Plant Cell SoftGenetics
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

First published online January 19, 2005; 10.1105/tpc.104.026716

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
17/2/464    most recent
tpc.104.026716v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via ISI Web of Science (43)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Foo, E.
Right arrow Articles by Beveridge, C. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Foo, E.
Right arrow Articles by Beveridge, C. A.
Agricola
Right arrow Articles by Foo, E.
Right arrow Articles by Beveridge, C. A.
The Plant Cell 17:464-474 (2005)
© 2005 American Society of Plant Biologists

The Branching Gene RAMOSUS1 Mediates Interactions among Two Novel Signals and Auxin in Pea

Eloise Fooa,1, Erika Bullierb, Magali Goussotb, Fabrice Foucherb,2, Catherine Rameaub and Christine Anne Beveridgea,3

a Australian Research Council Centre of Excellence for Integrative Legume Research, University of Queensland, St. Lucia, Queensland, 4072, Australia
b Station de Génétique et d'Amélioration des Plantes, Institut J.P. Bourgin, Institut National de la Recherche Agronomique, 78026, Versailles Cedex, France

3 To whom correspondence should be addressed. E-mail c.beveridge{at}botany.uq.edu.au; fax 61-0-7-3365-1699.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 METHODS
 REFERENCES
 
In Pisum sativum, the RAMOSUS genes RMS1, RMS2, and RMS5 regulate shoot branching via physiologically defined mobile signals. RMS1 is most likely a carotenoid cleavage enzyme and acts with RMS5 to control levels of an as yet unidentified mobile branching inhibitor required for auxin inhibition of branching. Our work provides molecular, genetic, and physiological evidence that RMS1 plays a central role in a shoot-to-root-to-shoot feedback system that regulates shoot branching in pea. Indole-3-acetic acid (IAA) positively regulates RMS1 transcript level, a potentially important mechanism for regulation of shoot branching by IAA. In addition, RMS1 transcript levels are dramatically elevated in rms3, rms4, and rms5 plants, which do not contain elevated IAA levels. This degree of upregulation of RMS1 expression cannot be achieved in wild-type plants by exogenous IAA application. Grafting studies indicate that an IAA-independent mobile feedback signal contributes to the elevated RMS1 transcript levels in rms4 plants. Therefore, the long-distance signaling network controlling branching in pea involves IAA, the RMS1 inhibitor, and an IAA-independent feedback signal. Consistent with physiological studies that predict an interaction between RMS2 and RMS1, rms2 mutations appear to disrupt this IAA-independent regulation of RMS1 expression.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 METHODS
 REFERENCES
 
The timing of axillary bud outgrowth and the ensuing growth of axillary shoots is one of the key determinants of shoot architecture. Before the last decade, decapitation was used almost exclusively as a means to study shoot branching. The term apical dominance was coined to describe the suppression of axillary bud outgrowth by the shoot tip (e.g., Cline, 1996Go). In many species, replacement of the shoot apex with exogenous indole-3-acetic acid (IAA) can maintain branching inhibition (e.g., Cline, 1996Go). Several studies have suggested that IAA may suppress axillary shoot branching via an acropetally moving second messenger. For example, in pea (Pisum sativum) plants with two decapitated shoots, replacement of one shoot tip with exogenous IAA can lead to inhibition of branching in both shoots even though very little auxin moves between shoots (Morris, 1977Go). Furthermore, in bean (Phaseolus vulgaris), IAA supplied to the shoot tip fails to enter axillary buds below (Hall and Hillman, 1975Go), and direct IAA application to axillary buds was unable to inhibit their growth (Yeang and Hillman, 1982Go). These and other studies have led to the hypothesis that apically derived auxin maintains apical dominance via a second signal, such as cytokinin (Bangerth, 1994Go; Li et al., 1995Go).

Characterization of relatively nonpleiotropic increased branching mutants in pea (ramosus [rms]; e.g., Beveridge et al., 1997bGo), Arabidopsis thaliana (more axillary growth [max]; Stirnberg et al., 2002Go; Turnbull et al., 2002Go; Sorefan et al., 2003Go), and petunia (Petunia hybrida) (decreased apical dominance [dad]; Napoli, 1996Go; Napoli and Ruehle, 1996Go) has highlighted the influence of tissues outside of the shoot apex on axillary branching. Grafting studies with these mutants indicate that root and stem tissue can influence branching in the shoot via long-distance signal(s) (Napoli, 1996Go; Turnbull et al., 2002Go; Beveridge et al., 2003Go; Leyser, 2003Go; Sorefan et al., 2003Go; Booker et al., 2004Go). The long-distance signal(s) involved may be potent because just 5 mm of wild-type epicotyl or stem grafted between an rms1 or dad1 rootstock and scion can inhibit branching in scions even to the extent of an entire wild-type rootstock (Napoli, 1996Go; Foo et al., 2001Go).

Hormonal characterization of the rms mutants of pea has revealed that the long-distance signal(s) are most likely novel (Figure 1; Beveridge, 2000Go; Morris et al., 2001Go). RMS1, RMS2, and RMS5 regulate mobile branching signal(s) produced in shoot and rootstock tissue. Branching in rms1, rms2, and rms5 scions can be suppressed by grafting to respective wild-type rootstocks and is not promoted in reciprocal grafts of wild-type scions with rms mutant rootstocks (reviewed in Beveridge et al., 2003Go). RMS1 and RMS5 are likely to be associated with the same signal (Morris et al., 2001Go), and experiments with two-shoot grafts indicate that this signal is a branching inhibitor that may move only acropetally in shoots (Foo et al., 2001Go). Intact rms1 and rms5 mutant plants do not display increased xylem sap cytokinin levels or altered IAA level or transport (Beveridge, 2000Go; Morris et al., 2001Go). Based on this physiological evidence, the novel long-distance branching signal regulated by RMS1 will be referred to as the branching inhibitor (Figure 1).



View larger version (21K):
[in this window]
[in a new window]
  		Figure 1. Working Model of Branching Control in Pea.

Adapted from Beveridge (2000)Go, Morris et al. (2001)Go, and Beveridge et al. (2003)Go. The model includes a branching inhibitor and a feedback signal that are novel long-distance signals produced in shoot and rootstock. The feedback signal is induced under conditions of low levels or reduced response to the branching inhibitor and upregulates branching inhibitor synthesis and downregulates xylem cytokinin (CK) content. IAA depresses cytokinin both in xylem sap and shoot and increases branching inhibitor levels. Lines with flat ends indicate suppression; arrows represent promotion.

