Molecular Plant Advance Access originally published online on February 11, 2008
Molecular Plant 2008 1(2):388-400; doi:10.1093/mp/ssn007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Interactions between Axillary Branches of Arabidopsis
a Department of Biology, University of York, PO Box 373—Area 11,York YO10 5YW, UK
b Present address: Institute of Plant Sciences, University of Bern, CH-3013 Bern, Switzerland
1 To whom correspondence should be addressed. E-mail hmol1{at}york.ac.uk, fax 00–44–1904–328682, tel. 00–44–1904328680
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONRESULTSDISCUSSIONMETHODS |
|---|
Studies of apical dominance have benefited greatly from two-branch assays in pea and bean, in which the shoot system is trimmed back to leave only two active cotyledonary axillary branches. In these two-branch shoots, a large body of evidence shows that one actively growing branch is able to inhibit the growth of the other, prompting studies on the nature of the inhibitory signals, which are still poorly understood. Here, we describe the establishment of two-branch assays in Arabidopsis, using consecutive branches on the bolting stem. As with the classical studies in pea and bean, these consecutive branches are able to inhibit one another's growth. Not only can the upper branch inhibit the lower branch, but also the lower branch can inhibit the upper branch, illustrating the bi-directional action of the inhibitory signals. Using mutants, we show that the inhibition is partially dependent on the MAX pathway and that while the inhibition is clearly transmitted across the stem from the active to the inhibited branch, the vascular connectivity of the two branches is weak, and the MAX pathway is capable of acting unilaterally in the stem.
Key Words: shoot branching • auxin • MAX • vascularization
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODS |
|---|
The plant shoot system is formed by the action of the primary shoot apical meristem, which is established during embryogenesis. The final architecture of the shoot is highly variable and this is largely because, in addition to the primary shoot apical meristem, secondary meristems can arise in the axils of the leaves produced by the primary meristem. These axillary meristems have the same developmental potential as the primary meristem, but frequently they arrest after forming only a few unexpanded leaves and remain dormant as an axillary bud. It is variation in the subsequent reactivation and outgrowth of these buds that underlies much of the observed diversity in shoot system architecture.
The activity of axillary buds is affected by many factors. One important regulator is the primary shoot apex. If the apex is removed, dormant buds below it can activate. This phenomenon is known as apical dominance (Cline, 1997). In 1933, Thimann and Skoog demonstrated that exogenous auxin, applied to the stump of a decapitated shoot, is able to substitute for the apex in inhibiting bud outgrowth, and a large body of data supports a role for apically derived auxin in the inhibition of bud outgrowth in intact plants (Leyser, 2003). Auxin is synthesized in the young expanding leaves of the apical bud (Ljung et al., 2001) and is transported down the plant in the polar transport stream, in the xylem-associated parenchyma of the main stem. This process is dependent on auxin transporters, particularly those of the PIN family of auxin efflux carriers (Blakeslee et al., 2005; Paponov et al., 2005). Auxin transport in the polar transport stream is strictly basipetal, and auxin moving down the stem is not transported laterally and acropetally into the buds, indicating that it must inhibit bud growth indirectly (Hall and Hillman, 1975; Booker et al., 2003).
This paradox was noted very early in the study of apical dominance. Of particular importance was the development of the so-called two-branched pea or bean system (Snow, 1929). In this system, plants are decapitated above the cotyledonary node. This results in the activation of the cotyledonary axillary buds to produce a plant with two active branches, similar to the cartoons below each bar in Figure 1F. Although both branches usually activate, often, one continues to grow actively, while the other stops growing. In these plants, if the active shoot is removed, the inactive shoot frequently reactivates, indicating that its growth is inhibited by the dominant shoot. Thus, the inhibitory influence of the dominant shoot can be transmitted some considerable distance up the subordinate shoot to influence its activity (Snow, 1931).
|
An extension of this experiment was developed by Snow in 1937. Plants were first divided in half from the root system up through the basal part of the stem to a point above the cotyledonary node. The plants were then treated as for a classical two-branch bean. The plants were decapitated with consequent activation of the cotyledonary buds to produce branches. At this stage, the plants had a W configuration, similar to that illustrated below each bar in Figure 2F, with the two active apices forming the left and right ends of the W, connected by their stems, down to the cotyledonadry nodes, and then up through the split halves of the primary stem, to meet just below the stump of the decapitated primary shoot. One or both of the cotyledonary axillary branches were then excised just above the first leaves, and the outgrowth of axillary buds in the axils of these leaves was measured. Snow's results show that where only one branch was excised, new buds located on that side of the W in general did not activate, where as when both active branches were removed, the buds always activated. Thus, in this system, the inhibitory influence of the active branch must in some way be transmitted across the W to inhibit buds on the other side.
|
These experiments emphasized the indirect effect of auxin and the requirement for both basipetal and acropetal effects in auxin-mediated bud inhibition. Recent results have suggested at least two independent mechanisms through which auxin traveling basipetally in the main stem can acropetally inhibit bud activity. Firstly, auxin can down-regulate the synthesis of cytokinin both in the main stem and in the root, and, in Arabidopsis, this process is dependent on AXR1-mediated auxin signaling (Bangerth, 1994; Nordström et al., 2004; Tanaka et al., 2006). Cytokinin can move acropetally in the transpiration stream and can act directly in buds to promote their outgrowth (Cline, 1991). Thus, one mechanism for auxin action is to restrict cytokinin delivery to the bud.
