; Pidal et al. 2009). In this study, we show that polymorphisms in the VRN-A1 Additional differences in wheat vernalization require- RIP3 region segregating in an F population are associated ment have been detected at the VERNALIZATION2 (VRN2) with differences in VRN-A1 transcript levels and heading and VERNALIZATION3 (VRN3) loci. VRN2 encodes a pro- time independently of differences in VRN-A1 copy number tein with a putative zinc finger and a CCT domain (ZCCT) and protein sequence. We also show that polymorphisms in that acts as a dominant long-day repressor of VRN3 and the RIP3 region in a panel of winter wheat varieties from dif- flowering. VRN2 deletions or loss-of-function mutations ferent geographic regions are associated with differences in result in spring growth habit in diploid wheat and barley heading time. The distribution of RIP3 alleles and VRN-A1 accessions (Yan et al. 2004b), but have not been observed copy number among winter wheat varieties and the utiliza- so far in natural polyploid wheat species. Combination of tion of these variants in winter wheat breeding are discussed. deletions or non-functional vrn2 alleles at all homologs in 1 3 Molecular Genetics and Genomics Allele Specific PCR (KASP) fluorescent assays were used Materials and methods to detect polymorphism among the 127 accessions at wheat vernalization loci. PCR was run according to manufacturer’s Plant material instructions, using a reaction volume of 4 µL, which con- sisted of 2 µL 2× KASPar reaction mix, 0.05 µL 72× assay The parental lines of the F population analyzed in this −1 mix, and 2 µL of template DNA (10 ng µL ). Endpoint study were the winter wheat lines Triple Dirk C (TDC) and genotyping was conducted using the software KlusterCaller CS5402 (Kippes et al. 2014, 2015). TDC is part of a set of (LGC Genomics, Hoddeson, UK). near isogenic lines for vernalization alleles described in Pug- Assays developed from published sequences of causal sley 1971 and 1972. CS5402 is a substitution line of chromo- genes were used to distinguish lines in the core collection some 5D of Chinese Spring (dominant Vrn-D1 allele) by the possessing spring alleles at Vrn-A1, Vrn-B1, Vrn-D1, and Ae. tauschii chromosome 5D (recessive vrn-D1 allele). Both Vrn-B3 loci (Supplemental Table 1). The exception was the lines have a single copy of the VRN-A1 gene homozygous KASP assay wMAS000033 used for detection of the Vrn- for the recessive vrn-A1 allele, which encodes for identical A1a spring allele developed from the contextual sequences VRN-A1 proteins (alanine residue at position 180, Li et al. of iSelect SNP marker IWA0001 determined to be associated 2013). The two accessions differ in the sequence of the RIP3 with Vrn-A1a. Accessions carrying alleles for spring growth region in intron 1. Relative to the RIP3 canonical sequence, habit at any VRN locus were excluded from further analysis. the VRN-A1 haplotype from TDC has one SNP (henceforth The VRN-A1 RIP3 region was sequenced in the 127 win- 1_SNP, GenBank AY747600) and the one from CS5402 ter accessions by Sanger sequencing using primers listed in three SNPs (henceforth 3_SNPs, GenBank KR422423, Kip- Table 1. Genotyping of the RIP3 alleles in the F population pes et al. 2015). was conducted with a KASP assay (Table 1) run on a 7500 For the association study, we used a panel of 127 win- Fast Real-Time PCR system (Applied Biosystems). VRN-A1 ter lines from the USDA National Germplasm Collection copy number was determined using a TaqMan assay from from diverse geographical origin. Sixty-five percent of the four biological replications per genotype as described in lines were collected from 31 different countries in Europe, Kippes et al. (2015). Differences in flowering time between 15% from 9 countries in Asia, 18% from 4 countries in the VRN-A1 copy number classes were analyzed using Tukey’s Americas, and 1% from Australia. These 127 accessions test (P < 0.05). The VRN-A1 promoter from Triple Dirk C were selected from a larger set after filtering out accessions (TDC) was sequenced by Sanger sequencing (GenBank carrying alleles associated with a spring growth habit using MH347747) using four overlapping PCR products ampli- available molecular markers (Yan et al. 2004b, 2006; Fu fied with primers listed in Table S3. et al. 2005). Principal component analysis Molecular markers and VRN1 polymorphisms Accessions from the winter panel were previously genotyped Genomic DNA was isolated from individual plants for the F with the Illumina Infinium 9K iSelect platform as part of study and from bulks of leaves from four seedlings of each the characterization of the NSGC core collection of winter accession for the winter wheat panel (MAG-Bind Plant DNA and facultative common wheat (Bonman et al. 2015). Geno- Plus 96 kit, Omega Bio-Tek, Norcross, GA). Kompetitive types of 4,483 markers for all accessions were obtained from Table 1 List of primers used Use Name Primer sequence (5′–3′) for RIP3 characterization and VRN-A1 transcripts abundance RIP3 A genome sequencing RIP-3_F1AAT CAC ACC TCA GGA TTT CAT by RT-qPCR RIP-3_R1GAT GGG TCA TAA GGT TTT GC RIP3 A genome Kasp assay 1_SNP_FTCT CAC AGT CAT TGT TGT TGG TAT G 3_SNPs_FTCT CAC AGT CAT TGT TGT TGG TAT C ReverseAGC AAT CAA GTT GTA ACA TAA ATA ATTA VRN1 RT-qPCR full tran- q.VRNA1-L-F1TCC ACC GAG TCA TGT ATG GA script q.VRNA1-L-R1GAG AAC CTT TTC TGC ATA AGAA VRN1 RT-qPCR short q.VRNA1-S-F1CAC CAA GGG AAA GCT CTA CG transcript q.VRNA1-S-R1GTT AAC TTG TAA CTG GGA GCTAA KASP assay primers do not include tail sequences 1 3 Molecular Genetics and Genomics The Triticeae Toolbox (https ://triti ceaet oolbo x.org/wheat ). photoperiod. Heading date was noted as the