Carrots require a certain number of cold hours to become vernalized and proceed to the reproductive stage, and this phenomenon is genotype-dependent. Annual carrots require less cold than biennials to flower; however, quantitative variation within annuals and biennials also exists, defining a gradient for vernalization requirement (VR). The flowering response of carrots to day length, after vernalization has occurred, is controversial. This vegetable has been described both as a long-day and a neutral-day species. The objective of this study was to evaluate flowering time and frequency in response to different cold treatments and photoperiod regimes in various carrot genotypes. To this end, three annual genotypes from India, Brazil, and Pakistan, and a biennial carrot from Japan, were exposed to 7.5 °C during 30, 60, 90, or 120 days, and then transferred to either long day (LD) or short day (SD) conditions. Significant variation (p < 0.05) among the carrot genotypes and among cold treatments were found, with increased flowering rates and earlier onset of flowering being associated with longer cold exposures. No significant differences in response to photoperiod were found, suggesting that post-vernalization day length does not influence carrot flowering. These findings will likely impact carrot breeding and production of both root and seed, helping in the selection of adequate genotypes and sowing dates to manage cold exposure and day-length for different production purposes.
Background: Carrot vernalization requirement (VR), the amount of cold exposure needed for plants to become vernalized, strongly influences geographical adaptation of the crop and many agricultural practices, and this trait defines annual vs. biennial cultivars. Despite its importance, the genetic and molecular mechanisms governing carrot VR remain elusive. Here, we integrated inheritance, linkage mapping, genome wide association studies (GWAS), phylogenetic analysis, and transcriptomics of coding and noncoding mRNA data to dissect the genetic control of carrot VR. Results: A simply-inherited dominant locus controlling VR, denominated Vrn1 , was identified and genetically mapped at the distal end of chromosome 2, in a highly-saturated genotyping-by-sequencing (GBS)-derived linkage map. GWAS results revealed a ~ 694 kb region with strong marker-trait associations that overlapped Vrn1 . In this region, we identified two duplicated SOC1 (SUPPRESSOR OF OVEREXPRESSION OF CONSTANS 1) genes, termed DcSOC1-1 and DcSOC1-2 , that were phylogenetically related to SOC1 genes from other species, and were strongly upregulated in vernalized plants relative to non-vernalized plants of annual and biennial carrots. Further, the identification and transcriptome analysis of carrot homologs to Arabidopsis vernalization and flowering genes revealed additional key genes from these pathways exhibiting correlated expression profiles with DcSOC1-1 and DcSOC1-2 , including other SOC1, FUL (FRUITFUL) and FLC (FLOWERING LOCUS C) genes. The expression profiles of DcSOC1-1 , DcSOC1-2 , and five other vernalization-related genes were validated by RT-qPCR analysis. No differentially expressed ncRNAs directly related to the Vrn1 candidates were identified. Conclusions: Overall, this study provides strong linkage-mapping and functional evidences for the candidacy of DcSOC1-1 and DcSOC1-2 as the primary regulators of VR in carrot.