The second proposed mechanism involves competition between the bud and the primary apex for auxin transport capacity in the stem (Bennett et al., 2006). This model is based on the assumptions that (1) for a branch to grow, it must be able to export auxin into the main stem, (2) auxin export from the bud into the main stem requires that the stem acts as an auxin sink, allowing canalization of auxin transport out of the bud, and (3) auxin synthesis in the bud is under feedback control, such that inability to export auxin restricts auxin production. This model is particularly cogent for early-stage buds, which are not connected to the vascular system of the shoot. Such vascular connections require canalization of auxin export from the bud, which is likely to require a strong auxin sink in the main stem. The model builds on the observations of Morris and colleagues, who demonstrated that branch activity correlates strongly with auxin transport activity in the branch (Morris, 1977; Morris and Johnson, 1990), and the observations of Sachs (1968, 1969, 1981), who demonstrated that vascular development triggered by a local auxin source would connect to an existing vascular strand only when the existing strand was not transporting substantial amounts of auxin. This was shown to be true both for explant systems and for the vascularization of axillary buds.
An interesting feature of this model is that it does not require movement of a signal from the stem into the bud, but rather relies on competition between the bud and the main stem for limited auxin transport capacity in the main stem. In this respect, it is similar to the auxin transport autoinhibition (ATA) model of Bangerth and co-workers (Li and Bangerth, 1999). However, a critical difference is that ATA predicts that increased auxin transport in the stem would increase bud inhibition, where as the canalization-based model predicts that increased transport capacity would result in bud activation. This is what is observed in the Arabidopsis max mutants. Mutations at the four MAX loci in Arabidopsis result in increased shoot branching, and bud growth resistant to apical auxin application (Stirnberg et al., 2002; Sorefan et al., 2003; Booker et al., 2004, 2005; Bennett et al., 2006). These phenotypes are associated with increased auxin transport capacity in the main stem (Bennett et al., 2006). Reduction in auxin transport capacity back to wild-type levels restores both wild-type branching and wild-type bud auxin response, suggesting that the increased auxin transport is the cause of these other phenotypes (Bennett et al., 2006). The max mutants have greatly increased expression of the auxin-responsive reporter DR5 in the main stem vasculature (Bennett et al., 2006), which, in combination with their auxin-resistant buds and increased expression of auxin transporter genes, suggests that more auxin is moving in the stems of max mutants than in wild-type plants. If all shoot branching is regulated by an auxin concentration sensing mechanism in the main stem, for example, through modulating cytokinin synthesis as described above, then one would expect max mutants to have reduced shoot branching. However, if the max mutants have increased branching because of canalization-based effects, then the paradox of high auxin-responsive gene expression, with auxin-resistant buds, is resolved. Consistent with this, double mutants between axr1, which is defective in auxin signaling, and max result in additive phenotypes (Bennett et al., 2006), suggesting that these genes act at least somewhat independently to regulate bud growth.
Thus, these recent developments point to two mechanisms by which basipetally moving auxin can affect bud growth acropetally. These discoveries provide a framework within which to investigate further the mechanism of communication between buds in two-branch systems. In order to take advantage of the impressive genetic and molecular resources available in Arabidopsis, we sought to establish a two-branch system in Arabidopsis and use it to test the role of the MAX pathway in mediating interactions between the branches. Here, we report our progress in this area.
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODS |
|---|
A Two-Branch Arabidopsis System
In Arabidopsis, the only axillary branches readily available for experimental manipulation are those in the nodes of cauline leaves, which are carried up on the elongating bolting stem. To determine whether branches in successive cauline nodes were able to influence one another's growth, as in the two-branched pea or bean, we established a two-branched Arabidopsis system. Stem segments from plants of the Columbia ecotype spanning two cauline nodes with their associated axillary buds (designated ‘top’ and ‘bottom’) were excised. Following excision of the two-node segments, the subsequent outgrowth of the branches derived from the two axillary buds was measured over an 11-d period. Examples are shown in Figure 1A. In general, only one of the two branches grew vigorously throughout the experiment. Thus, at day 11, the mean length of the longer branch was 49.02 mm ± 1.14 SE, whilst the mean length of the shorter branch was only 10.81 mm ± 1.62 SE (Figure 1D). These data can be expressed as the proportion of growth in the longest branch, to give a relative growth index (RGI), where equal growth would give an index of 0.5, and one branch completely and immediately dominating would give an index of 1. The mean RGI at day 11 for wild-type was 0.839 ± 0.019 SE, illustrating the dominance of one branch over the other (Figure 1D and 1E). The mean RGI at day 7 was lower, at 0.779 ± 0.019 SE, indicating that the degree of dominance increases over time (Figure 1C–1E). When the mean lengths for the top and bottom branches were compared, they were found to be surprisingly similar. For example, despite the high relative growth index, at day 11, the mean length of the top branch was 25.67 mm ± 2.87 SE and the bottom branch was 34.16 mm ± 3.094 SE (Figure 1G), indicating that there was a slight bias in favour of the bottom branch becoming dominant. Thus, our results are similar to the two-branched pea system, where both buds activate following removal of the main shoot, but, over time, one often becomes dominant and the other subordinate.
To determine whether this was due to the influence of one branch over the other, we tested the effect of removing one of the buds at day 0 (Figure 1B). The remaining branch grew to a mean length approximately double that of the branch in the equivalent position when both branches were left on the stem (Figure 1F and 1G), suggesting that the dominant shoot was indeed responsible for the inhibition of growth of the other shoot.
To determine whether the MAX pathway, and thus presumably auxin transport capacity in the stem, affected the ability of branches to influence one another in this two-branch system, we performed the same experiments described above for WT using the max4 mutant. All the effects described for wild-type were observed in the max4 mutant, but they were significantly weaker. For example, the RGI for the max4 was only 0.692 mm ± 0.019 SE at day 7 and it did not significantly change over the following 4 d, being 0.689 mm ± 0.021 SE at day 11, suggesting that both branches were actively growing during this period (Figure 1E). There was a noticeably stronger bias in max4 mutants as to which branch was longer. The bottom branch was usually longer than the top branch, so, for example, at 11 d, the mean length of the bottom branch was 43.97 mm ± 1.86 SE compared with only 28.97 mm ± 2.43 SE for the top branch (Figure 1G). As for wild-type, removal of one bud resulted in the remaining branch achieving a greater mean length than the branch in the equivalent position on explants from which neither bud was removed. The resultant increase in outgrowth was greater for the top branch than the bottom branch (Figure 1F and 1G). These results suggest that branch–branch communication is mediated in part, but not entirely, through MAX-dependent restriction of main stem auxin transport capacity.