number of days Principal component analysis was conducted in Tassel v5.0 after transplanting when the spike fully emerged from the (http://www.maize genet ics.net/tasse l) using the covariance boot (Zadoks 60, Zadoks et al. 1974). The experiment was matrix for markers having minor allele frequency greater ended 175 days after planting. than or equal to 0.05. Lines missing more than 10% of data were removed. Greenhouse After the vernalization treatment described above, plants RT‑qPCR were grown at greenhouse facilities at UCD (four replica- tions) and NCSU (up to six replications) in a completely The last expanded leaf of four biological replications per randomized design. Supplemental lighting was used to pro- genotype was collected in liquid nitrogen, and RNA was vide 16-h light/8-h dark photoperiod. During the growth extracted using the Spectrum Plant Total RNA Kit (Sigma- period, the minimum temperatures during the night varied Aldrich). RNA samples were treated with DNase I (RQ1 from 10 to 18 °C and the maximum temperatures during the RNase-Free DNase, Promega) and first strand cDNAs were day ranged from 21 to 32 °C. The experiment was ended 210 synthesized from 1 µg of total RNA using the High Capacity days after planting. Heading date is reported as the number Reverse Transcription Kit (Applied Biosystems). Quantita- of days after transplanting when the spike fully emerged tive RT-PCR was performed using SYBR Green and a 7500 from the boot (Zadoks 60, Zadoks et al. 1974). Fast Real-Time PCR system (Applied Biosystems). ACTIN was used as an endogenous control. Primers are detailed Field in Table 1; primers for ACTIN were previously described (Dubcovsky et al. 2006). Transcript levels were expressed Three grams of seed of each entry were sown in a single as linearized fold-ACTIN levels calculated by the formula (ACTIN CT − TARGET CT) 1.5  m row at the NCSU Research Farm at Raleigh, NC 2 ± standard error (SE) of the mean. (35.73°N, 78.68°W; elevation: 116.5  m) on October 17, The resulting number indicates the ratio between the initial 2015. Heading date was recorded as the number days after number of molecules of the target gene and the number of January 1st, when 50% of the plants in a plot were at Zadoks molecules of ACTIN. 60 growth stage. Phenotyping Results Vernalization The RIP3 region at the first intron of VRN‑A1 Seeds were germinated in 2 × 2 plastic inch pots filled with is associated with differences in flowering time vermiculite in a greenhouse with 16-h light and 8-h dark photoperiod. One week after planting, seedlings were moved To determine the effect of the two VRN-A1 RIP3 natural to a growth chamber for 3 weeks of vernalization at 4 °C alleles on heading time, we intercrossed TDC (1_SNP allele) under 16-h light/8-h dark photoperiod. Similar conditions with CS5402 (3_SNPs allele). These two lines have a single were used in the facilities at the University of California, VRN-A1 copy encoding identical proteins. We generated an Davis (UCD) and North Carolina State University (NCSU). F population of 142 plants segregating for these haplotypes, After vernalization, up to 10 seedlings were transplanted genotyped them for the RIP3 alleles and recorded heading into 0.7 L black plastic containers (Stuewe and Sons, Tan- time under three different vernalization treatments (no ver - gent, OR) with 1:1 Metro Mix 2 and soil. Two grams of nalization, 3 weeks vernalization and 7 weeks vernalization). slow release fertilizer (Multicote 14–14–16, N-P-K) was In the absence of vernalization, the parental line incorporated with the soil in each cone. Cones were placed CS5402 headed 30.2 days earlier than TDC, whereas F in greenhouses or growth chambers under the conditions plants homozygous for the 3_SNPs haplotype flowered described below. 16.9 days earlier than plants homozygous for the 1_SNP haplotype (P < 0.0001, Fig. 1a). When plants were vernal- Growth chamber ized for 3 weeks, the difference in heading time between parental lines was reduced to 12 days (TDC later than After vernalization, four replications of each entry were CS5402) and those between F homozygous plants to grown in a controlled-environment chamber at NCSU in a 11.3 days (1_SNP haplotype later than 3_SNPs haplo- completely randomized design. Plants were grown under type, P < 0.0001, Fig.  1b). After 7 weeks of vernaliza- 20/18  °C day/night temperature and 16-h light/8-h dark tion, the difference in heading time between the parental 1 3 Molecular Genetics and Genomics Flowering time differences are linked to differences in VRN‑A1 expression levels To study the effects of RIP3 polymorphisms on VRN-A1 transcript levels, we sampled plants from the F popula- tion homozygous for the 1_SNP and 3_SNPs haplotypes at different time points of the partial vernalization treat- ment. Leaf samples were collected from 3-week-old plants immediately before vernalization (3w), 48  h after the plants were transfer to 4 °C (48 h), after 1 and 3 weeks of vernalization (1wV and 3wV) and 3 weeks after the plants were returned to room temperature (3wR). Since two alternative splice variants of VRN-A1 were detected in the study of Xiao et al. (2014), we designed specific primers to amplify each variant separately. The long VRN-A1 transcript variant (henceforth, long vari- ant) encodes the complete gene, whereas the short VRN- A1 transcript variant (600 bp, henceforth, short variant) includes the complete first exon (185 bp) and a small por - tion of the first intron. The short transcript ends a few base pairs downstream of the RIP3 region, located 2,767 bp downstream of the VRN-A1 start codon in TDC (Fig. 2a). Figure 2b presents the transcript levels of the