Feedback regulation of xylem cytokinin content is conserved in pea and Arabidopsis

Increased-branching mutants of garden pea (Pisum sativum; ramosus [rms]) and Arabidopsis (Arabidopsis thaliana; more axillary branches) were used to investigate control of cytokinin export from roots in relation to shoot branching. In particular, we tested the hypothesis that regulation of xylem sap cytokinin is dependent on a long-distance feedback signal moving from shoot to root. With the exception of rms2, branching mutants from both species had greatly reduced amounts of the major cytokinins zeatin riboside, zeatin, and isopentenyl adenosine in xylem sap compared with wild-type plants. Reciprocally grafted mutant and wild-type Arabidopsis plants gave similar results to those observed previously in pea, with xylem sap cytokinin down-regulated in all graft combinations possessing branched shoots, regardless of root genotype. This long-distance feedback mechanism thus appears to be conserved between pea and Arabidopsis. Experiments with grafted pea plants bearing two shoots of the same or different genotype revealed that regulation of root cytokinin export is probably mediated by an inhibitory signal. Moreover, the signaling mechanism appears independent of the number of growing axillary shoots because a suppressed axillary meristem mutation that prevents axillary meristem development at most nodes did not abolish long-distance regulation of root cytokinin export in rms4 plants. Based on double mutant and grafting experiments, we conclude that RMS2 is essential for long-distance feedback regulation of cytokinin export from roots. Finally, the startling disconnection between cytokinin content of xylem sap and shoot tissues of various rms mutants indicates that shoots possess powerful homeostatic mechanisms for regulation of cytokinin levels.     

Long-distance signaling and the control of branching in the rms1 mutant of pea

The ramosus (rms) mutation (rms1) of pea (Pisum sativum) causes increased branching through modification of graft-transmissible signal(s) produced in rootstock and shoot. Additional grafting techniques have led us to propose that the novel signal regulated by Rms1 moves acropetally in shoots and acts as a branching inhibitor. Epicotyl interstock grafts showed that wild-type (WT) epicotyls grafted between rms1 scions and rootstocks can revert mutant scions to a WT non-branching phenotype. Mutant scions grafted together with mutant and WT rootstocks did not branch despite a contiguous mutant root-shoot system. The primary action of Rms1 is, therefore, unlikely to be to block transport of a branching stimulus from root to shoot. Rather, Rms1 may influence a long-distance signal that functions, directly or indirectly, as a branching inhibitor. It can be deduced that this signal moves acropetally in shoots because WT rootstocks inhibit branching in rms1 shoots, and although WT scions do not branch when grafted to mutant rootstocks, they do not inhibit branching in rms1 cotyledonary shoots growing from the same rootstocks. The acropetal direction of transport of the Rms1 signal supports previous evidence that the rms1 lesion is not in an auxin biosynthesis or transport pathway. The different branching phenotypes of WT and rms1 shoots growing from the same rms1 rootstock provides further evidence that the shoot has a major role in the regulation of branching and, moreover, that root-exported cytokinin is not the only graft-transmissible signal regulating branching in intact pea plants.     

Branching Mutant rms-2 in Pisum sativum (Grafting Studies and Endogenous Indole-3-Acetic Acid Levels)

Isogenic lines of pea (Pisum sativum L.) were used to determine the physiological site of action of the Rms-2 gene, which maintains apical dominance, and its effect on endogenous free indole-3-acetic acid (IAA) levels. In mutant rms-2 scions, which normally produce lateral branches below node 3 and above node 7, apical dominance was almost fully restored by grafting to Rms-2 (wild-type) stocks. In the reciprocal grafts, rms-2 stocks did not promote branching in wild-type shoots. Together, these results suggest that the Rms-2 gene inhibits branching in the shoot of pea by controlling the synthesis of a translocatable (hormone-like) substance that is produced in the roots and/or cotyledons and in the shoot. At all stages, including the stage at which aerial lateral buds commence outgrowth, the level of IAA in rms-2 shoots was elevated (up to 5-fold) in comparison with that in wild-type shoots. The internode length of rms-2 plants was 40% less than in wild-type plants, and the mutant plants allocated significantly more dry weight to the shoot than to the root in comparison with wild-type plants. Grafting to wild-type stocks did not normalize IAA levels or internode length in rms-2 scions, even though it inhibited branching, suggesting that the involvement of Rms-2 in the control of IAA level and internode length may be confined to processes in the shoot.     