A W Arabidopsis System
To investigate the properties of communication between successive branches in more detail, inspired by Snow's W assay, we adapted the assay developed for beans to Arabidopsis. Decapitated stem sections with two buds were isolated as described above. The stems were then split in half, leaving only a small piece of stem above the upper node connecting the two halves (Figure 2A and 2B). An important difference between this system and beans/peas is that the spiral phyllotaxy of the Arabidopsis shoot means that the W is highly asymmetric, with one short and one long central spur. As for the simple two-branch system, the growth of branches across this asymmetric W was analyzed, comparing explants where both buds were left to grow with those from which either the top or bottom bud was removed at the start of the experiment.
As for the simple two-branch system, there is evidence of one branch dominating the other in both genotypes, with this effect again being greater in WT than in the max4 mutant (Figure 2C and 2D). The RGI was still significantly above 0.5 for both WT and max4 (Figure 2E), although significantly lower than when the stems were intact (Figure 1E). Interestingly, the RGI for both genotypes reduced between day 7 and day 11, suggesting that the dominance of one branch over the other was weakening over time.
Both genotypes show bias as to which branch dominates, with the bottom branch being on average longer than the top branch, particularly at day 7 (Figure 2F and 2G). Furthermore, removal of the top bud had little (WT) or no (max4) effect on the mean length of the bottom branch, indicating that the ability of the top branch to influence the growth of the bottom branch was greatly reduced by the division of the stem (Figure 2G). In contrast, the top branch was on average significantly longer in both genotypes when the bottom bud was removed (Figure 2F and 2G), suggesting that even though the stem was split, there was still some communication between the two branches.
During these experiments, further evidence of communication became apparent through the development of accessory branches. Although the norm for Arabidopsis is for a single axillary bud, and thus a single branch, to form in each axil, additional axillary meristems can initiate adjacent to the main axillary bud, further up the petiole towards the leaf. The formation and activation of these so-called accessory meristems are rare in wild-type plants. We never observed accessory branches in the simple two-branch system described above in either genotype in this study. However, in the W system, excision of a bud often resulted in the development and outgrowth of an accessory bud in the denuded axil (Figure 2H). This suggests that when the stem in not split, an active branch in one axil can inhibit the development of accessory branches in the other axil, and this communication is reduced when the stem is divided. For both genotypes, accessories were produced with lower frequency in the bottom axil than the top axil (Figure 2H). If accessory formation is indeed due to reduced communication with the remaining branch, the strength of communication is the opposite of the situation with branch length (see above). For accessories, the bottom branch has a weaker influence in suppressing accessories at the top node than the top branch has on the bottom node. Interestingly, accessory branching was generally less frequent in the max4 mutant than in WT—again, the opposite of the situation for the branch length responses.
The Effect of Auxin
To assess whether auxin can mediate branch–branch communication in Arabidopsis, the top or bottom buds from two-branch stem segments were removed at day 0 and replaced by auxin, and the remaining branch was measured (Figure 3). In all cases—WT and max4, and top and bottom branch growth—there was an inhibitory effect on the growth of one branch when auxin was applied to the decapitated stump of the other. This effect was small but significant (ANOVA df = 1, F = 27.648). There was no significant difference in auxin response between the WT and max4, or between the top and bottom buds (ANOVA df = 1, F = 0.023, p = 0.88; df = 1, F = 0.58, p = 0.447).
|
Branch Vascularization
Many of the signals likely to mediate branch–branch communication, such as auxin and cytokinin, move in the vascularture or vascular-associated tissues. Thus, to investigate the possible anatomical pathways for branch–branch communication in more detail, we investigated the vascularization in MAX4 and max4 branches, using serial sections of wax-embedded tissue. We used plants mutant at the LFY locus to provide a greater number of nodes for sectioning. In our growth conditions, the basal region of the stem of axillary branches was found to have six vascular bundles, whilst the primary stem of the inflorescence had eight (Figure 4). By comparing successive transverse sections moving down the stem through a node, two alternative patterns were identified for vascularization of the axillary branches (Figure 4A). The first pattern, which we observed in both MAX4 and max4 plants, is essentially the same as that already described (Kang et al., 2003). The six vascular bundles of the axillary stems merge to form only two, derived from the three lateral vascular bundles from each side of the branch stem (Figure 4C). These then join the two closest vascular bundles of the primary stem, either side of the leaf trace. In the second pattern, only the more adaxial two vascular bundles on each side of the branch stem merge to form the two lateral vascular bundles, which, in turn, merge with the two adjacent vascular bundles of the primary stem, as above. The remaining two abaxial vascular bundles merge with the leaf trace (Figure 4B). The first of these patterns was the most common for both genotypes, with all but one of the max4 nodes adopting this anatomy, and 9/13 MAX4 nodes (Table 1). The second pattern was only observed in MAX4 nodes. We did not find any examples of max4 nodes in which branch vascular bundles merged with the leaf trace. In one case, one of the abaxial bundles had not merged with either the leaf trace or the adjacent stem vascular bundle in the most basal section of the series (Table 1). In contrast, for MAX4 nodes, in four out of 13 samples, both the abaxial vascular bundles of the branch stem merged with the leaf trace at the node. Thus, there was a greater tendency for vascular bundles to merge with stem vascular bundles versus the leaf trace in max4 mutants than in MAX4 WT nodes.
|
|
Interestingly, this trend was also observed for the smaller veins of the petiole (Table 1). The leaf petioles we sectioned had a large central midrib vein, but, in addition, they had smaller lateral veins running in parallel on either side of the midrib. In wild-type plants, in 11 out of 13 cases, these lateral petiole veins all joined the midrib above the node, such that the leaf vasculature was entirely connected to the leaf trace. In only two cases, one of the lateral petiole veins joined an adjacent vascular bundle in the stem, and not the leaf trace. The proportion of lateral veins adopting this second pattern in the max4 mutant was higher, at six out of 16. Thus, in general, in comparison to MAX4 WT, vasculature from lateral structures on max4 mutant primary stems has a greater tendency to merge with the stem vasculature bundles and a reduced tendency to merge with the leaf trace.
