Plant Physiology Preview. Published on July 29, 2015, as DOI: /pp

Transcription

1 Plant Physiology Preview. Published on July 29, 2015, as DOI: /pp Running head: Shoot branching and auxin transport Corresponding author: Christine A. Beveridge 1 The University of Queensland, School of Biological Sciences, St. Lucia, QLD 4072, Australia Phone: c.beveridge@uq.edu.au. Research area: Signaling and Response 1 Copyright 2015 by the American Society of Plant Biologists

2 Strigolactone inhibition of branching independent of polar auxin transport Philip B. Brewer, Elizabeth A. Dun, Renyi Gui 1, Michael G. Mason, and Christine A. Beveridge * The University of Queensland, School of Biological Sciences, St. Lucia, QLD 4072, Australia One-sentence summary (max 200 characters): Strigolactones act independently of auxin transport/canalization to inhibit bud outgrowth. 2

3 Present address: 1 Nurturing Station of State Key Laboratory of Sub-tropical Silviculture, Zhejiang Agriculture and Forestry University, Hangzhou, China * Corresponding author: Christine A. Beveridge, c.beveridge@uq.edu.au. Financial source: This research was funded by the Australian Research Council Discovery and Future Fellowship programs 3

4 Abstract The outgrowth of axillary buds into branches is regulated systemically via plant hormones and the demand of growing shoot tips for sugars. The plant hormone auxin is thought to act via two mechanisms. One mechanism involves auxin regulation of systemic signals, cytokinins and strigolactones, which can move into axillary buds. The other involves suppression of auxin transport/canalization from axillary buds into the main stem and is enhanced by a low sink for auxin in the stem. In this theory, the relative ability of buds and stem to transport auxin controls bud outgrowth. Here we evaluate whether auxin transport is required or regulated during bud outgrowth in pea (Pisum sativum). The profound, systemic and long-term effects of the auxin transport inhibitor N-1-naphthylphthalamic acid had very little inhibitory effect on bud outgrowth in strigolactone deficient mutants. Strigolactones can also inhibit bud outgrowth in N-1-naphthylphthalamic acid-treated shoots that have greatly diminished auxin transport. Moreover strigolactones can inhibit bud outgrowth despite a much diminished auxin supply in in vitro or decapitated plants. These findings demonstrate that auxin sink strength in the stem is not important for bud outgrowth in pea. Consistent with alternative mechanisms of auxin-regulation of systemic signals, enhanced auxin biosynthesis in Arabidopsis thaliana can suppress branching in yuc1d plants compared to wild-type plants, but has no effect on bud outgrowth in a strigolactone-deficient mutant background. Introduction Shoot branching is a highly plastic developmental trait that is greatly affected by genetic and environmental factors. Branches develop from buds that often enter a state of highly repressed growth, which can be rapidly re-activated to stimulate branch development in response to systemic cues from the plant such as sucrose (Mason et al., 2014). The mechanism of auxin action as the hormonal component of the shoot tip that suppresses axillary buds has led to a model comprised of two major hypotheses, auxin transport/canalization and direct hormone action (Fig. 1). Here we address the relative importance of these hypotheses in shoot branching. In the auxin transport/canalization hypothesis, auxin in the main stem is thought to act by inhibiting the flow of auxin from buds (Fig. 1; (Shinohara et al., 2013; Waldie 4

5 et al., 2014)). According to this hypothesis, the plant hormone strigolactone inhibits bud growth by dampening auxin transport and by creating a reduced sink for auxin in the auxin transporting cells in the vasculature of the main stem (Fig. 1). The case in support of this hypothesis is compelling due to experiments showing that auxin in the 5