long variant, Fig. 2c the short variant, and Fig. 2d the ratio between the two splice variants. For the VRN-A1 long variant, the transcript levels of the F plants carrying the 3_SNPs haplotype were 2- to 13-fold larger than those of plants carrying the 1_SNP haplotype. However, the differences were not significant at any of the time points or in the combined repeated meas- urements ANOVA (P = 0.18, Fig. 2b). For the VRN-A1 short variant, transcript levels of the F plants carrying the 3_SNPs haplotype were significantly higher than those of plants carrying the 1_SNP haplotype in the combined repeated measures ANOVA (P = 0.0003). However, for the individual time points the difference between haplotypes was significant only at 3 weeks after vernalization (P = 0.0177, Fig. 2c 3wV). The ratio between the short and long VRN-A1 variants showed a decrease during and after vernalization (Fig. 2d). This was the result of faster increases of the long variant Fig. 1 Difference in heading time between TDC (1_SNP) and relative to the short variant (Fig. 2d). After vernalization, CS5402 (3_SNPs) and between F plants homozygous for the RIP3 when the plants were returned to room temperature, the haplotypes. Data represents heading time means of at least eight plants per genotype under three different treatments (no vernalization short/long variant ratio was from 40- to 60-fold smaller and 3 or 7 weeks of vernalization). Different letters indicate signifi- than the same ratio before vernalization (Fig.  2d), sug- cant differences (Tukey’s test P < 0.05) gesting a decreasing importance of the short variant at this time point. Before vernalization, the ratio between lines was reduced to only 2 days (TDC later than CS5402, the short and long VRN-A1 variants was three times higher P = 0.0128) and those between homozygous F plants for in the plants carrying the 1_SNP RIP3 haplotype than in the two RIP3 homozygous plants were no longer signifi- those carrying the 3_SNPs haplotype (Fig. 2d, P = 0.0462). cant (0.9 d, P = 0.0578). These results show that there are No significant differences between haplotypes were significant differences in heading time linked to polymor - detected for the other time points (Fig. 2d). phisms at the RIP3 intronic region. 1 3 Molecular Genetics and Genomics Fig. 2 Transcript levels of VRN- A1 alternative splice variants during vernalization. Transcript levels of VRN-A1 alternative splice variants were studied in F plants homozygous for the two RIP3 haplotypes. Leaf samples were collected from 3-week-old plants immediately before vernalization (3w), 48 h after the plants were transfer to 4 °C (48 h), after 1 and 3 weeks of vernalization (1 and 3wV) and 3 weeks after the plants were returned to room temperature (3wR). (A) Schematic representation of the different VRNA1 transcripts studied. Arrowheads indicate regions complementary to the qRT-PCR primers utilized. The VRN-A1 alternative splice variant includes the RIP3 region (red). VRNA1-long correspond to the complete gene (B) and VRNA1-short to the alterna- tive splice variant (C). Average ratios of short/long transcript versions are presented in D. Bars represent means of four biological replications and error bars correspond to SEM (* P < 0.05) Fig. 3 Principal component analysis (PCA) of the winter panel used cles (1_SNP) or open diamonds (3_SNPs). a First (PC1) and second in this study. The PCA was based on 4483 SNP markers. Colors indi- (PC2) principal components. b First (PC1) and third (PC3) principal cate geographical origin and RIP3 genotypes are indicated by cir- components their effect in a panel of 127 winter lines from diverse Frequency and effect of the RIP3 haplotypes on heading time in a winter wheat panel geographical origins. The 1_SNP haplotype was found in 90.5% of the accessions whereas the 3_SNPs haplotype Based on the observed effect of the RIP3 haplotypes on was found in only 9.5% of the accessions. A principal component analysis showed that lines with the 3_SNPs heading time in the F population, we decided to study 1 3 Molecular Genetics and Genomics haplotype tend to cluster together in two separate groups Table 2 ANOVA for heading time of winter wheat lines carrying dif- ferent RIP3 haplotypes of Asian or European origin (Fig.  3, Supplemental a a b c Table S2). GH NC GH UCDGrowth chamber Field To study the effect of the RIP3 haplotypes on heading 3 SNPs 106.1 ± 2.3 104.3 ± 3.2 105.8 ± 2.7 104.8 ± 1.1 time the winter wheat panel was grown in three independent 1 SNP 161.0 ± 6.5 150.1 ± 4.4 142.6 ± 7.6 126.4 ± 4.5 experiments with partial vernalization conditions and one Dif. (days) 54.9 45.8 36.8 21.6 field experiment. The first experiment was conducted in a P value 1.11E-11 3.31E-06 1.01E-05 1.95E-08 grown chamber, where plants were first exposed to 3 weeks of vernalization and then moved to room temperature condi- GH = Greenhouse after 3-weeks vernalization tions. In two additional experiments performed at the Uni- Growth chamber after 3-week vernalization versity of California Davis (UCD) and North Carolina State Field conditions after natural vernalization (Raleigh, NC) University (NCSU), plants were transferred to greenhouses after 3 weeks of vernalization. The final experiment was experiment, plants with the 3_SNPs haplotype flowered conducted under field conditions in Raleigh, North Caro- lina. In the three experiments grown under controlled-envi- 21 days earlier than the plants with the 1_SNP haplotype (Fig. 4d; Table 2). ronments, plants carrying the 3_SNPs haplotype flowered significantly earlier (36–54 days, P < 0.0001) than plants with the 1_SNP haplotype (Fig. 4a–c; Table 2). In the field Fig. 4 Heading time