The MAX Signal Can Act Unilaterally
These and previous anatomical studies suggest that the vascular connectivity of successive axillary branches on the Arabidopsis bolting stem is relatively weak (Kang et al., 2003). The spiral phyllotaxy and pattern of vascularization mean that the vascular systems of the two sides of the plant could be partially independent, since successive branches vascularize to main stem vascular bundles on opposite sides of the stem. This observation is in contrast to the results described above for the two-branch and W assays, where there is clear evidence for communication between successive branches. To assess the degree of independence of the two sides of the stem with respect to shoot branching control, we constructed plants in which MAX4 expression could be induced by the topical application of dexamethasone (DEX) in a max4 mutant background, using the system developed by Craft et al. (2005), as described in Materials and Methods.
To determine whether unilateral MAX4 activity could allow unilateral branch suppression, the following experiment was carried out. Plants were grown under short days for 8 weeks and, after this period, DEX in lanolin paste was applied either encircling the hypocotyl or unilaterally. The plants were then transferred to long days for 4.5 weeks, promoting flowering. After this period, the number of rosette branches on each side of the plants was recorded.
DEX application in this system should restore wild-type MAX4 expression and hence a wild-type shoot branching habit. As expected, 1 mM DEX in lanolin paste applied fully encircling the hypocotyl suppressed shoot branching in comparison to untreated controls (Figure 5A and 5B). When DEX was applied on only one side of the hypocotyl, branching was suppressed preferentially on that side, indicating that there is substantial independence between the two sides of the stem (Figure 5C and 5D).
|
The construct used for DEX-inducible MAX4 expression includes a GUS gene under the control of the same DEX-dependent promoter; thus, GUS expression can be used to monitor DEX activity. When the unilaterally treated plants were stained for GUS activity, unilateral GUS expression was observed, consistent with the idea that DEX is carried up the plant in the transpiration stream and solutes transported in this way cross the stem relatively poorly (Figure 5E).
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODS |
|---|
In this study, we have provided evidence for communication between successive branches on the Arabidopsis bolting stem. Because of its spiral phyllotaxy, these branches are on opposite sides of the stem so that communication requires radial transmission of signals. However, the vascular connectivity we observed suggests the possibility for substantial independence of the two sides of the stem. Consistent with this idea, we have demonstrated that if the MAX pathway is activated on only one side of a stem, then branching can be specifically suppressed on that side (Figure 5). It has been proposed that the MAX pathway regulates shoot branching by limiting auxin transport capacity in the main stem, thus indirectly limiting the ability of axillary buds to establish a canalized auxin efflux stream out of the bud (Bennett et al., 2006). Among other things, canalization of auxin out of the bud into the stem may be necessary for establishing and enhancing a vascular connection between the bud and the main stem vascular bundles, which is presumably necessary to sustain axillary branch growth. Our observations of different patterns of vascularization of max4 mutant versus MAX4 WT lateral structures are consistent with this model (Figure 4 and Table 1). Both the axillary branch vasculature and the petiole minor veins have a greater tendency to merge with the main stem vascular bundles than with the leaf trace in max4 mutants versus MAX4 WT plants. This is consistent with the increased auxin transporter expression and the increased auxin transport capacity observed in the main stem of the max mutants (Lazar and Goodman, 2006; Bennett et al., 2006). The increased auxin transport capacity in the main stem vascular bundles would provide a stronger sink for auxin and thus would attract canalization pathways from auxin sources, such as active axillary shoots, as observed by Sachs for both exogenous addition of auxin to stem explants and specifically in the context of vascularization of axillary branches (Sachs 1968, 1969, 1981). The differences in vascularization of the max mutants vs WT therefore support the proposed model for MAX action.
The fact that branching can be differentially affected on the two sides of the stem by unilateral MAX induction (Figure 5) suggests that the main site for competition between the bud and primary apex for transport capacity in the stem is along the major auxin transport conduits specifically on the bud's side of the shoot. Highly local action for MAX has previously been demonstrated using clonal sectors of max2 mutant tissue on a MAX2 background (Stirnberg et al., 2007), where individual mutant buds adjacent to mutant stem can activate ectopically. Such unilateral and/or local bud activation by local auxin transport modulation in the main stem could have physiological relevance in situations such as unilateral shading. In this context, it is interesting to note that max2 is defective in several light-quality-related responses (Stirnberg et al., 2002; Shen et al., 2007).
Despite the clear evidence for substantial independence of the two halves of the stem, our Arabidopsis two-branch assay systems clearly show active communication across the stem, suggesting, for example, some radial movement of auxin (Figures 1 and 2). It is interesting that when we applied auxin to the decapitated stump of one axillary branch, this was much less effective at inhibiting the next branch than the intact branch it was designed to replace (Figures 1 and 3). This could be because the applied auxin source was poor at entering the stem, or perhaps because the canalization process resulted in a higher proportion of it being exported along the leaf trace, radially even further from the next bud than auxin exported from an active branch, which is expected to export auxin along the two adjacent vascular bundles of the stem. Interestingly, WT and max4 did not responded significantly differently to applied auxin (Figure 3), although max4 mutants have buds resistant to apical auxin (Bennett et al., 2006), and clearly showed reduced branch–branch communication in the other two-branch assays. This supports the idea of a two mechanisms for auxin-mediated bud inhibition, only one of which is MAX-dependent.