 
In addition to physiological evidence, the recent identification of RMS1 and its predicted protein sequence supports the hypothesis that the branching inhibitor is a novel hormone-like substance (Sorefan et al., 2003Go). The RMS1 protein belongs to a family of enzymes that includes carotenoid cleavage dioxygenases (CCDs) found in plants and animals (Sorefan et al., 2003Go) and is homologous to MAX4 from Arabidopsis (Sorefan et al., 2003Go) and DAD1 from petunia (Snowden et al., 2005)Go. The Arabidopsis MAX4 protein has been shown to cleave specific carotenoids when expressed in bacterial cells and may act with another CCD, MAX3, to produce the branching inhibitor in Arabidopsis (Booker et al., 2004Go; Schwartz et al., 2004Go). Carotenoids and their derivatives are a diverse group of secondary metabolites, some of which, most notably abscisic acid (ABA; Schwartz et al., 1997Go), have hormone-like activity. In the Arabidopsis genome, nine CCD genes have been identified, and of these at least five encode 9-cis-epoxycarotenoid dioxygenase enzymes involved in ABA biosynthesis (Iuchi et al., 2001Go; Schwartz et al., 2003Go; Tan et al., 2003Go). The phenotypes of mutant rms1 and max4 plants are not consistent with ABA deficiency, and gas chromatography–mass spectrometry (GC-MS) analysis of ABA concentration in three pools (n > 20) of xylem sap of rms1 and wild-type plants has given no indication that the branching inhibitor is ABA (C.A. Beveridge, J. Ross, and C. Ngo, personal communication). Recently, another branching gene in Arabidopsis that might regulate the branching inhibitor has been cloned. MAX1 encodes a putative cytochrome P450 that acts downstream of MAX3 and MAX4 (Booker et al., 2005Go).

In pea, IAA may suppress axillary shoot branching, at least partly, via the branching inhibitor (Figure 1). Exogenous IAA is unable to suppress branching in decapitated rms plants (Beveridge, 2000Go; Beveridge et al., 2000Go). This response to exogenous auxin after decapitation is regained in rms1 and rms2 shoots grafted to wild-type rootstocks, indicating that the long-distance signal(s) regulated by RMS1 and RMS2 are required for IAA action in the shoot (Beveridge, 2000Go; Beveridge et al., 2000Go). Furthermore, a brief analysis indicates that changes in IAA level in the stem by decapitation and auxin application may influence RMS1 gene expression in pea (Sorefan et al., 2003Go). In Arabidopsis, MAX4 is also required for auxin response in in vitro bud outgrowth assays, although auxin-regulated MAX4 gene expression appears to be limited to root tips (Sorefan et al., 2003Go). Unlike pea, it is difficult to inhibit branching in Arabidopsis by exogenous auxin after decapitation (Cline, 1996Go), indicating the possibility of different functions for auxin in pea and Arabidopsis in decapitated plants.

In pea and Arabidopsis, branching genes that act primarily in the shoot rather than via mobile signals have also been identified. This includes the max2 mutant of Arabidopsis (Leyser, 2003Go) and rms3 and rms4 mutants of pea (Figure 1; Beveridge et al., 1996Go), whose branching phenotypes cannot be suppressed by grafting to wild-type rootstocks. It is possible that some of these genes may act to regulate the perception of the branching inhibitor. MAX2 encodes a protein that is a member of the F-box protein family (Woo et al., 2001Go; Stirnberg et al., 2002Go). This family contains members involved in targeting proteins for ubiquitin-mediated degradation, a process integral to many hormone signaling pathways (Gray et al., 2001Go; Xu et al., 2002Go; Potuschak et al., 2003Go).

In addition to the branching inhibitor associated with RMS1, a novel feedback signal may play a role in shoot branching control in pea. Xylem sap cytokinin export from the roots is dramatically reduced in rms1, rms3, rms4, and rms5 plants (Beveridge et al., 1997aGo; Beveridge, 2000Go; Morris et al., 2001Go). This downregulation of cytokinin export from the roots appears to be due to altered production of a shoot-derived mobile signal. For example, in reciprocal grafts between rms4 and the wild type, cytokinin export from wild-type roots is suppressed by grafting to rms4 scions and is normalized in rms4 roots grafted to wild-type scions (Beveridge et al., 1997aGo). A role for IAA in suppressing cytokinin export from the roots has been proposed (Bangerth, 1994Go); however, the long-distance feedback signal described here is probably not IAA because IAA levels and transport rates are not greatly elevated in the rms1, rms3, rms4, or rms5 mutant plants (Beveridge, 2000Go; Morris et al., 2001Go; S. Morris and C.A. Beveridge, unpublished data). Furthermore, decapitation may not lead to a restoration of cytokinin export from these mutant roots (C.A. Beveridge, unpublished data).

Feedback regulation of xylem sap cytokinin levels may be RMS2 dependent (Figure 1). rms2 and rms1 rms2 double mutant plants do not display reduced xylem sap cytokinin concentrations (Beveridge et al., 1997bGo; Dodd et al., 2004Go). Indeed, rms2 plants display a small but significant increase in xylem sap cytokinin concentration but because of lower xylem sap flow rates do not show enhanced xylem sap delivery to the shoot compared with the wild type (Dodd et al., 2004Go). Because rms2 plants display normal IAA transport and increased rather than depleted shoot IAA levels, it is clear that the feedback signal disrupted in rms2 plants is unlikely to be IAA (Beveridge et al., 1994Go, 2000Go). Therefore, like the branching inhibitor associated with RMS1, the feedback signal appears to be a novel hormone-like signal.

In pea, the feedback signal may stimulate activity of the branching inhibitor. Branching is inhibited in rms1 scions grafted to rms2 rootstocks but not in rms2 scions grafted to rms1 rootstocks (Beveridge et al., 1997bGo). Although a simple explanation for this result is that RMS1 and RMS2 act on sequential steps in the production of the branching inhibitor, additional studies indicate that RMS2 probably acts before RMS1. As mentioned above, xylem sap cytokinin levels are elevated in rms1 rms2 double mutant plants and depleted in rms1 plants (Beveridge et al., 1997bGo), indicating that RMS2 probably acts before RMS1 in the control of xylem sap cytokinin levels. RMS1 and RMS2 action on different pathways is also supported by pleiotropic traits observed in rms2 mutants but not rms1 plants, such as modified pod shape (Murfet and Symons, 2000aGo; I. Dodd and C.A. Beveridge, unpublished data).

Collective analysis of these results has led to a working model of branching control in pea (Figure 1; Beveridge et al., 2003Go). In this model, a feedback signal stimulates production of a branching inhibitor and suppresses cytokinin export from the roots (Beveridge, 2000Go). RMS1 and RMS5 are required for synthesis of the branching inhibitor, RMS2 is required for the feedback pathway, and RMS3 or RMS4 act in the shoot, possibly influencing perception of the branching inhibitor. Perturbation of this network, such as by mutation of the RMS3, RMS4, or RMS5 genes would result in activation of the feedback signal and upregulation of the branching inhibitor.