differences between RIP3 haplotypes in a win- ferred to greenhouses at UCD and NCSU after 3-weeks vernalization. ter wheat panel. Independent experiments were conducted under a–c c Plants were vernalized in a growth chamber at 4  °C for 3 weeks control environmental conditions (3-week vernalization, or d) under and then temperature settings were switched to warm conditions natural vernalization in the field (Raleigh, NC). a, b Plants trans- (20/18 °C day/night) until heading 1 3 Molecular Genetics and Genomics other linked SNP or genes affecting the trait. This is particu- Variation in VRN‑A1 copy number larly critical for the VRN-A1 locus that is tightly linked with the PHYC gene (0.02 cM, Yan et al. 2003). PHYC mutants To study the effect of VRN-A1 copy number variation on heading time we characterized all lines in the winter wheat affect heading time in wheat (Chen et al. 2014) and natural variation in this gene is associated with variation in flower - panel with a TaqMan assay developed by Diaz et al. (2012). VRN-A1 copy number varied from one (10.2% of the lines) ing time in Arabidopsis (Balasubramanian et al. 2006) and pearl millet (Saïdou et al. 2014). Fortunately, Li et al. (2013) to four copies (3.9% of the lines), with most lines carrying two (35.4%) or three copies (50.4%). Most of the winter were able to find recombination events between VRN-A1 and PHYC that demonstrated that the differences in vernalization lines with a single VRN-A1 copy have the RIP3 3_SNPs haplotype. The only exception was PI 627798, which has requirement between their parental lines ‘Jagger’ and ‘2174’ were linked to VRN-A1 and not to PHYC. a single VRN-A1 copy and the 1_SNP haplotype (same as TDC). PI 627798 heading time was more similar to plants with the 1_SNP haplotype and multiple VRN-A1 copies, than to plants with the 3_SNPs haplotype and a single VRN-A1 Different polymorphisms in VRN‑A1 may contribute to flowering time differences copy. Compared with the plants in the last class, PI 627798 was the latest flowering in two experiments (Fig.  4a, d) and Although there is agreement on the contribution of VRN-A1 the second latest in the remaining one (Fig. 4c, not tested in b). These results are consistent with the late flowering of the to the differences in vernalization requirement among winter wheats, there is no agreement on the current interpretation plants carrying the 1_SNP haplotype in the F population. We detected no significant differences in heading time of the causal polymorphisms. Diaz et al. (2012) suggested that the differences in heading time linked to VRN-A1 were among varieties carrying two, three or four copies of VRN- A1 in any of the three pairwise comparisons. Taken together, caused by die ff rences in VRN-A1 copy number among Claire (1 copy), Malacca (2 copies) and Hereward (3 copies). these results suggest that VRN-A1 copy number variation has limited effect on heading time under the conditions used in However, the F population they generated from the cross between Claire and Hereward segregated also for the Ala/Val our four experiments. polymorphism at position 180 of the VRN-A1 protein and for the 3_SNPs/1_SNP polymorphism in the RIP3 site of the Discussion first intron, complicating the interpretation of these results. Diaz et al. (2012) also analyzed a double haploid popu- VRN‑A1 is linked to differences in vernalization lation from the cross between Malacca and Hereward, and found that the plants carrying three VRN-A1 copies tended to requirement in winter wheat head later than the plants carrying two VRN-A1 copies after 4 weeks of vernalization. In this population, both parental The length of the vernalization period required to saturate the acceleration of flowering varies widely among winter lines have the valine residue at position 180 and the 1_SNP haplotype at the RIP3 site, increasing the chances that the wheat varieties. Some varieties reach the saturation point after only 3 weeks of vernalization (sometimes called ‘fac- observed differences in heading time were caused by the dif- ferences in VRN-A1 copy number. Similarly, Guedira et al. ultative’ types), but most varieties require approximately 6 weeks of vernalization to reach this point. In some excep- (2016) observed in a RIL population from a cross between cultivars 26R61 and AGS 2000, both having the Val 180, tional cases, vernalization treatments of up to 8 weeks are necessary to saturate the acceleration of flowering (Brooking that after 2 or 4 weeks of vernalization lines having three VRN-A1 copies flowered later than those having two VRN- 1996; Kosner and Pankova 2002). The genetic factors controlling differences in vernali- A1 copies. However, it is not possible to rule out completely the effect of a linked gene given the small size of these seg- zation requirement among winter varieties are not as well understood as those controlling differences between winter regating populations. Li et  al. (2013) proposed that the early heading time and spring varieties (Distelfeld et al. 2009a). However, stud- ies that genetically mapped genes controlling heading time observed after partial vernalization of the plants carrying the Jagger VRN-A1 allele (Ala 180) relative to 2174 (Val in winter wheat using partial vernalization treatments (3–4 weeks of cold treatment) found that at least part of these dif- 180) were caused by different amino acid residues at posi- tion 180. However, these two winter wheat varieties differed ferences were linked to the VRN-A1 locus (Diaz et al. 2012; Li et al. 2013). A similar conclusion is supported by this also in VRN-A1 copy number (Jagger one VRNA1 copy vs. 2174 two VRNA1 copies) and the RIP3 haplotype in