6 main stem repels bud vascular connections (Sachs, 1969) and that auxin flow from a growing shoot tip correlates with the repression of other buds or branches (Morris, 1977; Bangerth, 1989; Bennett et al., 2006). Experiments showing that strigolactone mutants can have enhanced levels of auxin and auxin transport in the main stem (Beveridge, 2000; Beveridge et al., 2000; Crawford et al., 2010; Agusti et al., 2011) and that strigolactone can only inhibit branching in vitro in the presence of an apical auxin supply support the hypothesis that strigolactones could act via repression of auxin canalization (Bennett et al., 2006; Ongaro and Leyser, 2008; Crawford et al., 2010; Liang et al., 2010). The polarization and localization of PIN auxin export proteins on cell membranes, which is important for polar auxin transport (Wiśniewska et al., 2006), is thought to promote the establishment of a channel of focused auxin flow that coincides with new vasculature formation (Sauer et al., 2006). As predicted, PINs are apolar in the vasculature of repressed buds in intact plants, but 2 to 6 h after removal of the main shoot tip (decapitation), auxin builds up in the buds (Gocal et al., 1991) and buds commence polar auxin transport as PINs polarize towards the stem (Balla et al., 2011). In Arabidopsis (Arabidopsis thaliana), strigolactones promote PIN internalization, reducing the abundance of PIN proteins available for auxin transport ((Shinohara et al., 2013); Fig. 1). This is a major finding in support of strigolactone action via auxin transport. The direct action hypothesis of bud outgrowth highlights the role of strigolactones independent of auxin transport and the role of strigolactone acting downstream of auxin (Fig. 1; (Beveridge, 2000; Beveridge et al., 2000; Brewer et al., 2009)). Combined grafting and hormone studies with strigolactone mutants led to the hypothesis that auxin acts indirectly via other signals, cytokinins and strigolactones (Beveridge, 2000; Beveridge et al., 2000; Brewer et al., 2009) as predicted by Snow (1937). In this hypothesis, cytokinins and strigolactones act as long-distance secondary messengers for auxin, but do not require auxin to act. Accordingly, strigolactone supplied directly to axillary buds of garden pea (Pisum sativum) can inhibit branching all along the stem of decapitated, auxin-depleted plants (Brewer et al., 2009; Dun et al., 2013).. Moreover, unless given a strigolactone supply by grafting, strigolactone deficient mutants respond very poorly to auxin added to the stump after decapitation (Beveridge et al., 2000), indicating that auxin requires strigolactones to elicit its inhibitory effect. Additionally, exogenous strigolactone can 6

7 reduce branching in auxin mutants and can reduce growth of large axillary buds without affecting radiolabeled auxin transport (Brewer et al., 2009). Higher auxin transport in strigolactone mutants (Bennett et al., 2006) may be a consequence of the feedback loop between strigolactones and auxin (Hayward et al., 2009) rather than the cause of branching, and is not always observed in pea strigolactone mutants (Brewer et al., 2009). The mechanism of the direct action of strigolactones has not been fully elucidated, but includes the bud-specific, transcription factor BRANCHED1 (BRC1 in pea and Arabidopsis, TEOSINTE BRANCHED1 in maize (Zea mays) and FINE CULM1 in rice (Oryza sativa); (Doebley et al., 1997; Takeda et al., 2003; Aguilar-Martínez et al., 2007; Finlayson, 2007; Braun et al., 2012). In support of this, brc1 mutants are highly branched (Doebley et al., 1997; Takeda et al., 2003; Aguilar-Martínez et al., 2007; Finlayson, 2007; Braun et al., 2012) and BRC1 is regulated by strigolactones without the need for protein synthesis (Dun et al., 2012). Indeed, BRC1 is broadly regarded as an integrator of branching signals and the environment (Kebrom et al., 2006; Aguilar-Martínez et al., 2007; Minakuchi et al., 2010; Braun et al., 2012; Dun et al., 2012). In pea, BRC1 is also regulated by cytokinins and sucrose (Braun et al., 2012; Dun et al., 2012; Mason et al., 2014) and it is possible that cell cycle progression is repressed by BRC1 in the bud (Martín-Trillo and Cubas, 2010). In order to unravel the relative importance of these hypotheses, we first sought to explore the isolated node in vitro system that shows a strigolactone response only in the presence of auxin (Bennett et al., 2006; Ongaro and Leyser, 2008; Crawford et al., 2010; Liang et al., 2010), because this might give information as to why apical auxin has been required to inhibit bud outgrowth in isolated nodes, but is not required for the same response in decapitated plants (Brewer et al., 2009; Dun et al., 2013). We are also intrigued by the observation that auxin can inhibit but outgrowth in nodal segments from strigolactone mutants in vitro (Young et al., 2014), which lack both shoot tip, other buds and a root-system, but cannot do so in decapitated plants (Beveridge et al., 2000), which lack only the shoot tip. We then conducted a range of experiments, in vivo, to further test the auxin canalization process in pea and finally in Arabidopsis. In so doing, we tested N-1-naphthylphthalamic acid (NPA) as an essentially qualitative, systemic inhibitor of auxin transport in pea. NPA inhibits 7