In this study, RMS1 gene expression was employed as a tool to dissect the hormonal and genetic regulation of shoot branching in pea. Our studies show that IAA stimulates expression of the RMS1 gene, a potentially important mechanism for auxin action in branching control. We also present evidence that RMS1 expression is stimulated by a mobile feedback signal that is largely independent from IAA and appears to require RMS2.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 METHODS
 REFERENCES
 
The RMS1 Gene Is Mutated in 10 Independent rms1 Lines
We reported previously that a pea homolog of MAX4 from Arabidopsis cosegregated with RMS1 in an F2 mapping population and was encompassed by a large deletion in rms1-2 and rms1-3 mutant alleles (Sorefan et al., 2003Go), suggesting that RMS1 is the ortholog of MAX4 in pea. We confirmed the identity of RMS1 by investigating an additional eight rms1 mutant alleles. Genomic deletions that encompass the putative RMS1 open reading frame were detected by DNA gel blot analysis and PCR in rms1-1, rms1-8, and rms1-9 mutant alleles (data not shown). Base pair changes that are predicted to result in disrupted RMS1 protein sequence were detected in rms1-7, rms1-10, rms1-11, rms1-12, and rms1-13 mutant alleles (Figure 2). Mutant alleles containing genomic deletions were induced by x-rays or {gamma}-rays, whereas substitution rms1 mutant alleles containing base pair changes were induced by either ethyl methanesulfonate or {gamma}-rays (Symons and Murfet, 1997Go; C. Rameau, personal communication). The RMS1 gene in pea contains six exons (Figure 2) and encodes a conceptual protein of 562 amino acids that belongs to a family of polyene chain dioxygenase proteins, including the MAX4 gene product from Arabidopsis (Sorefan et al., 2003Go).



View larger version (7K):
[in this window]
[in a new window]
  		Figure 2. Structure of the RMS1 Gene, Showing the Relative Positions of Mutations in Various rms1 Alleles.

Exons are represented by shaded boxes and introns by black lines.

 
RMS1 Expression in the Wild Type
RMS1 transcript level was monitored in wild-type tissues by real-time RT-PCR (Figure 3) and by RNA gel blot analysis (data not shown). The highest levels of RMS1 transcript were observed in root tissue. In comparison with root tissue, RMS1 transcript levels were ~10-fold lower in epicotyl tissue and were ~100-fold lower in internode tissue. Very low RMS1 transcript levels were observed in leaves and the shoot tip (Figure 3). Epicotyl tissue, which is the stem tissue between the cotyledons and the first node, was included in several subsequent analyses because previous epicotyl interstock grafting studies demonstrated that RMS1 action in this tissue can suppress bud outgrowth in mutant rms1 scions (Foo et al., 2001Go).



View larger version (20K):
[in this window]
[in a new window]
  		Figure 3. RMS1 Gene Expression in Different Tissues of Wild-Type Plants.

 
RMS1 Expression Is Altered in rms Mutant Plants
RMS1 expression in the epicotyl of wild-type and various rms single and double mutant seedlings was monitored by real-time RT-PCR (Figure 4). RMS1 transcript level is expressed relative to the level observed in wild-type epicotyl tissue, which is given the value of 1. Epicotyls from mutant rms3, rms4, and rms5 plants accumulated RMS1 transcript to levels up to three orders of magnitude greater than wild-type plants (Figure 4; see also Figures 7 and 8). Similarly, in rms1-10 and rms1-11 mutant plants, which contain a single base pair change in the RMS1 sequence (Figure 2), rms1 transcript levels were increased up to two orders of magnitude compared with their respective wild-type progenitors (data not shown). Upregulation of RMS1 expression in these rms mutants suggests that RMS1 transcript levels may be under feedback control. In contrast with other rms mutants, rms2 plants exhibited a small but reproducible reduction in RMS1 expression compared with the wild type (Figure 4; see also Figures 7 and 8). Similar patterns of RMS1 expression were observed in independent rms mutant alleles derived from different genetic backgrounds (data not shown).



View larger version (16K):
[in this window]
[in a new window]
  		Figure 4. RMS1 Gene Expression in Epicotyl of Wild-Type, Single rms Mutant (rms2 to rms5), and Double rms Mutant (rms2 rms3 to rms2 rms5) Plants.

 


View larger version (42K):
[in this window]
[in a new window]
  		Figure 7. RMS1 Expression in Internode 4 of Intact or Decapitated Wild-Type, rms2, rms3, and rms4 Plants 12 h after Treatment with 0 or 3000 mg·L–1 IAA.

 


View larger version (20K):
[in this window]
[in a new window]
  		Figure 8. RMS1 Expression in Internode 4 of Wild-Type, rms2, rms3, and rms4 Plants 24 h after Intact Plants Were Treated with Ring of 0 or 3000 mg·L–1 of TIBA in Lanolin around Internode 5 or Decapitated in Internode 5.

 
To test whether feedback upregulation of RMS1 gene expression in rms3, rms4, and rms5 plants requires RMS2, RMS1 expression was monitored in rms2 rms3, rms2 rms4, and rms2 rms5 double mutants. RMS1 transcript levels in rms2 double mutant seedlings were substantially reduced when compared with the respective rms3, rms4, and rms5 single mutants (Figure 4), consistent with a role for RMS2 in the feedback regulation of RMS1 expression. However, because RMS1 transcript levels in these rms2 double mutant seedlings were still much higher than observed in wild-type and rms2 mutant seedlings, rms2 clearly does not completely prevent feedback upregulation of RMS1 expression.

RMS1 Regulation by a Mobile Feedback Signal
The elevated RMS1 transcript levels in several rms mutants (e.g., Figure 4) are consistent with physiological studies that indicate that the RMS1 branching inhibitor may be stimulated by a feedback signal (Figure 1; Beveridge et al., 1997aGo; Beveridge, 2000Go). The postulated feedback signal (Figure 1) is mobile and appears to move in the direction of shoot to root. If rms4 mutant plants exhibit elevated RMS1 expression because of altered production of a shoot-to-root feedback signal, we would expect that RMS1 expression in wild-type rootstock tissue could be influenced by grafting to rms4 scions. In pea, grafts are generally performed in the epicotyl; a portion of epicotyl remains in both scion and rootstock. By monitoring RMS1 expression in the rootstock epicotyl of reciprocal grafts between rms4 and wild-type plants, we tested the possibility that rms4 scions produce altered levels of a shoot-derived feedback signal (Figure 5). The branching phenotype of these grafts was consistent with previous reports that the genotype of the rootstock has little effect on branching in the scion of these genotypes (Figure 5A; Beveridge et al., 1996Go).



View larger version (14K):
[in this window]
[in a new window]
  		Figure 5. Branching Phenotype and RMS1 Gene Expression in Reciprocal Grafts between Wild-Type and rms4 Plants 46 d after Grafting.

(A) Ratio of total lateral length to main stem length (n = 10 to 26).

(B) RMS1 expression in rootstock epicotyl expressed relative to the wild-type self-grafts.