Two-branch assays in pea and bean have been a mainstay of research into the mechanism of apical dominance since they were introduced in the first quarter of the last century. It was these assays that made clear the indirect and acropetal effect of auxin, which runs counter to its basipetal movement down the main stem. We have developed two different two-branch systems for Arabidopsis, simple and W, allowing the combination of this useful assay with the genetic resources available in Arabidopsis.
Our simple two-branch system gives results qualitatively similar to those known for pea. Removal of the primary apex releases buds at adjacent nodes from apical dominance, allowing active shoot growth, but, over time, one shoot comes to dominate the other, such that this dominant shoot remains active, but the subordinate shoot re-enters a dormant state. We clearly see this effect in wild-type Arabidopsis, with one shoot remaining active while growth of the other slows down (Figure 1A and 1D). That one shoot inhibits the other is confirmed by the observation that removal of one of the shoots at the start of the experiment results in the other achieving a much greater mean length (Figure 1F and 1G). It is worth pointing out that this is not due to direct competition for limited nutrient supply, since the explants are well fertilized, and each explant is clearly capable of supporting the growth of two long branches because, in the W configuration, this is exactly what occurs (Figure 2D). Furthermore, straightforward competition would be predicted to result in two smaller shoots of similar size, rather than a single active shoot and an inactive shoot (Figure 1A).
In wild-type, in the simple two-branched system, which branch dominates is near random, although there is a slight bias in favor of the more basal branch (Figure 1F and 1G). This could reflect the fact that the basal bud is generally slightly larger at the start of the experiment (WT top bud: 2 mm ± 0.07 SE and basal bud 2.24 mm ± 0.07 SE; max4 top bud: 1.9 mm ± 0.07 SE and basal bud: 2.3 mm ± 0.08 SE, giving it a growth advantage from the beginning. Nonetheless, the top branch can clearly inhibit the growth of the bottom branch and vice versa, demonstrating both acropetal and basipetal inhibitory effects, as observed in pea. In contrast to the two-branch pea, where acropetal transmission of inhibition is observed only in the inhibited branch itself, in the Arabidopsis system, the asymmetric positions of the nodes on the main stem make it easy to observe both acropetal and basipetal signal transmission in the main stem.
In the max4 mutant, both branches grow more quickly than in the WT (Figure 1F and 1G). There is still clear evidence that they can inhibit one another's growth, since the relative growth index is still high and removal of one branch increases the mean length achieved by the other (Figure 1G). However, the ability of one branch progressively to dominate is reduced, as evidenced by the observation that the relative growth index of the branches does not increase between day 7 and day 11 (Figure 1E). There is also a clearer bias in favour of the bottom branch (Figure 1F and 1G). One interpretation of this observation is that loss of MAX activity inherently affects the ability of the top branch to inhibit the bottom one more strongly than the reciprocal. Alternatively, it could be that reduction in the strength of inhibition in either direction results in the initial small bias in favor of the bottom branch being amplified. Consistent with this idea, in the two-branch pea system, various treatments that confer even a small growth advantage on one branch over the other bias the establishment of that branch as the dominant one, leading to the inhibition of the other pea (Snow, 1931). This idea is consistent with the observation that the behaviour of max4 explants in the simple two-branch system is very similar to that observed for wild-type explants in the W assay (Figures 1 and 2). Both branches grow more quickly than for the WT simple system; there is a tendency for one branch to dominate the other, although this is reduced and does not increase over time; and the ability of the top shoot to inhibit the bottom shoot is reduced more strongly than the reciprocal. The effects of stem division and max4 mutation appear to be essentially additive. For max4 mutants in the W configuration, both shoots are longer compared with the simple two-branch configuration, and the growth of the bottom shoot is completely insensitive to the presence of the top shoot, whereas the presence of the bottom shoot still has a weak inhibitory effect on the top shoot, but this diminishes over time.
The additivity of the effects of max4 mutation and stem division in compromising branch–branch communication suggest that two mechanisms may be involved, one of which is MAX4-dependent and the other of which requires an intact stem. The similarity of the effects of loss of max4 function and stem division suggests that the mechanisms have similar properties and are partially redundant. As described in the Introduction, two mechanisms have been proposed for auxin-mediated bud inhibition: one involves auxin-mediated down-regulation of cytokinin synthesis, which is AXR1/TIR1-dependent, and the other involves competition between auxin sources for limited auxin transport capacity in the main stem, which is MAX-dependent. If both these mechanisms are operating in the inhibition of one branch by the other branch, this could explain the additivity of max mutation and stem division.
As described above, in both genotypes, the ability of one branch to inhibit the other is greatly reduced by division of the stem to a point above the upper node, creating W explants (Figures 1 and 2). The serious weakening of communication in the W configuration compared to the intact stem is further illustrated by the observation of accessory branches in the denuded axils of W configuration explants (Figure 2H), but never in the simple two-branch configuration. This implies that in the simple system, the presence of the other shoot inhibits accessory development in the other axil, either acropetally or basipetally, and this inhibitory influence is reduced upon division of the stem. Here, in contrast to the main axillary branch, the top branch has a stronger effect on the bottom axil than the reciprocal, and max4 mutation, if anything, strengthens this effect rather than reducing it. Similar to the bias in favour of the bottom branch, it is possible that the bias in accessory development in favour of the top axil reflects a different potential of the nodes to produce accessories. However, there is no evidence for this in studies of intact plants, where accessory development has been observed in all nodes with no clear pattern (Grbic and Bleecker, 2000; Talbert et al., 1995). Thus, this bias may reflect asymmetries in signaling across the W, such as more robust transmission of basipetal inhibition than acropetal inhibition along the split stem.