the first study. One limitation of linkage studies using small segregating intron (Jagger 3_SNPs vs. 2174 1_SNP), complicating the interpretation of the results. populations is that they cannot rule out the possibility of 1 3 Molecular Genetics and Genomics The VRN-A1 allele in CS5402 is almost identical to the transcript (Xiao et al. 2014). During vernalization, GRP2 is alleles present in Jagger and Claire, all carrying a single O-GlcNAc modified and its levels in the nucleus decrease, VRN-A1 copy, the 3_SNPs RIP3 haplotype and Ala 180. allowing higher VRN1 mRNA accumulation (Xiao et al. Therefore, it is likely that the contrasting RIP3 haplotypes 2014). segregating in the Jagger (3_SNPs) × 2174 (1_SNP) (Li Differences in the speed and processivness of tran- et al. 2013) and Claire (3_SNPs) × Hereward (1_SNP) (Diaz scription affects the selection of alternative splicing sites et al. 2012) populations could have contributed to the differ - (de la Mata et al. 2003). Therefore, the stronger binding ences in heading time observed after partial vernalization in of the GRP2 protein to the translated RIP3 sites with the these studies. This conclusion, does not rule out the possibil- 1_SNP haplotype may favor the VRN-A1 short alterna- ity that the polymorphisms at position 180 or the differences tive splice variant, first described by Xiao et  al. (2014). in copy number could have also contributed to the observed We detected in silico a short alternative splice variant differences in heading time in the previous studies. for VRN-B1, which showed the same structure as the one The plants from the F population segregating for the described for VRN-A1 (http://plants. ensemb l.org, TGACv1, RIP3 haplotypes have a single VRN-A1 copy encoding iden- Traes_5BL_89636D032.1). However, we did not find this tical proteins, but differ in VRN-A1 transcript levels. There- short variant in the D genome of Chinese Spring, a result fore, polymorphisms at the VRN-A1 regulatory regions are that is consistent with a deletion encompassing the RIP3 site good candidates to explain the differences in heading time in the first intron of VRN-D1. linked to this gene. A comparison of the VRN-A1 promoter The binding of the GRP2 protein to the pre-mRNA RIP3 regions (2254 bp upstream from the start codon) from the site may explain the significantly higher short/long variant 1_SNP and 3_SNPs haplotypes revealed no-polymorphisms ratio observed before vernalization in the F plants carrying in the first 436 bp (Supplemental Figure S1). The rest of the 1_SNP haplotype relative to those carrying the 3_SNPs the promoter region (437–2254 bp) showed seven SNPs and haplotype in this study (Fig. 2d). The reduced GRP2 levels four indels (1–2 bp), but none of them were located within during and after vernalization would favor the long tran- known regulatory elements (Pidal et al. 2009; Kane et al. script variant and explain the decrease in the short/long 2007; Li and Dubcovsky 2008; Li et al. 2015), predicted variant ratio in these later time points. The higher relative binding sites of transcription factors, or evolutionary con- abundance of the short splice variant before vernalization served regions (Supplemental Figure S1). By contrast, the may contribute to maintain low levels of functional VRN-A1 RIP3 polymorphisms have been shown to affect the binding until the vernalization requirement is satisfied. Alternatively, of GRP2 proteins to the pre-mVRN-A1 transcripts (Kippes the 155 amino acids encoded by the short variant (includ- et al. 2015), a result consistent with the differences in rela- ing the MADS-box domain MADS_MEF2_like, cd00265) tive abundance of alternative splice variants described in the may interact with other MADS-Box proteins altering their following section. function. Transgenic experiments overexpressing this short variant will be required to test this hypothesis. Once the Differences in VRN‑A1 expression are consistent role of these alternative splice variants is better understood, with the proposed RIP3/GRP2 molecular it may provide an additional entry point to modulate the mechanisms vernalization requirement in wheat. Based on the previous results and discussion, we favor the Diaz et al. (2012) observed faster and higher VRN-A1 tran- hypothesis that the polymorphisms in the RIP3 region are script levels in Claire (3_SNPs haplotype) than in the two responsible for the differences in VRN-A1 transcript levels varieties carrying the 1_SNP haplotype. Re-analysis of and heading time between the lines carrying the 1_SNP and the expression data from Li et al. (2013) using a two-way 3_SNPs haplotypes. However, we cannot rule the possibility ANOVA with time (3 weeks and 6 weeks) and genotypes of effects caused by linked polymorphisms in the promoter (Jagger and 2174) as factors, revealed higher VRN-A1 region or outside the sequenced region. transcript levels in Jagger (3_SNPs) than in 2174 (1_SNP, P < 0.0001, Supplemental Table S2). These results are simi- lar to the ones presented here, and suggest that at least part Winter wheat varieties carrying the 3_SNPs of the differences in heading time between these two VRN- haplotype were detected at low frequencies A1 alleles are regulated at the transcriptional level. We have recently shown that the natural polymorphisms A recent study of 1,100 winter wheat lines from different found in the 3_SNPs haplotype result in reduced binding of regions of the world (with emphasis in European varieties) the GRP2 protein to the RIP3 site in the VRN-A1 first intron found a single VRN-A1 copy in only 7% of the varieties (Kippes et al. 2015). GRP2 has been previously shown to (Würschum et al. 2015). This