Evolutionary Analysis and Functional Identification of Clock-Associated PSEUDO-RESPONSE REGULATOR (PRRs) Genes in the Flowering Regulation of Roses

Abstract

Pseudo-response regulators (PRRs) are the important genes for flowering in roses. In this work, clock PRRs were genome-wide identified using Arabidopsis protein sequences as queries, and their evolutionary analyses were deliberated intensively in Rosaceae in correspondence with angiosperms species. To draw a comparative network and flow of clock PRRs in roses, a co-expression network of flowering pathway genes was drawn using a string database, and their functional analysis was studied by silencing using VIGS and protein-to-protein interaction. We revealed that the clock PRRs were significantly expanded in Rosaceae and were divided into three major clades, i.e., PRR5/9 (clade 1), PRR3/7 (clade 2), and TOC1/PRR1 (clade 3), based on their phylogeny. Within the clades, five clock PRRs were identified in Rosa chinensis. Clock PRRs had conserved RR domain and shared similar features, suggesting the duplication occurred during evolution. Divergence analysis indicated the role of duplication events in the expansion of clock PRRs. The diverse cis elements and interaction of clock PRRs with miRNAs suggested their role in plant development. Co-expression network analysis showed that the clock PRRs from Rosa chinensis had a strong association with flowering controlling genes. Further silencing of RcPRR1b and RcPRR5 in Rosa chinensis using VIGS led to earlier flowering, confirming them as negative flowering regulators. The protein-to-protein interactions between RcPRR1a/RcPRR5 and RcCO suggested that RcPRR1a/RcPRR5 may suppress flowering by interfering with the binding of RcCO to the promoter of RcFT. Collectively, these results provided an understanding of the evolutionary profiles as well as the functional role of clock PRRs in controlling flowering in roses.

Keywords: clock PRRs; evolution; flowering; rose.

Figure 1

Clock PRRs in basal angiosperms, Brassicaceae and Rosaceae. (A) List of the copy number of identified clock PRRs in our study; (B) boxplots of the copy numbers of clock PRRs in basal angiosperms, Brassicaceae and Rosaceae. Clock PRRs-A, Clock PRRs-B, and Clock PRRs-R denote the number of clock PRRs in Basal Angiosperms, Brassicaceae, and Rosaceae, respectively; (C) boxplots of the copy numbers of clock PRRs in subfamilies of Rosaceae. Clock PRRs-R (Rose) and clock PRRs-R (Amgd) denote subfamily Rosoideae and Amygdaloideae, respectively.

Figure 2

Phylogeny of clock PRR genes in Rosaceae. (A) Phylogenetic tree of all clock PRRs from 2 basal angiosperm species, 18 Rosaceae species, 8 Brassicaceae species, and 2 basal Rosids species. Clock PRRs of Rosaceae classified into three major clades on the basis of their phylogeny shown in different branch colors; (B) species tree constructed on the basis of clock PRRs phylogenetic relation and the number of clock PRRs in each clade of every species.

Figure 3

Conserved motifs and domains, gene structure organization, and chromosomal localization of clock PRRs in Rosa chinensis. (A) discovered motifs; (B) discovered domains; (C) gene structure organization; (D) logos of the identified motifs; and (E) chromosomal localization.

Figure 4

cis-regulatory elements present in the promoter region of clock PRRs of rose.

Figure 5

The interaction network of known miRNA of Rosa chinensis with clock PRRs. The network was performed by psRNAtarget tool and Cytoscape.

Figure 6

The interaction network of chemical compounds with clock PRRs. The network was performed by STRING and Cytoscape.

Figure 7

Co-expression network and functional annotation of clock PRRs genes with other flowering pathways genes in Rosa chinensis. Different colors of nodes represent genes sharing biological processes, while the edges/lines connecting the nodes represent the protein-to-protein interaction between genes. The addition of edges/lines between two nodes signifies a more significant interaction.

Figure 8

Silencing of RcPRR1a and RcPRR5 in Rosa chinensis. (A) Flowering phenotype; (B) flowering time (days) of Rosa chinensis; and (C) expression levels of RcPRR1a and RcPRR5 in two independent gene silenced lines; (D) expression levels of RcCO in two independent gene silenced lines; (E) expression levels of RcFT in two independent gene silenced lines. Gray bars, pink bars, and light green bars indicated the two independent lines of control, silenced lines of RcPRR1a and silenced lines of RcPRR5, respectively. Rose GAPDH gene was used as a reference. Three biological replicates were performed for each experiment. Asterisks above the bars indicate significant differences between gene silenced lines and the control as determined by the LSD test, ** p < 0.01 and * p < 0.05.

Figure 9

The protein-to-protein interactions of RcPRR1a and RcPRR5 with RcCO in Nicotiana benthamiana leaves. (A) 35:LUC-N+35:RcCO:LUC-C and 35:RcPRR1a:LUC-N+35:RcCO:LUC-C represent the interaction of LUC with RcCO and RcPRR1a, respectively, while 35:RcPRR1a:LUC-N+35:RcCO:LUC-C represent the interaction RcPRR1a with RcCO (B) 35:LUC-N+35:RcCO:LUC-C and 35:RcPRR5:LUC-N+35:RcCO:LUC-C represent the interaction of LUC with RcCO and RcPRR5, respectively, while 35:RcPRR5:LUC-N+35:RcCO:LUC-C represent the interaction RcPRR5 with RcCO.