Apical wilting and petiole xylem vessel diameter of the rms2 branching mutant of pea are shoot controlled and independent of a long-distance signal regulating branching

RMS2 (RAMOSUS2) affects the level or transport of a graft-transmissible signal produced in the shoot and root that controls axillary bud outgrowth in pea (Pisum sativum L.). The shoot apex of rms2 transiently wilts under high evaporative demand. The origin of this phenotype was investigated to determine whether it was involved in the regulation of branching. Wild-type (WT) and rms2 leaves showed a similar stomatal conductance at both low and high evaporative demand in vivo, indicating normal stomatal function. Leaves of both genotypes had similar ABA content and response to ABA. Although root hydraulic conductance (determined by pressure-induced flow) of rms2 plants was normal, more xylem vessels per vascular bundle were identified in cross-sections of fully expanded rms2 petioles compared with those of the WT. However, the diameter of these vessels was nearly half that of the WT. Since the conductance of each vessel is proportional to the fourth power of the vessel radius (according to the Hagen-Poiseulle law), the theoretical (calculated) petiole hydraulic conductance of rms2 was greatly decreased compared with WT plants. Under high evaporative demand, this would cause a temporary imbalance between water supply to, and demand from, rms2 shoots, directly resulting in the wilting phenotype of the mutant. Reciprocal grafting showed that xylem vessel development in rms2 shoots is strictly shoot controlled, probably via elevated auxin levels. This altered xylem vessel development, though causing wilting in rms2 shoot tips, does not appear to affect shoot branching.     

Mutational Analysis of Branching in Pea. Evidence That Rms1 and Rms5 Regulate the Same Novel Signal

The fifth increased branching ramosus (rms) mutant, rms5, from pea (Pisum sativum), is described here for phenotype and grafting responses with four other rms mutants. Xylem sap zeatin riboside concentration and shoot auxin levels in rms5 plants have also been compared with rms1 and wild type (WT). Rms1 and Rms5 appear to act closely at the biochemical or cellular level to control branching, because branching was inhibited in reciprocal epicotyl grafts between rms5 or rms1 and WT plants, but not inhibited in reciprocal grafts between rms5 and rms1 seedlings. The weakly transgressive or slightly additive phenotype of the rms1 rms5 double mutant provides further evidence for this interaction. Like rms1, rms5 rootstocks have reduced xylem sap cytokinin concentrations, and rms5 shoots do not appear deficient in indole-3-acetic acid or 4-chloroindole-3-acetic acid. Rms1 and Rms5 are similar in their interaction with other Rms genes. Reciprocal grafting studies with rms1, rms2, and rms5, together with the fact that root xylem sap cytokinin concentrations are reduced in rms1 and rms5 and elevated in rms2 plants, indicates that Rms1 and Rms5 may control a different pathway than that controlled by Rms2. Our studies indicate that Rms1 and Rms5 may regulate a novel graft-transmissible signal involved in the control of branching.     

Auxin inhibition of decapitation-induced branching is dependent on graft-transmissible signals regulated by genes Rms1 and Rms2

Decapitation-induced axillary bud outgrowth is a vital mechanism whereby shoots are able to continue normal growth and development. In many plants, including wild-type garden pea (Pisum sativum L.), this process can be inhibited by exogenous auxin. Using the ramosus (rms) increased branching mutants of pea, we present evidence that this response to auxin is dependent on graft-transmissible substance(s) regulated by the genes Rms1 and Rms2. The response to exogenous auxin is massively diminished in decapitated rms1 and rms2 mutant plants. However, basipetal auxin transport is not reduced in intact or decapitated mutants. Grafting rms1 or rms2 shoots onto wild-type rootstocks restored the auxin response, indicating that Rms1 and Rms2 gene action in the rootstock is sufficient to enable an auxin response in mutant shoots. We conclude that Rms1 and Rms2 act in the rootstock and shoot to control levels of mobile substance(s) that interact with exogenous auxin in the inhibition of bud outgrowth after decapitation. At least for rms1, the reduced auxin response is unlikely to be due to an inability of auxin to decrease xylem sap cytokinin content, as this is already low in intact rms1 plants. Consequently, we have genetic evidence that auxin action in decapitated plants depends on at least one novel long-distance signal.     