percentage is similar to the be a repressor of flowering that binds the pre-mVRN-A1 10.2% of varieties with a single VRN-A1 copy found in this 1 3 Molecular Genetics and Genomics study. Unfortunately, Würschum et al. (2015) did not have and within the range of the accession with multiple VRN- information about the RIP3 haplotypes. A1 copies and the 1_SNP haplotype. The late flowering of Since 92.3% of the accessions with a single VRN-A1 PI 627798 is also consistent with the late flowering of the copy in our winter panel have the 3_SNPs haplotype, we plants carrying the 1_SNP haplotype in our F population, will assume for the following discussion that in the win- which suggests that the RIP3 haplotypes correlate better ter panel from Würschum et al. (2015) most of the acces- with heading time in winter wheat than the differences in sions carrying a single VRN-A1 copy also carry the 3_SNPs VRN-A1 copy number. allele. Würschum et al. (2015) found that the frequency of the varieties with a single VRN-A1 copy was larger in South- VRN‑A1 copy number variants within the 1_ ern Europe and the UK where the winters are milder (Wür- SNP haplotype showed limited association schum et al. 2015). This distribution is consistent with the with differences in heading time milder vernalization requirement of the varieties carrying the 3_SNPs haplotype. We present the geographical origins Once we removed the effect of the 3_SNPs haplotype, we of 12 accessions with 3_SNPs haplotypes detected in this did not detect differences in heading time among the win - study in Supplemental Table S4, but due to the small sample ter wheat varieties with two, three or four VRN-A1 copies size, it is difficult to draw any solid conclusion. in any of the field or controlled environment experiments The only accession in this survey with one VRN-A1 copy with partial vernalization (Fig. 5). Even the single variety and the 1_SNP haplotype (PI 627798) was the latest or sec- found with one VRN-A1 copy with the 1_SNP haplotype ond latest flowering line when compared with the acces- (PI 627798) flowered within the range of the varieties with sions with a single VRN-A1 copy and the 3_SNPs haplotype, multiple VRN-A1 copies. times among groups were compared using a Tukey’s test. Different Fig. 5 Heading time differences among winter wheats with different letters above the box-plots indicate significant differences among VRN-A1 copy number. The relative copy number and VRN-A1 hap- lotypes are described in supplemental Table S4. Lines were grouped groups (P < 0.05) based on the estimated VRN-A1 copy number and average heading 1 3 Molecular Genetics and Genomics The study by Würschum et al. (2015) also failed to detect edu and kswheat.com). Although we do not know how much differences in heading time among varieties with two, three the 3_SNPs VRN-A1 haplotype contributed to Jagger suc- or four VRN-A1 copies. However, their heading time stud- cess, it would be interesting to characterize the presence ies were performed under field conditions in Germany of this allele in the multiple varieties derived from Jagger. (> 48°N), where winter conditions were likely sufficient to Additionally, it will be informative to monitor the changes satisfy completely the vernalization requirement. We can- in the 3_SNPs allele frequency as new varieties are released not rule out a role of VRN-A1 copy number under different in this region. conditions. In fact, in a double haploid population from the In summary, we have shown a significant effect of the cross between Malacca (two VRN-A1 copies) and Hereward RIP3 haplotypes on wheat heading times, both under con- (three VRN-A1 copies) subjected to partial vernalization, trolled environments with partial vernalization and in field Diaz et al. (2012) showed that the lines with three VRN-A1 experiments. Our results and those from Würschum et al. copies flowered later than those with two copies. Guedira (2015) suggest that one VRN-A1 copy with the 3_SNPs et al. (2016) observed that RILs from the AGS 2000 × 26R61 haplotype may have an adaptive value in regions with mild RIL population with two VRN-A1 copies flowered earlier winters. The confirmation of the role of the 3_SNPs allele than the lines with three copies when the population was on heading time and in the modulation of the vernaliza- grown in the field at locations with mild winters in the south- tion requirement can provide winter wheat breeders new eastern United States. It would be interesting to test if the genetic tools to improve wheat adaptation to new or chang- same effect can be detected in other biparental populations ing environments. segregating for VRN-A1 copy number variants. Funding This project was supported by the Agriculture and Food Although the frequencies of varieties with different VRN- Research Initiative Competitive Grants 2016-67013-24617 and 2017- A1 copy number alleles in different geographical regions 67007-25939 (WheatCAP) from the USDA National Institute of Food suggest a possible adaptive role, additional studies using and Agriculture and by the Howard Hughes Medical Institute. isogenic lines or biparental populations will be necessary to quantify better the effect of the number of VRN-A1 copies Compliance with ethical standards on the adaptation to different environments. Adaptation to these environments may depend not only on the effect of the Conflict of interest All authors declare that they have no conflicts of interest. different VRN-A1 copy number variants on heading time, but also in their interactions with FR-A2 alleles for frost toler- Ethical approval This article does not contain any studies with human ance. Zhu et al. (2014) showed that