 
As observed in the epicotyl of intact plants (Figure 4), RMS1 transcript levels were elevated more than 1000-fold in the rootstock epicotyl of rms4 self-grafted plants relative to comparable wild-type tissue (Figure 5B). This high level of RMS1 expression in rms4 rootstocks was greatly reduced by grafting to wild-type scions (Figure 5B). Conversely, wild-type rootstocks grafted to rms4 scions exhibited higher RMS1 expression than those of wild-type self-grafts. This influence of scion genotype on RMS1 expression in rootstock of grafted plants is consistent with rms4 scions producing altered levels of a feedback signal.

It is clear that the RMS4 gene also acts in rootstocks to suppress RMS1 expression. Although RMS1 expression is upregulated in the epicotyls of wild-type rootstocks grafted to rms4 scions, the level of expression was still substantially less than observed with rms4 self-grafts (Figure 5A). In addition, grafting to wild-type scions fails to completely suppress RMS1 expression in rms4 rootstock epicotyls, indicating that mutation of the RMS4 gene in rootstock influences RMS1 expression in that tissue.

Auxin Regulation of RMS1 Expression
Several approaches were taken to investigate the interaction between IAA and RMS1. Temporal and spatial changes in RMS1 expression and endogenous IAA levels were monitored in stem tissue after decapitation and replacement of the apex with exogenous IAA (Figure 6, Table 1). Previous studies have shown that application of similar doses of IAA to decapitated wild-type plants can substantially reduce branching (Beveridge et al., 2000Go; data not shown). Decapitation resulted in a substantial decrease in RMS1 expression ranging several orders of magnitude. After decapitation, the first decrease in RMS1 expression was observed in the uppermost internode (internode 5) at 3 h, at 6 h in internode 4, and at 12 h in epicotyl tissue, which is at the base of the shoot (Figure 6). IAA application to the cut stump prevented the decline in RMS1 transcript level in all tissues tested over the 12-h time course. Moreover, this IAA treatment was correlated with a small increase in RMS1 expression at some time points, relative to intact controls.



View larger version (15K):
[in this window]
[in a new window]
  		Figure 6. RMS1 Expression in Wild-Type Plants over the Course of a Decapitation Experiment.

RMS1 gene expression at different times in intact plants or plants decapitated in internode 5 treated with 0, 500, or 3000 mg·L–1 IAA to the decapitated stump; internode 5 (A), internode 4 (B), and epicotyl (C). Values are average ± SE of two or three pools.

 

View this table:
[in this window]
[in a new window]
  		Table 1. IAA Level in Wild-Type Plants after Decapitation

 
IAA levels were monitored in replicate tissue pools from the same experiment described above (Table 1). IAA levels in the uppermost internode fell by ~40% within 3 h of decapitation (Table 1), the same time period that an ~90% drop in RMS1 expression was observed for this tissue (Figure 6A). An ~600% increase in stem IAA levels was recorded in decapitated plants treated with 500 mg L–1 IAA for 3 h (Table 1) and was accompanied by a doubling in RMS1 transcript levels in this tissue (Figure 6A). Relatively small decreases in stem IAA content therefore appear associated with large decreases in RMS1 transcript level, whereas large increases in stem IAA content are associated with relatively small increases in RMS1 transcript level.

The possibility that IAA regulation of RMS1 transcript level is specific to decapitated plants was investigated by monitoring the level of the RMS1 transcript in internode 4 of intact plants after treatment with exogenous IAA or 2,3,5-triiodobenzoic acid (TIBA), an auxin transport inhibitor, in a ring around internode 5. Exogenous IAA application to intact plants caused a small increase in RMS1 expression levels (Figure 7), whereas TIBA caused a massive decrease compared with intact untreated plants (Figure 8).

Auxin-Regulated RMS1 Expression in rms Mutants
The altered RMS1 expression in rms2, rms3, and rms4 plants (Figures 4 and 5) does not appear to be due to altered endogenous IAA levels or transport (Beveridge et al., 1994Go, 1996Go, 2000Go; Morris et al., 2001Go). As outlined previously, these mutants have reduced ability to suppress branching in response to exogenous IAA (Beveridge, 2000Go; Beveridge et al., 2000Go). Because IAA clearly stimulates RMS1 expression (Figures 6 to 8GoGo), the possibility that IAA regulation of RMS1 expression is disrupted in the rms mutants was explored.

The expression of the RMS1 gene in rms mutants was monitored in response to various treatments that alter auxin level in the stem, including decapitation and TIBA and auxin treatment (Figures 7 and 8). Both decapitation and TIBA dramatically reduced RMS1 expression in wild-type, rms3, and rms4 plants (Figures 7 and 8). Indeed, TIBA was as effective as decapitation in causing a two to three order of magnitude decrease in RMS1 expression in rms3 and rms4 plants (Figure 8). Decapitation or treatment with TIBA had no effect on RMS1 transcript levels in rms2 plants, which already exhibited very low RMS1 expression levels before treatment (Figures 7 and 8).

The decapitation-induced reduction in RMS1 transcript level in wild-type, rms3, and rms4 mutant plants could be largely or wholly prevented by replacement of the apical bud with exogenous IAA (Figure 7). IAA application to intact rms3 and rms4 plants resulted in a relatively small increase in RMS1 expression compared with the already high expression levels in intact mutant plants (Figure 7). IAA application to intact rms2 plants restored RMS1 expression to a level similar to that observed in untreated wild-type plants (Figure 7). Importantly, IAA application to wild-type plants never resulted in a more than 10-fold increase in expression compared with intact untreated wild-type plants (Figures 6 and 7), whereas up to 1000-fold increases in RMS1 transcript levels were observed in untreated rms3 and rms4 plants (Figures 4, 5, 7, and 8).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
METHODS
 REFERENCES
 
The expression studies presented here indicate that RMS1, which regulates levels of a novel hormone-like branching inhibitor, plays a central role in the signal cross talk that regulates shoot branching. We propose that IAA interacts with RMS1, at least partly, by stimulating expression of the RMS1 gene. We also provide molecular evidence for an additional IAA-independent feedback signal that stimulates expression of the RMS1 gene. Such an interaction between a feedback signal and the branching inhibitor is consistent with previous physiological studies in pea (Figure 1; Beveridge et al., 2003Go) and may play an important role in branching control in this species.