Long-distance signalling and a mutational analysis of branching in pea

Four ramosus mutants with increased branching at basal andaerial nodes have been used to investigate the genetic regulation of budoutgrowth in Pisum sativum L. (garden pea). Studies oflong-distance signalling, xylem sap cytokinin concentrations, shootauxin level, auxin transport and auxin response are discussed. A modelof branching control is presented that encompasses twograft-transmissible signals in addition to auxin and cytokinin. Mutantsrms1 through rms4 are not deficient in indole-3-aceticacid (IAA) or in the basipetal transport of this hormone. Three of thefour mutants, rms1, rms3 and rms4, have veryreduced cytokinin concentrations in xylem sap from roots. This reductionin xylem sap cytokinin concentration appears to be caused by a propertyof the shoot and may be part of a feedback mechanism induced by anaspect of bud outgrowth. The shoot-to-root feedback signal is unlikelyto be auxin itself, as auxin levels and transport are not correlatedwith xylem sap cytokinin concentrations in various intact and graftedmutant and wild-type plants. Rms1 and Rms2 act inshoot and rootstock to regulate the level or transport ofgraft-transmissible signals. Various grafting studies and double mutantanalyses have associated Rms2 with the regulation of theshoot-to-root feedback signal. Rms1 is associated with a secondunknown graft-transmissible signal that is postulated to move in thedirection of root-to-shoot. Exogenous auxin appears to interact withboth of the signals regulated by Rms1 and Rms2 in theinhibition of branching after decapitation. The action of Rms3and Rms4 is less apparent at this stage, although both appearto act largely in the shoot.     

Pea rms6 mutants exhibit increased basal branching

Our studies on two branching mutants of pea ( Pisum sativum L.) have identified a further Ramosus locus , Rms6, with two recessive or partially recessive mutant alleles: rms6-1 (type line S2-271) and rms6-2 (type line K586). Mutants rms6-1 and rms6-2 were derived from dwarf and tall cultivars, Solara and Torsdag, respectively. The rms6 mutants are characterized by increased branching from basal nodes. In contrast, mutants rms1 through rms5 have increased branching from both basal and aerial (upper stem) nodes. Buds at the cotyledonary node of wild-type (WT) plants remain dormant but in rms6 plants these buds were usually released from dormancy. Their growth was either subsequently inhibited, sometimes even prior to emergence above ground, or they grew into secondary stems. The mutant phenotype was strongest for rms6-1 on the dwarf background. Although rms6-2 had a weak single-mutant phenotype, the rms3-1 rms6-2 double mutant showed clear transgression and an additive branching phenotype, with a total lateral length almost 2-fold greater than rms3-1 and nearly 5-fold greater than rms6-2 . Grafting studies between WT and rms6-1 plants demonstrated the primary action of Rms6 may be confined to the shoot. Young WT and rms6-1 shoots had similar auxin levels, and decapitated plants had a similar magnitude of response to applied auxin. Abscisic acid levels were elevated 2-fold at node 2 of young rms6-1 plants. The Rms6 locus mapped to the R to Gp segment of linkage group V (chromosome 3). The rms6 mutants will be useful for basic research and also have possible agronomical value.     

Branching in Pea (Action of Genes Rms3 and Rms4)

The nonallelic ramosus mutations rms3-2 and rms4 of pea (Pisum sativum L.) cause extensive release of vegetative axillary buds and lateral growth in comparison with wild-type (cv Torsdag) plants, in which axillary buds are not normally released under the conditions utilized. Grafting studies showed that the expression of the rms4 mutation in the shoot is independent of the genotype of the root-stock. In contrast, the length of the branches at certain nodes of rms3-2 plants was reduced by grafting to wild-type stocks, indicating that the wild-type Rms3 gene may control the level of a mobile substance produced in the root. This substance also appears to be produced in the shoot because Rms3 shoots did not branch when grafted to mutant rms3-2 rootstocks. However, the end product of the Rms3 gene appears to differ from that of the Rms2 gene (C.A. Beveridge, J.J. Ross, and I.C. Murfet [1994] Plant Physiol 104: 953-959) because reciprocal grafts between rms3-2 and rms2 seedlings produced mature shoots with apical dominance similar to that of rms3-2 and rms2 shoots grafted to wild-type stocks. Indole-3-acetic acid levels were not reduced in apical or nodal portions of rms4 plants and were actually elevated (up to 2-fold) in rms3-2 plants. It is suggested that further studies with these branching mutants may enable significant progress in understanding the normal control of apical dominance and the related communication between the root and shoot.     