winter wheat varieties participants or animals performed by any of the authors. carrying three VRN-A1 copies were more frost tolerant than varieties with two VRN-A1 copies when the FROST TOLER- Open Access This article is distributed under the terms of the Crea- ANCE 2 allele T (FR-A2-T) was present. tive Commons Attribution 4.0 International License (http://creat iveco mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- tion, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Conclusions and practical applications Creative Commons license, and indicate if changes were made. It is still not clear if the relatively low frequency of the 3_ SNPs allele in the Western breeding programs is a result of its recent introduction or the effect of a narrow adap- References tive value, limited to a small range of environments. If the 3_SNPs allele introgression happened recently, it is possi- Balasubramanian S, Sureshkumar S, Agrawal M et al (2006) The PHY- TOCHROME C photoreceptor gene mediates natural variation in ble that this VRN-A1 allele has not reached yet its optimal flowering and growth responses of Arabidopsis thaliana. Nat frequency. It would be interesting to test the value of this Genet 38:711–715 allele in modern winter wheat breeding programs, particu- Bonman JM, Babiker EM, Cuesta-Marcos A, Esvelt-Klos K, Brown- larly in regions with mild winters. It is also possible, that as Guedira G, Chao S, See D, Chen J, Akhunov E, Zhang J, Bock- elman HE (2015) Genetic diversity among wheat accessions global temperatures increase and winters become milder, the from the USDA National Small Grains Collection. Crop Sci 3_SNPs allele will became more valuable in certain regions. 55:1243–1253 The presence of the 3_SNPs haplotype in the success- Brooking IR (1996) Temperature response of vernalization in wheat: A ful winter wheat variety Jagger may represent an example developmental analysis. Ann Bot London 78:507–512 Cang AF, Wilson AA, Wiens JJ (2016) Climate change is projected to of the potential of this allele. Jagger was grown in more outpace rates of niche change in grasses. Biol Lett 12:20160368 than 25% of the Oklahoma acreage for 11 years and was Chen A, Dubcovsky J (2012) Wheat TILLING mutants show that the the dominant wheat variety in Kansas between 1998 and vernalization gene VRN1 down-regulates the flowering repressor 2010 (USDA/NASS Oklahoma Field Office wheat.okstate. 1 3 Molecular Genetics and Genomics VRN2 in leaves but is not essential for flowering. PLoS Genet Konopatskaia I, Vavilova V, Kondratenko EY, Blinov A, Goncharov NP 8:e1003134 (2016) VRN1 genes variability in tetraploid wheat species with a Chen A, Li C, Hu W, Lau M, Lin H, Rockwell NC, Martin SS, Jernstedt spring growth habit. BMC Plant Biol S3:2 JA, Lagarias JC, Dubcovsky J (2014) PHYTOCHROME C plays a Kosner J, Pankova K (2002) Vernalisation response of some winter major role in the acceleration of wheat flowering under long days. wheat cultivars (Triticum aestivum L.). Czech J Genet Plant Breed Proc Natl Acad Sci USA 111:10037–10044 38:97–103 Chu CG, Tan CT, Yu GT, Zhong S, Xu SS, Yan L (2011) A novel Li C, Dubcovsky J (2008) Wheat FT protein regulates VRN1 transcrip- retrotransposon inserted in the dominant Vrn-B1 allele confers tion through interactions with FDL2. Plant J 55:543–554. https:// spring growth habit in tetraploid wheat (Triticum turgidum L.). doi.org/10.1111/j.1365-313X.2008.03526 .x G3-Genes Genom Genet 1:637–645 Li G, Yu M, Fang T, Cao S, Carver BF, Yan L (2013) Vernalization Cook BI, Wolkovich EM, Davies JT, Ault TR, Betancourt JL, Allen requirement duration in winter wheat is controlled by TaVRN-A1 JM, Bolmgren K, Cleland EE, Crimmins TM, Kraft NJB, Lan- at the protein level. Plant J 76:742–753 caster LT, Mazer SJ, McCabe GJ, McGill BJ, Parmesan C, Pau S, Li C, Lin H, Dubcovsky J (2015) Factorial combinations of protein Regetz J, Salamin N, Schwartz MD, Travers SE (2012) Sensitivity interactions generate a multiplicity of florigen activation com- of spring phenology to warming across temporal and spatial cli- plexes in wheat and barley. Plant J 84:70–82 mate gradients in two independent databases. Ecosystems 15:1283 Liu B, Asseng S, Müller C, Ewert F, Elliott J, Lobell D, Martre P, Corbesier L, Vincent C, Jang SH, Fornara F, Fan QZ, Searle I, Giak- Ruane A, Wallach D, Jones J, Rosenzweig C, Aggarwal P, Alder- ountis A, Farrona S, Gissot L, Turnbull C, Coupland G (2007) FT man P, Anothai J, Basso B, Biernath C, Cammarano D, Chal- protein movement contributes to long-distan ce signaling in floral linor A, Deryng D, Sanctis G, Doltra J, Fereres E, Folberth C, induction of Arabidopsis. Science 316:1030–1033 Garcia-Vila M, Gayler S, Hoogenboom G, Hunt L, Izaurralde R, de la Mata M, Alonso CR, Kadener S, Fededa JP, Blaustein M, Pelisch Jabloun M, Jones C, Kersebaum K, Kimball B, Koehler A, Kumar F, Cramer P, Bentley D, Kornblihtt AR (2003) A slow RNA poly- S, Nendel C, O’Leary G, Olesen J, Ottman M, Palosuo T, Prasad merase II affects alternative splicing in vivo. Mol Cell 12:525–532 P, Priesack E, Pugh T, Reynolds M, Rezaei E, Rötter R, Schmid Diaz A, Zikhali M, Turner AS, Isaac P, Laurie DA (2012) Copy num- E, Semenov M, Shcherbak I, Stehfest E, Stöckle C, Stratonovitch ber variation affecting the Photoperiod-B1 and Vernalization-A1 P, Streck T, Supit I, Tao F, Thorburn P, Waha K, Wall G, Wang genes is associated with altered flowering time in wheat (Triticum E, White J, Wolf J, Zhao Z, Zhu Y (2016) Similar estimates of aestivum). Plos One 7:e33234 temperature impacts on global wheat yield by three independent Distelfeld A, Li C, Dubcovsky J (2009a) Regulation of flowering in methods. Nat Clim Change 6:1130–1136 