The fact that max4 mutants have fewer accessories than WT (Figure 2H) suggests that the MAX pathway, and thus restricted auxin transport capacity, is not required to transmit accessory inhibition, in contrast to the situation with basic branch length inhibition. This suggests that the cytokinin-dependent mechanism may be primarily responsible. The reduced accessory formation observed in max4 mutants is consistent with this idea, because max mutants have reduced xylem sap cytokinin content compared with wild-type (Foo et al., 2007). Furthermore, the cytokinin-dependent mechanism is proposed to operate through AXR1/TIR1-mediated auxin signaling down-regulating cytokinin synthesis. It has previously been observed that axr1 mutants have an increased frequency of accessory branching on fully intact plants (Stirnberg et al., 1999), as does the cytokinin over-producing supershoot mutant (Tantikanjana et al., 2001). These observations support further the idea that accessory production can be regulated through this pathway. Therefore, our data are consistent with a model in which branch–branch communication involves both the MAX and AXR1/TIR1/cytokinin pathways, whilst branch-accessory communication involves only the AXR1/TIR1/cytokinin pathway.
Thus, our observations of branch length and accessory development in W configuration plants at least partially confirm the results of Snow that an inhibitory effect can be transmitted between branches via a tortuous route down one branch up the divided stem to the top, down the other side of the divided stem and up into the other axil (Snow, 1937). Our data demonstrate that restricted auxin transport capacity is required along at least part of this transmission route. Furthermore, they support a role for the AXR1/TIR1/cytokinin pathway, too. However, there are currently insufficient data to explain fully how signals can be transmitted across the W. Furthermore, it is unclear whether any of the asymmetries observed in the transmission of inhibitory signals, such as inhibition of accessories in the bottom versus top nodes, can be explained by the robustness of the mechanism of signal transmission along the different parts of the route. The genetic tools available in Arabidopsis, along with these new assays, make it possible to address these questions in the future.
|
METHODS |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODS |
|---|
Plant Material
Seeds of Arabidopsis thaliana (L.) Heynh ecotype Columbia WT, max4, lfy, lfy max4, and DEX-inducible MAX4 were sown on P40 trays on F2+sand (Scotts Levington) treated with intercept 70 WG. Trays were kept at 4°C for at least 2 d, and all the genotypes apart from DEX-inducible MAX4 were then transferred to the greenhouse under the following conditions: 16 h light /8 h dark and temperature between 16 and 21°C. DEX::MAX4 plants were placed in a growth cabinet under short-day conditions (8 h light /16 h dark) for 8 weeks and later shifted to long-day conditions (16 h light /8 h dark) in the greenhouse.
Two-Branch Arabidopsis Stem Segments
WT and max4 plants with 1.5–3 cm bolting stems that carried cauline buds (top and bottom) close together were selected. The whole bolting stem was excised with a scalpel and placed under the dissecting microscope. The shoot apex was removed above the top cauline node and the stem was trimmed, leaving approx 1 cm below the bottom bud, leaving two buds, which were then measured. At day 0, buds were between 1.5 and 3 mm. In experiments in which one bud was removed, this was carried out using a scalpel. For the W system, stems were bisected using a scalpel to a point above the top node. After the apex was removed, lanolin (Sigma) was applied to the decapitated stump using a sharpened cocktail stick. When one bud was removed, lanolin or NAA in lanolin paste was applied.
Stem segments were then inserted through a hole in the closed cap of a 1.5 ml microfuge tube containing cotton wool soaked in ATS without sucrose (Wilson et al., 1990). For W plants, each half-stem was placed in a well of a 96-well plate with 2-ml wells, which also contained cotton wool saturated with ATS without sucrose.
Microfuge tubes and 96-well plates were placed in a growth room on a tray with wet paper on the bottom and with a propagator lid, under the following condition: 16 h light /8 h dark and temperature between 18 and 21°C. Bud outgrowth was measured with a millimeter ruler.
To prepare the lanolin solutions, lanolin was placed in a microfuge tube and heated up to 60°C. NAA in 70% ethanol was added to a final concentration of 1 mM NAA. Food dye was also added to ensure an even distribution of the hormone in the lanolin paste. In controls, the lanolin contained food dye and 70% ethanol.
Anatomical Studies
For anatomical studies, 5-week-old lfy and lfymax4 plants were selected. The processes of tissue clearing, embedding, and staining were adapted from Ruzin (1999). lfy and lfy max4 cauline nodes, each carrying a branch and a leaf, were collected from 5-week-old plants. The top part of the bud and the leaf blade were excised, and nodes were fixed in 2.5% NaOH for 48 h after being transferred to Chloral hydrate (2.7 g ml–1) for 11 d. The tissue was then washed twice in distilled water for 15 min and then left overnight (ON) in water. The tissue was dehydrated through an ethanol series (30, 50, 70, 90, 95 and 100% ethanol, each for 1 h) and the ethanol was then replaced with histoclear through 3:1, 1:1, 1:3 ethanol:histoclear washes, each for 1 h, followed by two 100% histoclear washes for 1 h each. Paraffin chips (Paraplast plus, Sigma) were added, and the tissue was left ON at room temperature. The samples were warmed to 42°C and the histoclear was saturated with paraffin. Then, a third of the volume was replaced by paraffin and tubes were incubated at 58°C ON. The next day, half of the tube was emptied and replaced with paraffin and this was repeated twice at 12-h intervals. Finally, the whole content of the tube was replaced with paraffin three times at 12-h intervals. Paraffin blocks were made and cut into 10-mm sections, which were transferred onto glass microscope slides. The slides were washed twice in histoclear for 15 min each time, and then hydrated through an ethanol series (95, 70, 50, 30% ethanol for 5 min each), after which they were stained in 1% safranin for 1 h. The tissue was dehydrated through an ethanol series (30, 50, 70, and 95% for 5 min each) and counterstained with 0.5% fast green in 95% ethanol for 5 s. Finally, two washes with 100% ethanol and another two with 100% histoclear were carried out (1 min each). Slides were mounted in entellan solution and viewed under a standard light microscope.
