Branching genes are conserved across species. Genes controlling a novel signal in pea are coregulated by other long-distance signals

Abstract

Physiological and genetic studies with the ramosus (rms) mutants in garden pea (Pisum sativum) and more axillary shoots (max) mutants in Arabidopsis (Arabidopsis thaliana) have shown that shoot branching is regulated by a network of long-distance signals. Orthologous genes RMS1 and MAX4 control the synthesis of a novel graft-transmissible branching signal that may be a carotenoid derivative and acts as a branching inhibitor. In this study, we demonstrate further conservation of the branching control system by showing that MAX2 and MAX3 are orthologous to RMS4 and RMS5, respectively. This is consistent with the long-standing hypothesis that branching in pea is regulated by a novel long-distance signal produced by RMS1 and RMS5 and that RMS4 is implicated in the response to this signal. We examine RMS5 expression and show that it is more highly expressed relative to RMS1, but under similar transcriptional regulation as RMS1. Further expression studies support the hypothesis that RMS4 functions in shoot and rootstock and participates in the feedback regulation of RMS1 and RMS5 expression. This feedback involves a second novel long-distance signal that is lacking in rms2 mutants. RMS1 and RMS5 are also independently regulated by indole-3-acetic acid. RMS1, rather than RMS5, appears to be a key regulator of the branching inhibitor. This study presents new interactions between RMS genes and provides further evidence toward the ongoing elucidation of a model of axillary bud outgrowth in pea.

Figure 1.

Predicted amino acid sequence of the PsMAX2/RMS4 protein and alignment with Arabidopsis and rice MAX2 predicted protein sequences. The F-box is shown with one of the degenerate F-box consensus sequences above (Patton et al., 1998). The LRRs conserved between the three proteins are underlined and highlighted in gray. Numbers corresponding to the different mutant alleles denote relative positions of mutations with a list of the different mutant alleles for rms4 indicating the change to the nucleotide and the resulting amino acid.

Figure 2.

Neighbor-joining tree of well-characterized F-box proteins and calculated sequence identity with PsRMS4. The bootstrap value for 1,000 trials is indicated for each branch. Accession numbers for protein sequences: M. truncatula cyclin-like F-box (ABD32619); AtTIR1 (NP567135); AtCOI1 (O04197); HsSKP2 (NP005974); OsD3 (BAD69288); PsRMS4 (ABD67495); AtMAX2 (NP565979); AtFBL4 (AAM60829); LeEBF2 (ABC24972); and AtEBF1(NP565597).

Figure 3.

Alignment of the predicted amino acid sequence of PsCCD7 cv PI 26918 (DQ403160) compared with Arabidopsis MAX3 (NP_182026), rice HTD1 (AL663000.4), Nostoc sp. PCC7120/BAB75983, Synechocystis sp. SynACO (2BIX_B), pea CCD1 (BAC10549), and pea CCD8/RMS1 (AAS66907). Intron positions corresponding to the genomic DNA sequence are denoted by triangles and positions of rms5 mutations by a shaded residue and asterisk (*). Conserved His (H) residues implicated in binding of Fe²⁺ in the active site are highlighted in green; residues highlighted in gray are implicated in the active site of the structurally characterized Synechocystis sp. apocarotenoid-15,15′-oxygenase (Kloer et al., 2005).

Figure 4.

A, RMS4 gene expression in the different tissues of wild-type pea plants (cv Torsdag) determined by real-time PCR. RNA was extracted from the dissected plants at the five-node stage. Values are average ± se of two biological replicates (except for roots; n = 1) of pools of eight plants. B, Scheme of a node showing the different parts of the pea compound leaf.

Figure 5.

RMS5 and RMS1 gene expression. A, Tissue profile of RMS5 gene expression. B, RMS5 and RMS1 gene expression in the vascular tissue and remaining stem tissue from the upper internode (2.5–3.0 cm from apex) of wild-type plants. RNA was extracted from the dissected plant tissues at the five-node stage (14 to 15 d old). Values are average ± se of two biological replicates of pools of 12 plants, except for leaves (n = 8) and roots (n = 4). RMS1 and RMS5 expression shown is relative to RMS1 in the vascular bundle.

Figure 6.

RMS1 and RMS5 gene expression after 24 h at node 5 of intact, decapitated IAA (3,000 mg L⁻¹) and NPA-treated plants relative to RMS1 and RMS5 gene expression in intact plants, respectively. RNA was extracted from the dissected plant tissues at the five-node stage (14 to 15 d old). Values are average ± se of two biological replicates of pools of eight plants. RMS1 and RMS5 expression shown is relative to RMS1 and RMS5 expression in intact plants, independently.

Figure 7.

RMS5 gene expression in epicotyls of wild-type (cv Torsdag), rms1, rms2, and rms1 rms2 double-mutant plants. RNA was extracted from the dissected plant tissues at the five-node stage (14 to 15 d old). Values are average ± se of three biological replicates of pools of six plants.

Figure 8.

Gene expression and branching phenotype in reciprocal grafts between wild type and rms4 grown under short-day conditions (12-h photoperiod) measured 40 d after grafting. A, RMS1 and RMS5 gene expression in the epicotyls of rootstocks relative to the wild-type self grafts. Values are average ± se of two biological replicates of five to six plants. B, Number of buds or branches >2 mm in length of reciprocal grafts between wild type and rms4 (n = 10–12). C, Total lateral length (mm) of branches of reciprocal grafts between wild type and rms4 (n = 10–12).