temperate cereals. Curr Opin Plant Biol 12:178–184 Loukoianov A, Yan L, Blechl A, Sanchez A, Dubcovsky J (2005) Reg- Distelfeld A, Tranquilli G, Li C, Yan L, Dubcovsky J (2009b) Genetic ulation of VRN-1 vernalization genes in normal and transgenic and molecular characterization of the VRN2 loci in tetraploid polyploid wheat. Plant Physiol 138:2364–2373 wheat. Plant Physiol 149:245–257 Muterko A, Balashova I, Cockram J, Kalendar R, Sivolap Y (2015) Dubcovsky J, Loukoianov A, Fu D, Valarik M, Sanchez A, Yan L The new wheat vernalization response allele Vrn-D1s is caused (2006) Effect of photoperiod on the regulation of wheat vernaliza- by DNA transposon insertion in the first intron. Plant Mol Biol tion genes VRN1 and VRN2. Plant Mol Biol 60:469–480 Rep 33:294–303 FAOSTAT (2015) Food and Agriculture Organization (FAO Statistical Muterko A, Kalendar R, Salina E (2016) Novel alleles of the VER- Pocketbook 2015, http://www.fao.org/3/a-i4691 e.pdf) NALIZATION1 genes in wheat are associated with modulation Franks SJ, Sim S, Weis AE (2007) Rapid evolution of flowering time of DNA curvature and flexibility in the promoter region. BMC by an annual plant in response to a climate fluctuation. Proc Natl Plant Biol 16:9 Acad Sci USA 104:1278–1282 Pidal B, Yan L, Fu D, Zhang F, Tranquilli G, Dubcovsky J (2009) The Fu D, Szucs P, Yan L, Helguera M, Skinner JS, von Zitzewitz J, Hayes CArG-box in the promoter region of wheat vernalization gene PM, Dubcovsky J (2005) Large deletions within the first intron VRN1 is not necessary to mediate the vernalization response. J in VRN-1 are associated with spring growth habit in barley and Hered 100:355–364 wheat. Mol Genet Genomics 273:54–65 Pugsley AT (1971) A genetic analysis of the spring-winter habit of Guedira M, Xiong M, Hao YF, Johnson J, Harrison S, Marshall D, growth in wheat. Aust J Agric Res 22:21–31 Brown-Guedira G, Zhang A (2016) Heading Date QTL in Winter Pugsley AT (1972) Additional genes inhibiting winter habit in wheat. Wheat (Triticum aestivum L.) Coincide with Major Developmen- Euphytica 21:547–552 tal Genes VERNALIZATION1 and PHOTOPERIOD1. PLoS One Saïdou AA, Clotault J, Couderc M, Mariac C, Devos KM, Thuillet AC, 11(5):e0154242 Amoukou IA, Vigouroux Y (2014) Association mapping, patterns Kane NA, Agharbaoui Z, Diallo AO, Adam H, Tominaga Y, Ouellet of linkage disequilibrium and selection in the vicinity of the PHY- F, Sarhan F (2007) TaVRT2 represses transcription of the wheat TOCHROME C gene in pearl millet. Theor Appl Genet 127:19–32 vernalization gene TaVRN1. Plant J 51:670–680 Tamaki S, Matsuo S, Wong HL, Yokoi S, Shimamoto K (2007) Kippes N, Zhu J, Chen A, Vanzetti L, Lukaszewski A, Nishida H, Hd3a protein is a mobile flowering signal in rice. Science Kato K, Dvorak J, Dubcovsky J (2014) Fine mapping and epistatic 316:1033–1036 interactions of the vernalization gene VRN-D4 in hexaploid wheat. Taoka K, Ohki I, Tsuji H, Furuita K, Hayashi K, Yanase T, Yamaguchi Mol Genet Genomics 289:47–62 M, Nakashima C, Purwestri YA, Tamaki S, Ogaki Y, Shimada Kippes N, Debernardi J, Vasquez-Gross HA, Akpinar BA, Budak H, C, Nakagawa A, Kojima C, Shimamoto K (2011) 14-3-3 pro- Kato K, Chao S, Akhunov E, Dubcovsky J (2015) Identification of teins act as intracellular receptors for rice Hd3a florigen. Nature the VERNALIZATION 4 gene reveals the origin of spring growth 476:332–335 habit in ancient wheats from South Asia. Proc Natl Acad Sci USA Trevaskis B, Bagnall DJ, Ellis MH, Peacock WJ, Dennis ES (2003) 112:E5401-E5410 MADS box genes control vernalization-induced flowering in cere- Kippes N, Chen A, Zhang X, Lukaszewski AJ, Dubcovsky J (2016) als. Proc Natl Acad Sci USA 100:13099–13104 Development and characterization of a spring hexaploid Woods D, Mckeown M, Dong Y, Preston JC, Amasino RM (2016) wheat line with no functional VRN2 genes. Theor Appl Genet Evolution of VRN2/GhD7-like genes in vernalization-mediated 129:1417–1428 repression of grass flowering. Plant Physiol 170:2124–2135 1 3 Molecular Genetics and Genomics Würschum T, Boeven PHG, Langer SM, Longin CFH, Leiser WL Yan L, Loukoianov A, Blechl A, Tranquilli G, Ramakrishna W, San- (2015) Multiply to conquer: Copy number variations at Ppd-B1 Miguel P, Bennetzen JL, Echenique V, Dubcovsky J (2004b) The and Vrn-A1 facilitate global adaptation in wheat. BMC Genet wheat VRN2 gene is a flowering repressor down-regulated by ver - 16:96 nalization. Science 303:1640–1644 Xiao J, Xu S, Li C, Xu Y, Xing L, Niu Y, Huan Q, Tang Y, Zhao C, Yan L, Fu D, Li C, Blechl A, Tranquilli G, Bonafede M, Sanchez A, Wagner D, Gao C, Chong K (2014) O-GlcNAc-mediated interac- Valarik M, Yasuda S, Dubcovsky J (2006) The wheat and barley tion between VER2 and TaGRP2 elicits TaVRN1 mRNA accumu- vernalization gene VRN3 is an orthologue of FT. Proc Natl Acad lation during vernalization in winter wheat. Nat Commun 5:4572 Sci USA 103:19581–19586 Yan L, Loukoianov A, Tranquilli G, Helguera M, Fahima T, Dubcovsky Zadoks J, Chang T, Konzak C (1974) A decimal code for the growth J (2003) Positional cloning of wheat vernalization gene VRN1. stages of cereals. Weed Res 14:415–421 Proc Natl Acad Sci USA 100:6263–6268 Zhu J, Pearce S, Burke A, See DR, Skinner DZ, Dubcovsky J, Camp- Yan L, Helguera M, Kato K, Fukuyama S, Sherman J, Dubcovsky J bell KG (2014) Copy number and haplotype variation at the VRN- (2004a) Allelic variation at the VRN-1 promoter region in poly- A1 and central FR-A2 loci are associated with frost tolerance in ploid wheat. Theor Appl Genet 109:1677–1686 hexaploid wheat. Theor Appl Genet 127:1183–1197 1 3

Journal

Molecular Genetics and GenomicsSpringer Journals

Published: Jun 5, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off