Generation of MAX4 DEX Inducible Line
To generate plants in which MAX4 expression could be exogenously induced with dexamethasone (DEX), a two-component system was used (Craft et al., 2005). In this system, the DEX receptor is fused N-terminally to a chimeric transcription factor consisting of the lac repressor as a DNA-binding domain and the GAL4 transcriptional activation domain. This protein, designated LhG4-N, has been introduced into plants under the control of the CaMV 35S promoter (Craft et al., 2005). On a separate construct, any gene of interest can be fused to a promoter including lac repressor binding elements, thus rendering it DEX-inducible in the presence of LhG4-N. An expression cassette allowing such fusion to be generated (pH–TOP–GUS) also includes a GUS gene expressed divergently from the same promoter, such that DEX inducibility can be monitored using GUS activity (Craft et al., 2005).
WT plants, homozygous for the LhG4-N trans-acting construct (Craft et al., 2005), were crossed to max4 mutants and F2 plants with the max4 phenotype and homozygous for the LhG4-N transgene were identified. The MAX4 cDNA was cloned into the pH–TOP–GUS vector. The MAX4 coding sequence was amplified by PCR from MAX4 cDNA in pUC18. The forward primer (5'–GCCTCGAGATGGCTTCTTTGATCACAA–3’) introduced an XhoI site immediately 5’ to the MAX4 ATG. The reverse primer (5'–GCGGATCCTGTGATCAAAGAAGC–3’) included a BglII site 3’ to the MAX4 stop codon. The PCR product was cloned into the pCR2.1–TOPO vector (Invitrogen). The resultant MAX4 XhoI and BglII fragment was cloned into pH–TOP–GUS (which confers hygromycin resistance), cut with BamHI and SalI. The resultant plasmid was transformed into Agrobacteria and introduced into max4, LhG4-N background by floral dip (Clough and Bent, 1998). Three independent transformed lines were selected for further analysis and, from these, one was chosen for future studies based on uniform GUS induction with DEX application.
Unilateral MAX4 Induction
DEX inducible MAX4 plants (designated DEX, MAX4) were grown under short days (8 h light/16 h dark) for 8 weeks. 1 mM dexamethasone (DEX) was applied in lanolin paste, either fully encircling the hypocotyl, or to only one side of it. Controls were treated with lanolin only. The plants were then shifted to long days to induce flowering, and allowed to grow for 4.5 weeks. The number of rosette branches was then scored for each half of the rosette. The DEX solution in lanolin paste was prepared in the same way as the NAA solution already described. These experiments were repeated at least once. GUS-staining was carried out according to Jefferson et al. (1987).
|
Acknowledgements |
|---|
We would like to thank the University of York Horticultural staff for expert plant care, and Patrick Crozier and Stephen Day for critical reading of the manuscript. This work was funded by BBSRC. No conflict of interest declared.
-
Bangerth F. Response of cytokinin concentration in the xylem exudates of bean (Phaseolus vulgaris L.) plants to decapitation and auxin treatment, and relationship to apical dominance. Planta (1994) 194:439–442.[Web of Science]
Bennett T, Sieberer T, Willett B, Booker J, Luschnig C, Leyser O. The Arabidopsis MAX pathway controls shoot branching by regulating auxin transport. Curr. Biol. (2006) 16:553–563.[CrossRef][Web of Science][Medline]
Blakeslee JJ, Pee WA, Murphy AS. Auxin Transport. Curr. Opin. Plant Biol. (2005) 8:494–500.[CrossRef]
Booker J, Auldridge M, Wills S, McCarty D, Klee H, Leyser O. MAX3/CCD7 is a carotenoid cleavage diosygenase required for the synthesis of a novel plant signaling molecule. Curr. Biol. (2004) 14:1232–1238.[CrossRef][Web of Science][Medline]
Booker J, Chatfield S, Leyser O. Auxin acts in xylem-associated or medullary cells to mediate apical dominance. Plant Cell (2003) 15:495–507.
Booker J, Sieberer T, Wright W, Williamson L, Willett B, Stirnberg P, Turnbull C, Srinivasan M, Goddard P, Leyser O. MAX1 encodes a cytochrome P450 family member that acts downstream of MAX3/4 to produce a carotenoid-derived branch-inhibiting hormone. Dev. Cell (2005) 8:443–449.[CrossRef][Web of Science][Medline]
Cline MG. Apical dominance. Bot. Rev. (1991) 57:318–358.[CrossRef]
Cline MG. Concepts and terminology of apical dominance. Am. J. Bot. (1997) 84:1064–1069.[Abstract]
Clough SJ, Bent AF. Floral dip: a simplified method for Agrobacterium- mediated transformation of Arabidopsis thaliana. Plant J. (1998) 16:735–743.[CrossRef][Web of Science][Medline]
Craft J, Samalova M, Baroux C, Townley H, Martinez A, Jepson I, Tsiantis M, Moore I. New pOp/LhG4 vectors for stringent and efficient glucocorticoid-inducible gene expression in Arabidopsis. Plant J. (2005) 41:899–918.[CrossRef][Web of Science][Medline]
Foo E, Morris SE, Parmenter K, Young N, Wang H, Jones A, Rameau C, Turnbull CGN, Beveridge CA. Feedback regulation of xylem cytokinin content is conserved in pea and Arabidopsis. Plant Physiol. (2007) 143:1418–1428.
