A carotenoid-derived hormonal signal that inhibits shoot branching in plants has long escaped identification. Strigolactones are compounds thought to be derived from carotenoids and are known to trigger the germination of parasitic plant seeds and stimulate symbiotic fungi. Here we present evidence that carotenoid cleavage dioxygenase 8 shoot branching mutants of pea are strigolactone deficient and that strigolactone application restores the wild-type branching phenotype to ccd8 mutants. Moreover, we show that other branching mutants previously characterized as lacking a response to the branching inhibition signal also lack strigolactone response, and are not deficient in strigolactones. These responses are conserved in Arabidopsis. In agreement with the expected properties of the hormonal signal, exogenous strigolactone can be transported in shoots and act at low concentrations. We suggest that endogenous strigolactones or related compounds inhibit shoot branching in plants. Furthermore, ccd8 mutants demonstrate the diverse effects of strigolactones in shoot branching, mycorrhizal symbiosis and parasitic weed interaction.

Shoot branching is inhibited by auxin transported down the stem from the shoot apex. Auxin does not accumulate in inhibited buds and so must act indirectly. We show that mutations in the MAX4 gene of Arabidopsis result in increased and auxin-resistant bud growth. Increased branching in max4 shoots is restored to wild type by grafting to wild-type rootstocks, suggesting that MAX4 is required to produce a mobile branch-inhibiting signal, acting downstream of auxin. A similar role has been proposed for the pea gene, RMS1. Accordingly, MAX4 and RMS1 were found to encode orthologous, auxin-inducible members of the polyene dioxygenase family.Supplemental material is available at http://www.genesdev. org.Received December 6, 2002; revised version accepted March 20, 2003. Variation in shoot branching is an important cause of diversity in plant form. Individual species have a characteristic branching pattern, which can change through the life cycle in response to developmental cues and to environmental conditions (Cline 1991;Beveridge et al. 2003). Branching control therefore requires the integration of many signals, both known and unknown.Shoot branches arise from axillary meristems that form in the axils of leaves on the primary shoot axis. The axillary meristems themselves initiate leaves to form a bud. Bud growth can arrest but has the potential to reactivate to produce a shoot branch. Removal of the primary shoot apex results in activation of arrested axillary buds. The ability of the shoot apex to repress axillary bud growth is termed apical dominance. Thimann and Skoog (1933) reported that a compound, derived from the shoot apex, and later identified as auxin (indole-3-acetic acid), could inhibit the growth of lateral buds when applied to the stump of a decapitated plant. Subsequent work has provided multiple lines of evidence in support of auxinmediated bud inhibition in planta. However, a second messenger must relay the auxin signal into the bud because apically derived auxin is not transported into buds (Morris 1977) and exogenous auxin applied directly to buds does not inhibit their growth (Cline 1996).One model proposes that the effect of auxin on bud growth is mediated by cytokinin. Cytokinin can directly promote bud growth (Cline 1991); transgenic plants with increased auxin levels have reduced cytokinin levels (Eklö f et al. 2000), and cytokinin export from roots increases after decapitation, with this increase being abolished by application of auxin to the decapitated stump (Bangerth 1994). However, there is also good evidence for novel regulators of bud growth downstream of auxin. The ramosus mutants (rms1 to rms5) of pea (for reviews, see Beveridge 2000; Beveridge et al. 2003) have increased lateral branching, but this phenotype can be almost completely rescued by grafting a wild-type (WT) rootstock to an rms1, rms2, or rms5 mutant scion. Such grafting studies show that RMS1 and RMS5 are required for the production of a graft transmissible signal that moves from root to shoot and inhibits branching ...

Cytokinin (CK) has long been implicated as a promoter of bud outgrowth in plants, but exactly how this is achieved in coordination with other plant hormones is unclear. The recent discovery of strigolactones (SLs) as the long-sought branchinhibiting hormone allowed us to test how CK and SL coordinately regulate bud outgrowth in pea (Pisum sativum). We found that SL-deficient plants are more sensitive to stimulation of bud growth by low concentrations of locally applied CK than wildtype plants. Furthermore, in contrast with SL mutant plants, buds of wild-type plants are almost completely resistant to stimulation by CK supplied to the vasculature. Regardless of whether the exogenous hormones were supplied locally or to the xylem stream, SL and CK acted antagonistically on bud outgrowth. These data suggest that SLs do not affect the delivery of CK to axillary buds and vice versa. Rather, these data combined with dose-response experiments suggest that SLs and CK can act directly in buds to control their outgrowth. These hormones may converge at a common point in the bud outgrowth regulatory pathway. The expression of pea BRANCHED1, a TCP transcription factor expressed strongly in buds and thought to act downstream of SLs in shoot branching, is regulated by CK and SL without a requirement for protein synthesis and in a manner that correlates with observed bud growth responses.Shoot branching is a major determinant of plant shoot architecture. Many factors contribute to the ability of an axillary bud to grow out to form a branch, including developmental, positional, genetic, hormonal, and environmental factors. Auxin, cytokinin (CK), and strigolactones (SLs) are implicated in the hormonal regulation of bud outgrowth; auxin and SLs as inhibitors of bud outgrowth and CK as a promoter of bud outgrowth (Dun et al., 2009a;Leyser, 2009;Beveridge and Kyozuka, 2010). Many studies over a number of decades have investigated the antagonistic action of auxin and CK in bud outgrowth control (ShimizuSato et al., 2009) and, more recently, the relationships between auxin and SL (Brewer et al., 2009;Crawford et al., 2010;Liang et al., 2010), but how SL and CK integrate to antagonistically control bud outgrowth remains unclear.Prior to their identification as hormones involved in shoot branching, certain properties of SLs were characterized based on studies of the long-distance branchinhibiting signal in a series of increased branching mutants. These mutants include ramosus (rms) in pea (Pisum sativum), more axillary growth (max) in Arabidopsis (Arabidopsis thaliana), decreased apical dominance (dad) in Petunia hybrida, and dwarf (d) and high tillering dwarf (htd) in rice (Oryza sativa; for review, see Dun et al., 2009a;Beveridge and Kyozuka, 2010;Domagalska and Leyser, 2011). Grafting studies demonstrated that the branch-inhibiting signal can be synthesized in root or shoot tissue, moves upward to inhibit bud outgrowth, and that a subset of the branching mutants are unable to synthesize the signal (now named SL synthesis mutants; rms1...