RMS1 Expression Is Feedback Regulated
The altered level of RMS1 expression in different rms mutants supports the notion that RMS1 expression levels are under feedback control. As discussed later, this feedback process is largely IAA independent. In four rms mutants, RMS1 transcript levels were greatly elevated in the stem and epicotyl (Figures 4, 5, 7, and 8; data not shown). Previous studies indicate that these mutants may have reduced levels of, or reduced response to, the RMS1 branching inhibitor. Mutants rms1 and rms5 are thought to block synthesis of the branching inhibitor (Foo et al., 2001Go; Morris et al., 2001Go), and rms3 and rms4 are thought to act after RMS1, perhaps by modulating response to the branching inhibitor (Figure 1; Beveridge, 2000Go). In these mutants, feedback upregulation of RMS1 expression may occur in response to actual or perceived reduction in the level of the branching inhibitor. The elevated RMS1 expression in these mutants does not appear to be an indirect effect of altered development because apart from increased bud outgrowth, all plants and harvested tissues were at similar developmental stages (Beveridge et al., 1996Go). Positive feedback is a common feature in the regulation of the synthesis of many plant hormones, including gibberellin and ethylene (Olszewski et al., 2002Go; Wang et al., 2002Go). Feedback regulation of RMS1 activity may offer an effective means to maintain branching homeostasis in response to changing environmental and endogenous factors.

Grafting studies reveal that feedback regulation of RMS1 expression occurs, at least partly, via transmission of a mobile feedback signal produced in shoot and rootstock. In reciprocal grafts between wild-type and rms4 plants, RMS1 expression in the rootstock was influenced by the genotype of the scion, indicating that a mobile signal produced in shoot tissue can regulate activity of the RMS1 gene in the rootstock (Figure 5B). The signal does not appear to be the RMS1 transcript itself because no RMS1 transcript could be detected in null allele rms1-1 rootstocks grafted to wild-type scions (data not shown).

Grafting studies have also demonstrated that RMS4 acts in shoot and rootstock tissue to regulate RMS1 (Figure 5B). Previous hypotheses for RMS4 action were based solely on the shoot phenotype in reciprocal grafting studies and predicted that RMS4 acted in or near axillary buds to suppress their outgrowth, possibly by influencing response to the branching inhibitor (Beveridge et al., 1996Go, 1997aGo). By monitoring RMS1 gene expression in grafted plants, we have shown that RMS4 may indeed control response to the branching inhibitor but acts in both scion and rootstock tissue. Epicotyls of rms4 self-grafts accumulated substantially more RMS1 transcript than wild-type rootstocks grafted to rms4 scions (Figure 5B). Similarly, rms4 rootstocks grafted with wild-type scions accumulated more RMS1 transcript than rootstocks of wild-type self-grafts. Thus, RMS4-mediated feedback regulation of RMS1 expression may operate in the scion and rootstock.

Confirmation that changes in RMS1 transcript level reflect changes in the level of the mobile branching inhibitor awaits identification and characterization of this novel signal. However, in pea, there is some evidence that RMS1 transcript levels may correlate with branching inhibitor levels. As might be expected if elevated RMS1 expression was associated with increased levels of the branching inhibitor, mutant rms3 and rms4 rootstocks are more effective than wild-type rootstocks at inhibiting branching when grafted to other rms scions (Beveridge et al., 1996Go, 1997bGo; Morris et al., 2001Go). Similarly, under short days where branching is enhanced in all genotypes, branching in cv Térèse (wild type) scions can be inhibited by grafting to rms4 rootstocks (E. Harding, K.E. Ng, and C.A. Beveridge, unpublished data). By contrast, transgenic Arabidopsis plants overexpressing MAX4 do not appear to display less shoot branching than wild-type plants, possibly because of differences in the regulation of the levels of, or response to, the branching inhibitor in pea and Arabidopsis (Sorefan et al., 2003Go). Alternatively, perhaps regulation of the branching inhibitor pathway occurs at multiple enzymatic steps.

Several pieces of evidence indicate that the feedback signal that activates RMS1 expression may be RMS2 dependant. Mutant rms2 seedlings display reduced RMS1 transcript levels in the stem compared with wild-type plants (Figures 4, 7, and 8). Furthermore, the ability of rms3, rms4, and rms5 mutations to elevate RMS1 expression is greatly diminished by the rms2 mutation (Figure 4). Nevertheless, RMS1 expression levels are not completely repressed in rms2 rms3, rms2 rms4, and rms2 rms5 double mutant plants, indicating that rms2 mutants may retain some RMS2 function or that factors independent of RMS2 may also upregulate RMS1 expression in rms3, rms4, and rms5 mutant plants. As discussed later, this effect of rms2 on suppression of RMS1 expression does not appear to be mediated by IAA.

Additional support for the hypothesis that RMS1 and RMS2 act on different pathways comes from the discovery that the rms1-1 mutation is caused by a genomic deletion and is effectively null. The strongly additive phenotype of the rms1-1 rms2-2 double mutant (Beveridge et al., 1997bGo) cannot be the result of leaky mutations in the same biosynthetic pathway and is likely to be due to RMS1 and RMS2 acting independently to influence branching. By contrast, rms1 rms5 double mutants exhibit only a weakly transgressive phenotype (Morris et al., 2001Go), consistent with these genes acting on the same biosynthetic pathway.

As outlined previously, analysis of xylem sap from reciprocal grafts between rms mutants and the wild type demonstrates that cytokinin export from the roots is regulated by a shoot-derived signal (Figure 1; Beveridge et al., 1997aGo; Beveridge, 2000Go). The low xylem sap cytokinin levels observed in rms1, rms3, and rms4 plants are mediated by a shoot-derived signal that is probably not IAA and may be influenced by RMS2 (Beveridge et al., 1997bGo; Beveridge, 2000Go). Preliminary studies indicate that a direct interaction between cytokinin level and RMS1 expression is unlikely because injection of the synthetic cytokinin, benzyl adenine, failed to suppress RMS1 expression in rms4 plants (data not shown). Considering the parallels between the downregulation of xylem sap cytokinin level and upregulation of expression of the RMS1 gene, an attractive hypothesis is that RMS2 regulates both processes via a common feedback signal.

Auxin Regulation of RMS1
In pea, the acropetally moving branching inhibitor controlled by RMS1 is required for IAA inhibition of branching (Beveridge et al., 2000Go; Foo et al., 2001Go). Here, we demonstrate that in wild-type pea, changes in IAA levels are associated with changes in RMS1 transcript levels in the stem. In wild-type plants, treatments that reduce IAA level, such as decapitation or auxin transport inhibition, also result in falls in RMS1 transcript levels in the stem (Figures 6 to 8GoGo). Replacement of the apex with exogenous IAA prevents falls in stem IAA level and maintains intact levels of RMS1 gene expression (Figures 6 and 7). Treatment of isolated stem segments with IAA also increases RMS1 expression (data not shown). Because decapitation and TIBA may affect the levels and/or transport of other long-distance signals (Napoli et al., 1999Go; Geldner et al., 2001Go), the small increase in RMS1 expression in intact plants in response to IAA was also an important observation (Figure 7).