Grbic V, Bleecker AB. Axillary meristem development in Arabidopsis thaliana. Plant J. (2000) 21:215–223.[CrossRef][Web of Science][Medline]
Hall SM, Hillman J. Correlative inhibition of lateral bud growth in Phaseolus vulgaris L.: timing of bud growth following decapitation. Planta (1975) 123:137–143.[CrossRef][Web of Science]
Jefferson RA, Kavanagh TA, Bevan MW. GUS fusions: beta-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO J. (1987) 6:3901–3907.[Web of Science][Medline]
Kang J, Tang T, Donnelly P, Dengler N. Primary vascular pattern and expression of ATHB-8 in shoots of Arabidopsis. New Phytol. (2003) 158:443–454.[CrossRef][Web of Science]
Lazar G, Goodman HM. MAX1, a regulator of the flavonoid pathway, controls vegetative axillary bud outgrowth in Arabidopsis. Proc. Natl Acad. Sci. U S A (2006) 103:472–476.
Leyser O. Regulation of shoot branching by auxin. Trends Plant Sci. (2003) 8:541–545.[CrossRef][Web of Science][Medline]
Li C-J, Bangerth F. Autoinhibition of indoleacetic acid transport in the shoot of two–branched pea (Pisum sativun) plants and its relationship to correlative dominance. Physiol. Plant (1999) 106:15–20.
Ljung K, Bhalerao RP, Sandberg G. Sites and homeostatic control of auxin biosynthesis in Arabidopsis during vegetative growth. Plant J. (2001) 28:465–474.[CrossRef][Web of Science][Medline]
Morris DA. Transport of exogenous auxin in two-branched dwarf pea seedlings (Pisum sativun L.). Planta (1977) 136:91–96.[CrossRef][Web of Science]
Morris DJ, Johnson CF. The role of auxin efflux carriers in the reversible loss of polar auxin transport in the pea (Pisum sativum L.) stem. Planta (1990) 181:117–124.[Web of Science]
Nordström A, Tarkowski P, Tarkowska D, Norbaek R, Åstot C, Dolezal K, Sandberg G. Auxin regulation of cytokinin biosynthesis in Arabidopsis thaliana: a factor of potential importance for auxin–cytokinin-regulated development. Proc. Natl Acad. Sci U S A (2004) 101:8039–8044.
Paponov IA, Teale WD, Trebar M, Blilou I, Palme K. The PIN auxin efflux facilitarors: evolutionary and functional perspectives. Trends Plant Sci. (2005) 10:170–177.[CrossRef][Web of Science][Medline]
Ruzin SE. (1999) Plant Microtechnique and Microscopy (Oxford University Press).
Sachs T. On the determination of the pattern of vascular tissue in peas. Ann. Bot. (1968) 32:781–790.
Sachs T. Polarity and the induction of organized vascular tissue. Ann. Bot. (1969) 33:263–275.
Sachs T. The control of patterned differentiation of vascular tissues. Ad. Bot. Res. (1981) 9:151–162.[CrossRef]
Shen H, Luong P, Huq E. The F-box protein MAX2 functions as a positive regulator of photomorphogenesis in Arabidopsis. Plant Phys. Preview published 19 October 2007, doi 10.1104/pp.107.107227 (2007).
Sorefan K, et al. MAX4 and RMS1 are orthologous dioxygenase-like genes that regulate shoot branching in Arabidopsis and pea. Genes Dev. (2003) 17:1469–1474.
Snow R. The transmission of inhibition through dead stretches of stem. Ann. Bot. (1929) 43:261–267.
Snow R. Experiments on growth inhibition. Part II: New phenomena of inhibition. Proc. Roy. Soc. (1931) 108:305–316.
Snow R. On the nature of correlative inhibition. New Phyt. (1937) 36:283–300.[CrossRef]
Stirnberg P, Chatfield SP, Leyser HMO. AXR1 acts after lateral bud formation to inhibit lateral bud growth in Arabidopsis. Plant Physiol. (1999) 121:839–847.
Stirnberg P, Furner IJ, Leyser HMO. MAX2 participates in an SCF complex which acts locally at the node to suppress shoot branching. Plant J. (2007) 50:80–94.[CrossRef][Web of Science][Medline]
Stirnberg P, van de Sande K, Leyser HMO. MAX1 and MAX2 control shoot lateral branching in Arabidopsis. Development (2002) 129:1131–1141.
Talbert PB, Adler HT, Parks DW, Comai L. The REVOLUTA gene is necessary for apical meristem development and for limiting cell divisions in the leaves and stems of Arabidopsis thaliana. Development (1995) 121:2723–2735.[Abstract]
Tanaka M, Takei K, Kojima M, Sakakibara H, Mori H. Auxin controls local cytokinin biosynthesis in the nodal stem in apical dominance. Plant J. (2006) 45:1028–1036.[Web of Science][Medline]
Tantikanjana T, Yong JWH, Letham DS, Griffith M, Hussain M, Ljung K, Sandberg G, Sundaresan V. Control of axillary bud initiation and shoot architecture in Arabidopsis through the SUPERSHOOT gene. Genes and Dev. (2001) 15:1577–1588.
Thimann KV, Skoog F. Studies on the growth hormone of plants III: The inhibitory action of the growth substance on bud development. Proc. Natl Acad. Sci. USA (1933) 19:714–716.
Wilson AK, Pickett FB, Turner JC, Estelle M. A dominant mutation in Arabidopsis confers resistance to auxin, ethylene and abscisic acid. Mol. Gen. Genet. (1990) 222:377–383.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Prusinkiewicz, S. Crawford, R. S. Smith, K. Ljung, T. Bennett, V. Ongaro, and O. Leyser Control of bud activation by an auxin transport switch PNAS, October 13, 2009; 106(41): 17431 - 17436. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. B. Brewer, E. A. Dun, B. J. Ferguson, C. Rameau, and C. A. Beveridge Strigolactone Acts Downstream of Auxin to Regulate Bud Outgrowth in Pea and Arabidopsis Plant Physiology, May 1, 2009; 150(1): 482 - 493. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. J. Ferguson and C. A. Beveridge Roles for Auxin, Cytokinin, and Strigolactone in Regulating Shoot Branching Plant Physiology, April 1, 2009; 149(4): 1929 - 1944. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||