The shoot controls zeatin riboside export from pea roots. Evidence from the branching mutant rms4

The rms4 mutant of pea ( Pisum sativum L.) was used in grafting studies and cytokinin analyses of the root xylem sap to provide evidence that, at least for pea, the shoot can modify the import of cytokinins from the root. The rms4 mutation, which confers a phenotype with increased branching in the shoot, causes a very substantial decrease (down to 40-fold less) in the concentration of zeatin riboside (ZR) in the xylem sap of the roots. Results from grafts between wild-type (WT) and rms4 plants indicate that the concentration of cytokinins in the xylem sap of the roots is determined almost entirely by the genotype of the shoot. WT scions normalize the cytokinin concentration in the sap of rms4 mutant roots, whereas mutant scions cause WT roots to behave like those of self-grafted mutant plants. The mechanism whereby rms4 shoots of pea cause a down-regulation in the export of cytokinins from the roots is unknown at this time. However, our data provide evidence that the shoot transmits a signal to the roots and thereby controls processes involved in the regulation of cytokinin biosynthesis in the root.     

Strigolactones promote nodulation in pea

Strigolactones are recently defined plant hormones with roles in mycorrhizal symbiosis and shoot and root architecture. Their potential role in controlling nodulation, the related symbiosis between legumes and Rhizobium bacteria, was explored using the strigolactone-deficient rms1 mutant in pea (Pisum sativum L.). This work indicates that endogenous strigolactones are positive regulators of nodulation in pea, required for optimal nodule number but not for nodule formation per se. rms1 mutant root exudates and root tissue are almost completely deficient in strigolactones, and rms1 mutant plants have approximately 40% fewer nodules than wild-type plants. Treatment with the synthetic strigolactone GR24 elevated nodule number in wild-type pea plants and also elevated nodule number in rms1 mutant plants to a level similar to that seen in untreated wild-type plants. Grafting studies revealed that nodule number and strigolactone levels in root tissue of rms1 roots were unaffected by grafting to wild-type scions indicating that strigolactones in the root, but not shoot-derived factors, regulate nodule number and provide the first direct evidence that the shoot does not make a major contribution to root strigolactone levels.     

Computational Modeling and Molecular Physiology Experiments Reveal New Insights into Shoot Branching in Pea

Bud outgrowth is regulated by the interplay of multiple hormones, including auxin, cytokinin, strigolactones, and an unidentified long-distance feedback signal that moves from shoot to root. The model of bud outgrowth regulation in pea (Pisum sativum) includes these signals and a network of five RAMOSUS (RMS) genes that operate in a shoot-root-shoot loop to regulate the synthesis of, and response to, strigolactones. The number of components in this network renders the integration of new and existing hypotheses both complex and cumbersome. A hypothesis-driven computational model was therefore developed to help understand regulation of shoot branching. The model evolved in parallel with stepwise laboratory research, helping to define and test key hypotheses. The computational model was used to verify new mechanisms involved in the regulation of shoot branching by confirming that the new hypotheses captured all relevant biological data sets. Based on cytokinin and RMS1 expression analyses, this model is extended to include subtle but important differences in the function of RMS3 and RMS4 genes in the shoot and rootstock. Additionally, this research indicates that a branch-derived signal upregulates RMS1 expression independent of the other feedback signal. Furthermore, we propose xylem-sap cytokinin promotes sustained bud outgrowth, rather than acting at the earlier stage of bud release.     

Modulation of hepatocyte nuclear factor-4alpha function by the peroxisome-proliferator-activated receptor-gamma co-activator-1alpha in the acute-phase response

Recent studies with the high-tillering mutants in rice (Oryza sativa), the max (more axillary growth) mutants in Arabidopsis thaliana and the rms (ramosus) mutants in pea (Pisum sativum) have indicated the presence of a novel plant hormone that inhibits branching in an auxin-dependent manner. The synthesis of this inhibitor is initiated by the two CCDs [carotenoid-cleaving (di)oxygenases] OsCCD7/OsCCD8b, MAX3/MAX4 and RMS5/RMS1 in rice, Arabidopsis and pea respectively. MAX3 and MAX4 are thought to catalyse the successive cleavage of a carotenoid substrate yielding an apocarotenoid that, possibly after further modification, inhibits the outgrowth of axillary buds. To elucidate the substrate specificity of OsCCD8b, MAX4 and RMS1, we investigated their activities in vitro using naturally accumulated carotenoids and synthetic apocarotenoid substrates, and in vivo using carotenoid-accumulating Escherichia coli strains. The results obtained suggest that these enzymes are highly specific, converting the C27 compounds beta-apo-10'-carotenal and its alcohol into beta-apo-13-carotenone in vitro. Our data suggest that the second cleavage step in the biosynthesis of the plant branching inhibitor is conserved in monocotyledonous and dicotyledonous species.