Recent studies in Arabidopsis have revealed that targeted protein degradation via an SCF complex (SKP1, Cullin/CDC53, F-box protein) is an important component of auxin-regulated gene expression (Kepinski and Leyser, 2002Go). In Arabidopsis, mutations in several members and targets of this auxin-regulated protein degradation machinery results in increased shoot branching. The genes involved include AXR1, a potential regulator of the SCF complex, and IAA28 and AXR3, two genes that may regulate auxin-responsive gene expression (Leyser et al., 1996Go; Stirnberg et al., 1999Go; Ouellet et al., 2001Go; Rogg et al., 2001Go; del Pozo et al., 2002Go). Double mutant studies in Arabidopsis indicate that AXR3 may play a role in the max4 mutant phenotype (Sorefan et al., 2003Go); additional studies are required to dissect which of the many proteins involved in auxin response specifically influence RMS1 expression levels in pea.

It must be noted that the regulation of RMS1 in pea may differ from the regulation of its ortholog, MAX4, in Arabidopsis. ß-Glucuronidase reporter studies in Arabidopsis indicate that IAA modulation of MAX4 expression is restricted to root tips exposed to exogenous IAA and may occur over a longer time frame than that observed in pea (Sorefan et al., 2003Go). This has led to the hypothesis that IAA may interact with MAX4 posttranscriptionally (Sorefan et al., 2003Go) or may act on other genes in the same biosynthetic pathway as MAX4.

Cross Talk: Auxin and a Feedback Signal Control RMS1
Effective regulation of axillary bud outgrowth requires the capacity to respond to decapitation by rapid reestablishment of a growing shoot tip. In addition, the branching control system needs a fine-tuning component enabling intact plants to modify branching in response to variation in the environment and endogenous factors. RMS1 may represent a point of cross talk in branching control. The control of the axillary branching by IAA produced in growing apical tip may be mediated, in part, by the enhancement of RMS1 expression. In addition, the RMS genes appear to regulate an IAA-independent feedback signal that enhances RMS1 expression. These interactions among IAA, RMS1, and a feedback signal provide the mechanism for a rapid response to decapitation, which is essential for plant survival, as well as a feedback mechanism for homeostatic control of shoot branching in intact plants. In this way, axillary buds of intact plants may be released from inhibition without necessitating a drop in endogenous auxin level. Such a specific branching control mechanism would not, like changes in IAA level, adversely influence other developmental processes, such as stem elongation.

The reduced response to IAA observed in rms3 and rms4 plants (Beveridge, 2000Go) does not appear to be due to disruption of IAA-regulated RMS1 expression. Although rms3 and rms4 plants exhibit very high levels of RMS1 transcript (Figures 4, 5, 7, and 8), this is not due to altered endogenous IAA levels or transport (Beveridge et al., 1996Go; Beveridge, 2000Go). In addition, rms3 and rms4 mutations do not prevent IAA regulation of RMS1 expression (Figures 7 and 8). The high RMS1 transcript levels observed in rms3 and rms4 plants could be dramatically reduced by treatments that lower IAA levels in the stem and were largely restored by exogenous IAA treatment (Figures 7 and 8). The IAA regulation of RMS1 transcript level in rms3 and rms4 plants (Figures 7 and 8), together with their relatively nonpleiotropic phenotypes compared with plants such as axr1 and axr3 that contain lesions in general auxin response components (Beveridge et al., 1996Go; Stirnberg et al., 1999Go; Ouellet et al., 2001Go), indicates that RMS3 and RMS4 are not primarily involved in IAA response. The failure of exogenous IAA to inhibit bud outgrowth in rms3 and rms4 plants (Beveridge, 2000Go), despite its ability to upregulate RMS1 transcript levels in these mutants, supports the hypothesis in Figure 1 that RMS3 and RMS4 act downstream of RMS1, possibly affecting response to the branching inhibitor.

Similarly, RMS2 does not appear to act primarily via disruption of IAA-mediated regulation of RMS1 expression. Treatment with exogenous IAA elevated RMS1 transcript levels in intact and decapitated rms2 plants (Figure 7). It is possible that the failure of decapitation or TIBA treatment to repress RMS1 expression in rms2 seedlings (Figures 7 and 8) is a consequence of the already reduced RMS1 transcript levels or that RMS2 has a more direct effect on RMS1 expression. The reduced RMS1 expression in intact rms2 plants is probably not due to reduced IAA level or polar transport because rms2 mutant shoots have up to a fivefold increase in IAA content (Beveridge et al., 1994Go) and near wild-type or elevated polar IAA transport (Beveridge et al., 2000Go).

The weight of evidence suggests that a mobile signal in addition to IAA is involved in regulating RMS1 expression. However, it could be argued that because the endogenous IAA and gene expression analyses presented here were based on whole tissues, they may not fully reflect changes in IAA content or RMS1 transcript levels in specific cells. The strongest evidence for IAA-independent regulation of RMS1 expression in the plant parts we have tested is the extremely high levels of RMS1 expression seen in rms3, rms4, and rms5 plants (Figures 4, 5, 7, and 8). This very high level of RMS1 expression could not be achieved in comparable wild-type plants by exogenous IAA treatments (Figures 6 and 7). Wild-type plants treated with exogenous IAA displayed clear increases in stem auxin content and reached a consistently higher level of RMS1 expression than untreated wild-type plants (Figures 6 and 7, Table 1). However, the levels of RMS1 expression reached in IAA-treated wild-type plants are severalfold less than those observed in untreated rms3, rms4, and rms5 plants (Figures 4, 5, 7, and 8). Given that RMS1 expression in rms3 and rms4 plants shows major changes in response to treatments that cause depleted or increased IAA level (Figures 7 and 8), it is clear that factors in addition to IAA level may cause the very high levels of RMS1 transcript in intact rms3 and rms4 plants.

Here, we have provided evidence that IAA and an as yet unidentified mobile feedback signal independently regulate RMS1 transcript level in pea. Steady state RMS1 transcript level may be a point for integration of various processes that influence bud outgrowth. Future studies will focus on the identification of the branching inhibitor and the feedback signal and characterization of their physiological functions.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 METHODS
REFERENCES
 
Plant Materials and Growth Conditions
The rms branching lines K524 (rms2-1), K487 (rms3-1), and K164 (rms4-1) were derived from cv Torsdag (Arumingtyas et al., 1992Go). rms5-3 on a Torsdag background was derived from the cross of Wt15241 (rms5-3) and Torsdag after backcrossing and was kindly provided by S. Morris and I.C. Murfet. The double mutants rms2-1 rms4-1 (Murfet and Symons, 2000bGo) and rms2-1 rms5-3 (Murfet and Symons, 2000aGo) display transgressive phenotypes compared with either single parent. The genotype rms2-1 rms3-1 was generated by following a cross between K524 (rms2-1) and K487 (rms3-1), but transgressive plants were not identifiable in the F2 (Murfet and Symons, 2000aGo). Seed from F2 plants that displayed a branching phenotype and the pod shape typical of rms2-1 plants were supplied by I.C. Murfet and grafted to wild-type rootstocks. Branching is inhibited in rms2 scions grafted to wild-type rootstocks but not in rms3-1 scions grafted to the wild type (Beveridge et al., 1996Go). We selected F3 scions that branched and confirmed them as genotype rms2-1 rms3-1 by backcrosses to both single mutant parents. The origin and progenitors of rms1 mutant lines are given by Symons and Murfet (1997)Go and Rameau et al. (1997)Go, with the exception of lines TB-703 (rms1-12) and TB-1525 (rms1-13) that were derived from Térèse by C. Rameau using ethyl methanesulfonate mutagenesis (personal communication).

