Chrysanthemum MAF2 regulates flowering by repressing gibberellin biosynthesis in response to low temperature

Multi-locus genome-wide association study and genomic prediction for flowering time in chrysanthemum

MAIN CONCLUSION

Multi-locus GWAS detected several known and candidate genes responsible for flowering time in chrysanthemum. The associations could greatly increase the predictive ability of genome selection that accelerates the possible application of GS in chrysanthemum breeding. Timely flowering is critical for successful reproduction and determines the economic value for ornamental plants. To investigate the genetic architecture of flowering time in chrysanthemum, a multi-locus genome-wide association study (GWAS) was performed using a collection of 200 accessions and 330,710 single-nucleotide polymorphisms (SNPs) via 3VmrMLM method. Five flowering time traits including budding (FBD), visible colouring (VC), early opening (EO), full-bloom (OF) and senescing (SF) stages, plus five derived conditional traits were recorded in two environments. Extensive phenotypic variations were observed for these flowering time traits with coefficients of variation ranging from 6.42 to 38.27%, and their broad-sense heritability ranged from 71.47 to 96.78%. GWAS revealed 88 stable quantitative trait nucleotides (QTNs) and 93 QTN-by-environment interactions (QEIs) associated with flowering time traits, accounting for 0.50-8.01% and 0.30-10.42% of the phenotypic variation, respectively. Amongst the genes around these stable QTNs and QEIs, 21 and 10 were homologous to known flowering genes in Arabidopsis; 20 and 11 candidate genes were mined by combining the functional annotation and transcriptomics data, respectively, such as MYB55, FRIGIDA-like, WRKY75 and ANT. Furthermore, genomic selection (GS) was assessed using three models and seven unique marker datasets. We found the prediction accuracy (PA) using significant SNPs identified by GWAS under SVM model exhibited the best performance with PA ranging from 0.90 to 0.95. Our findings provide new insights into the dynamic genetic architecture of flowering time and the identified significant SNPs and candidate genes will accelerate the future molecular improvement of chrysanthemum.

The RAV transcription factor TEMPRANILLO1 involved in ethylene-mediated delay of chrysanthemum flowering

TEMPRANILLO1 (TEM1) is a transcription factor belonging to related to ABI3 and VP1 family, which is also known as ethylene response DNA-binding factor 1 and functions as a repressor of flowering in Arabidopsis. Here, a putative homolog of AtTEM1 was isolated and characterized from chrysanthemum, designated as CmTEM1. Exogenous application of ethephon leads to an upregulation in the expression of CmTEM1. Knockdown of CmTEM1 promotes floral initiation, while overexpression of CmTEM1 retards floral transition. Further phenotypic observations suggested that CmTEM1 involves in the ethylene-mediated inhibition of flowering. Transcriptomic analysis established that expression of the flowering integrator CmAFL1, a member of the APETALA1/FRUITFULL subfamily, was downregulated significantly in CmTEM1-overexpressing transgenic plants compared with wild-type plants but was verified to be upregulated in amiR-CmTEM1 lines by quantitative RT-PCR. In addition, CmTEM1 is capable of binding to the promoter of the CmAFL1 gene to inhibit its transcription. Moreover, the genetic evidence supported the notion that CmTEM1 partially inhibits floral transition by targeting CmAFL1. In conclusion, these findings demonstrate that CmTEM1 acts as a regulator of ethylene-mediated delayed flowering in chrysanthemum, partly through its interaction with CmAFL1.

AtHSPR functions in gibberellin-mediated primary root growth by interacting with KNAT5 and OFP1 in Arabidopsis

KEY MESSAGE

AtHSPR forms a complex with KNAT5 and OFP1 to regulate primary root growth through GA-mediated root meristem activity. KNAT5-OFP1 functions as a negative regulator of AtHSPR in response to GA. Plant root growth is modulated by gibberellic acid (GA) signaling and depends on root meristem maintenance. ARABIDOPSIS THALIANA HEAT SHOCK PROTEIN-RELATED (AtHSPR) is a vital regulator of flowering time and salt stress tolerance. However, little is known about the role of AtHSPR in the regulation of primary root growth. Here, we report that athspr mutant exhibits a shorter primary root compared to wild type and that AtHSPR interacts with KNOTTED1-LIKE HOMEOBOX GENE 5 (KNAT5) and OVATE FAMILY PROTEIN 1 (OFP1). Genetic analysis showed that overexpression of KNAT5 or OFP1 caused a defect in primary root growth similar to that of the athspr mutant, but knockout of KNAT5 or OFP1 rescued the short root phenotype in the athspr mutant by altering root meristem activity. Further investigation revealed that KNAT5 interacts with OFP1 and that AtHSPR weakens the inhibition of GIBBERELLIN 20-OXIDASE 1 (GA20ox1) expression by the KNAT5-OFP1 complex. Moreover, root meristem cell proliferation and root elongation in 35S::KNAT5athspr and 35S::OFP1athspr seedlings were hypersensitive to GA3 treatment compared to the athspr mutant. Together, our results demonstrate that the AtHSPR-KNAT5-OFP1 module regulates root growth and development by impacting the expression of GA biosynthetic gene GA20ox1, which could be a way for plants to achieve plasticity in response to the environment.

The Cm14-3-3μ protein and CCT transcription factor CmNRRa delay flowering in chrysanthemum

Floral transition from vegetative to reproductive growth is pivotal in the plant life cycle. NUTRITION RESPONSE AND ROOT GROWTH (OsNRRa) as a CONSTANS, CONSTANS-LIKE, TOC1 (CCT) domain protein delays flowering in rice and an orthologous gene CmNRRa inhibits flowering in chrysanthemum; however, the mechanism remains unknown. In this study, using yeast two-hybrid screening, we identified the 14-3-3 family member Cm14-3-3µ as a CmNRRa-interacting protein. Biochemical assays using a combination of bimolecular fluorescence complementation (BiFC), pull-down, and Co-immunoprecipitation (Co-IP) were performed to confirm the physical interaction between CmNRRa and Cm14-3-3µ in chrysanthemum. In addition, expression analysis showed that CmNRRa, but not Cm14-3-3µ, responded to the diurnal rhythm, whereas both genes were highly expressed in the leaves. Moreover, the function in flowering time regulation of Cm14-3-3µ is similar to that of CmNRRa. Furthermore, CmNRRa repressed chrysanthemum FLOWERING LOCUS T-like 3 (CmFTL3) and APETALA 1 (AP1)/FRUITFULL (FUL)-like gene (CmAFL1), but induced TERMINAL FLOWER1 (CmTFL1) directly by binding to their promoters. Cm14-3-3µ enhanced the ability of CmNRRa to regulate the expression of these genes. These findings suggest that there is a synergistic relationship between CmNRRa and Cm14-3-3µ in flowering repression in chrysanthemum.