Unless otherwise stated, wild-type Torsdag plants with five leaves expanded (counting acropetally from the cotyledonary node as zero) were selected for experimentation after growing at one per pot in 15-cm slim line pots as described by Morris et al. (2001)Go.

Epicotyl-to-epicotyl wedge grafts were performed on 6-d-old seedlings planted two per two-liter pot as described by Beveridge et al. (1994)Go. Forty-six days after grafting, total lateral length (the sum of all lateral branches arising from the main shoot) and main stem length were measured and rootstock epicotyl tissue (~5 mm) collected.

Auxin Treatments
Plants were left intact, treated with a ring of 0 or 3000 mg·L–1 IAA in lanolin around internode 5 or decapitated in internode 5 and treated with 0, 500, or 3000 mg·L–1 IAA in lanolin at the cut stump (Beveridge et al., 2000Go). In other studies, intact plants were treated with a ring of 0 or 5000 mg·L–1 TIBA in lanolin around internode 5.

Endogenous IAA Analysis
IAA was extracted from frozen tissue (0.03 to 0.5 g) as described by Morris et al. (2001)Go, based on a modification of Batge et al. (1999)Go. [13C6]IAA (Cambridge Isotope Laboratories, Hampshire, UK) internal standard was added at 100 ng·g–1 for intact samples, 50 ng·g–1 for decapitated samples, and 200 ng·g–1 for IAA-treated internode samples. IAA analysis was performed by GC-MS-selected ion monitoring as described by Ross (1998)Go. Endogenous IAA and [13C6]IAA were quantified as described by Morris et al. (2001)Go, and IAA is expressed as ng·g–1 of fresh weight.

Sequencing of the RMS1 Gene
Overlapping fragments covering the full RMS1 coding sequence were amplified and sequenced from genomic DNA of various rms1 mutant lines and their corresponding wild-type progenitors. Sequence alignment and single nucleotide polymorphism detection were performed using the software Genalys, developed at the Centre National de Recherche en Génomique, Evry, France (http://software.cng.fr).

RMS1 Expression Analysis
For all data shown, results were verified in at least one independent experiment. Tissue segments were collected into one or more pools of 6 to 12 segments.

Total RNA was extracted using a modification of the hot-phenol method (Kreig, 1996Go), and cDNA was synthesized from 4.5 µg of total RNA with Superscript II reverse transcriptase (Invitrogen Life Technologies, Carlsbad, CA). PCR reactions were performed in duplicate with 10% of each cDNA sample using TaqMan Universal PCR master mix (Applied Biosystems, Foster City, CA) or Thermostart QPCR master mix with ROX (ABgene, Surrey, UK) in an ABI Prism 7700 cycler (Applied Biosystems). PCR reactions were performed with 200 to 300 nM each primer and 125 nM probe in a 25-µL reaction under the following conditions: 50°C for 2 min and 95°C for 10 min followed by 45 two-step cycles at 95°C for 15 s and 60°C for 1 min. Gene-specific primers and Taqman probes were designed for RMS1 and actin (as a control) using Primer Express 1.5 software (Applied Biosystems) with one of the primer pairs designed over an RNA splice junction. In many experiments, the level of 18s rRNA was also monitored and was consistent with results obtained with actin (data not shown).

Sequences are as follows: RMS1 primers RMS1 F (5'-AAGGAGCTGTGCCCTCAGAA-3') and RMS1 R (5'-ATTATGGAGATCACCACACCATCA-3'), RMS1 probe (5'-CATTCTTTGTGCCTCGACCAGGAGCA-3'), actin primers ACT F (5'-GTGTCTGGATTGGAGGATCAATC-3') and ACT R (5'-GGCCACGCTCATCATATTCA-3'), and actin probe (5'-CACCTTCCAGCAGATGTGGATATCTAAGGC-3'). The cycle at which the level of fluorescence reaches a preset threshold (CT) is proportional to the amount of target RNA present in each sample. The average CT value was calculated for duplicate PCR reactions. A normalized value for RMS1 expression was obtained with the formula {Delta}CT = average CT(RMS1) – average CT(Actin). For Figure 6, an average {Delta}CT was calculated for biological replicates. Within an experiment, one treatment or genotype was nominated as a control, and relative RMS1 expression of each sample was calculated using the equation 2–(average {Delta}CT(treatment) – average {Delta}CT(control)).

Sequence data from this article have been deposited with the EMBL/GenBank data libraries under accession numbers AY557341 for RMS1 (Térèse haplotype) and AY557342 (Raman and Borek haplotype).


   Acknowledgments
 
We thank Colin Turnbull for helpful discussions and comments on the manuscript, John Ross for GC-MS-selected ion monitoring quantitation of IAA and helpful discussions, Chuong Ngo for technical assistance, Patrick Grillot for glasshouse management in Versailles, and Ian Murfet, Greg Symons, and Suzanne Morris for providing seed material. We also thank Jim Weller, Peter Davies, and reviewers for feedback and comments on the manuscript. This work was supported by the Union Nationale Interprofessionnelle des Plantes riches en proteine and the Australian Research Council. E.F. was supported by an Australian postgraduate award.


   Footnotes
 
1 Current address: School of Plant Sciences, University of Tasmania, Private Bag 55, Hobart, Tasmania, 7001, Australia. Back

2 Current address: Unité Mixte de Recherche GenHort, Institut National de la Recherche Agronomique, 42 rue G. Morel, 49071, Beaucouze, France. Back

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantcell.org) is: Christine Anne Beveridge (c.beveridge{at}botany.uq.edu.au).

Article, publication date, and citation information can be found at www.plantcell.org/cgi/doi/10.1105/tpc.104.026716.

Received August 23, 2004; accepted November 12, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 METHODS
 REFERENCES
 
Arumingtyas, E.L., Floyd, R.S., Gregory, M.J., and Murfet, I.C. (1992). Branching in Pisum: Inheritence and allelism test with 17 ramosus mutants. Pisum Genet. 24, 17–31.

Bangerth, F. (1994). Response of cytokinin concentration in the xylem exudate of bean (Phaseolus vulgaris L.) plants to decapitation and auxin treatment and relationship to apical dominance. Planta 194, 439–442.

Batge, S.L., Ross, J.R., and Reid, J.B. (1999). Abscisic acid levels in seeds of the gibberellin-deficent mutant lh-2 of pea (Pisum sativum). Physiol. Plant. 105, 485–490.[CrossRef]

Beveridge, C.A. (2000). Long-distance signalling and a mutational analysis of branching in pea. Plant Growth Regul. 32, 193–203.[CrossRef]

Beveridge, C.A., Murfet, I.C., Kerhoas, L., Sotta, B., Miginiac, E., and Rameau, C. (1997a). The shoot controls zeatin riboside export from pea roots: Evidence from the branching mutant rms4. Plant J. 11, 339–345.[CrossRef]

Beveridge, C.A., Ross, J.J., and Murfet, I.C. (1994). Branching mutant rms-2 in Pisum sativum. Grafting studies and endogenous indole-3-acetic acid levels. Plant Physiol. 104, 953–959.[Abstract]

Beveridge, C.A., Ross, J.J., and Murfet, I.C. (1996). Branching in pea. Action of genes Rms3 and Rms4. Plant Physiol. 110, 859–865.[Abstract]

Beveridge, C.A., Symons, G.M., Murfet, I.C., Ross, J.J., and Rameau, C. (1997b). The rms1 mutant of pea has elevated indole-3-acetic acid levels and reduced root-sap zeatin riboside content but increased branching controlled by graft-transmissible signals. Plant Physiol. 115, 1251–1258.[ISI]

Beveridge, C.A., Symons, G.M., and Turnbull, C.G.N. (2000). Auxin inhibition of decapitation-induced branching is dependent on graft-transmissible signals regulated by genes Rms1 and Rms2. Plant Physiol. 123, 689–697.[Abstract/Free Full Text]

Beveridge, C.A., Weller, J.L., Singer, S.R., and Hofer, J.M.I. (2003). Axillary meristem development. Budding relationships between networks controlling flowering, branching and photoperiod responsiveness. Plant Physiol. 131, 1–8.[Free Full Text]

Booker, J., Auldridge, M., Wills, S., McCarty, D., Klee, H., and Leyser, O. (2004). MAX3/CCD7 is a carotenoid cleavage dioxygenase required for the synthesis of a novel plant signalling molecule. Curr. Biol. 14, 1232–1238.[CrossRef][ISI][Medline]

Booker, J., Sieberer, T., Wright, W., Williamson, L., Willett, B., Stirnberg, P., Turnbull, C., Srinivasan, M., Goddard, P., and Leyser, O. (2005). The MAX1 gene of Arabidopsis encodes a putative cytochrome P450 that acts downstream of MAX3 and MAX4 in the synthesis of a novel branch-inhibiting hormone. Dev. Cell, in press.

Cline, M.G. (1996). Exogenous auxin effects on lateral bud outgrowth in decapitated shoots. Ann. Bot. 78, 255–266.[Abstract/Free Full Text]

Dodd, I.C., Ngo, C., Turnbull, C.G.N., and Beveridge, C.A. (2004). Effects of nitrogen supply on xylem cytokinin delivery, transpiration and leaf expansion of pea genotypes differing in xylem cytokinin concentration. Funct. Plant Biol. 31, 903–911.[CrossRef]

del Pozo, J.C., Dharmasiri, S., Hellman, H., Walker, L., Gray, W.M., and Estelle, M. (2002). AXR1-ECR1–dependant conjugation of RUB1 to the Arabidopsis cullin AtCUL1 is required for auxin response. Plant Cell 14, 421–437.[Abstract/Free Full Text]

Foo, E., Turnbull, C., and Beveridge, C.A. (2001). Long distance signalling and the control of branching in the rms1 mutant of pea. Plant Physiol. 126, 203–209.[Abstract/Free Full Text]

Geldner, N., Friml, J., Stierhof, Y.-D., Jurgens, G., and Palme, K. (2001). Auxin transport inhibitors block PIN1 cycling and vesicle trafficking. Nature 413, 425–428.[CrossRef][Medline]

Gray, W.M., Kepinski, S., Rouse, D., Leyser, O., and Estelle, M. (2001). Auxin regulates SCF (TIR1)-dependent degradation of AUX/IAA proteins. Nature 414, 271–276.[CrossRef][Medline]

Hall, S.M., and Hillman, J. (1975). Correlative inhibition of lateral bud growth in Phaseolus vulgaris L. Timing of bud growth following decapitation. Planta 123, 137–143.[CrossRef][ISI]

Iuchi, S., Kobayashi, M., Taji, T., Naramoto, M., Seki, M., Kato, T., Tabata, S., Kakubari, Y., Yamaguchi-Shinozaki, K., and Shinozaki, K. (2001). Regulation of drought tolerance by gene manipulation of 9-cis-epoxycarotenoid dioxygenase, a key enzyme in abscisic acid biosynthesis in Arabidopsis. Plant J. 27, 325–333.[CrossRef][ISI][Medline]

Kepinski, S., and Leyser, O. (2002). Ubiquitination and auxin signalling: A degrading story. Plant Cell 14 (suppl.), S81–S95.[Free Full Text]

Kreig, P.A. (1996). A Laboratory Guide to RNA Isolation, Analysis and Synthesis. (New York: John Wiley & Sons).

Leyser, H.M.O., Pickett, F.B., Dharmasiri, S., and Estelle, M. (1996). Mutations in the AXR3 gene of Arabidopsis result in altered auxin response including ectopic expression from the SAUR-AC1 promoter. Plant J. 10, 403–413.[CrossRef][ISI][Medline]

Leyser, O. (2003). Regulation of shoot branching by auxin. Trends Plant Sci. 8, 541–545.[CrossRef][ISI][Medline]

Li, C.J., Guevara, E., Gerrera, J., and Bangerth, F. (1995). Effect of apex excision and replacement by 1-naphthylacetic acid on cytokinin concentration and apical dominance in pea plants. Physiol. Plant. 94, 465–469.[CrossRef]

Morris, D.A. (1977). Transport of exogenous auxin in a two-branched dwarf pea seedling. Planta 36, 91–96.

Morris, S.E., Turnbull, C.G.N., Murfet, I.C., and Beveridge, C.A. (2001). Mutational analysis of branching in pea. Evidence that Rms1 and Rms5 regulate the same novel signal. Plant Physiol. 126, 1205–1213.[Abstract/Free Full Text]

Murfet, I.C., and Symons, G.C. (2000a). Double mutant rms2 rms5 expresses a transgressive, profuse branching phenotype. Pisum Genet. 32, 33–38.

Murfet, I.C., and Symons, G.C. (2000b). The pea rms2-1 rms4-1 double mutant phenotype is transgressive. Pisum Genet. 32, 59–60.

Napoli, C.A. (1996). Highly branched phenotype of the petunia dad1-1 mutant is reversed by grafting. Plant Physiol. 111, 27–37.[Abstract]

Napoli, C.A., Beveridge, C.A., and Snowden, K.C. (1999). Reevaluating concepts of apical dominance and the control of axillary bud outgrowth. Curr. Top. Dev. Biol. 44, 127–169.<