8 the transport function of PIN and ABCB (PGP) auxin transporters in a synergistic way (Petrášek et al., 2006; Blakeslee et al., 2007), but NPA does not remove PIN transporters from the cell membranes (Kleine-Vehn et al., 2006; Sauer et al., 2006). Thus, NPA would be expected to block auxin transport downstream of the action of strigolactones on PIN internalization. We then tested whether bud outgrowth could commence and persist under conditions of severely impaired auxin transport, and whether this prevented strigolactone action. 8

9 Results Strigolactone can act independently of auxin in vitro One of the findings commonly used to support the auxin transport/canalization model is the requirement for a competing auxin source for strigolactone-mediated bud inhibition (Crawford et al., 2010; Shinohara et al., 2013; Waldie et al., 2014). In a similar split-plate in vitro assay to that used for Arabidopsis (Chatfield et al., 2000; Bennett et al., 2006), we also found that pea buds do not respond to a basal supply of GR24, a synthetic strigolactone, unless they are also supplied with auxin apically (Fig. 2B,C). However, when we applied GR24 directly to buds in this in vitro system, we observed repressed growth without an auxin supply in both wild type (Fig. 2D) and a strigolactone deficient branching mutant (Fig. 2E). Consequently, apical auxin is not essential for bud growth inhibition by strigolactone in vitro or in vivo (decapitated plants; (Brewer et al., 2009; Dun et al., 2013)). However, experiments with a new in vitro method gave strikingly contrasting results to the split-plate assay. Comparable segments to those used in the split-plate assay (Fig. 2) were inserted into individual, small, open, agar-filled tubes (Fig. 3A) and exposed to the same concentration of GR24 as the previous in vitro experiment. These segments were then responsive to GR24 in the media, even in the absence of apical auxin (Fig. 3B). This was repeated in independent experiments including with a strigolactone deficient branching mutant (Fig. 3C). It is unclear exactly why basally supplied strigolactone is ineffective in the split-plate (Fig. 2), but very effective in individual, open, agar-filled tubes (Fig. 3). However, the new findings for pea show that direct application of strigolactones to buds in the split plate assay and strigolactone supplied basally in the individual tubes can indeed inhibit bud outgrowth in isolated segments without an exogenous auxin supply. This supports the hypothesis of direct action, that strigolactones act downstream of auxin. NPA is a rapid, long-lasting and systemic inhibitor of auxin transport Although exogenous apical auxin is not required for strigolactone to inhibit branching in pea in vitro (Figs 2,3) or in decapitated plants (Brewer et al., 2009; Dun et al., 2013), it is possible that fully intact systems may somehow differ in strigolactone action and auxin interplay. Furthermore, with respect to the canalization model, one might hypothesize that bud growth inhibition via auxin canalization might still be able 9

10 to operate in these systems without apical auxin due to exogenous GR24 dramatically enhancing PIN removal in buds (Waldie et al., 2014). We sought to test the prediction that auxin transport is required for bud growth in vivo (Shinohara et al., 2013; Waldie et al., 2014) by carefully observing auxin transport and branching after 10

11 pea plants were exposed to a high dose of NPA. As bud outgrowth occurs over several days and auxin canalization from buds may continue to be important during this time, we sought to determine whether NPA could act rapidly and over the long term. Radiolabeled auxin ( 3 H-indole-3-acetic acid; 3 H-IAA) transport from the shoot 11

12 tip was measured over time after a single treatment of 0.1% NPA in lanolin as a ring around the stem above the highest expanded leaf. Very little auxin transport was detected during the period of 2 h to 5 d after treatment with NPA (Fig. 4; Supplemental Fig. S1). For example, on day 2, when strigolactone deficient mutant 12

13 plants showed significant outgrowth, wild-type and mutant plants treated with NPA transported 0.4% or 0.3%, respectively, of the radiolabel from 3 H-IAA compared with control-treated plants (measured at 6 cm; Supplemental Fig. S1C). The amount of polar auxin transport in NPA treated plants is likely to be considerably less than this because the very small movements of radiolabel from the 3 H-IAA supplied to the NPA-treated plants do not show the classical polar transport shape (Supplemental Fig. S1). The movement is too fast to be caused by diffusion (Kramer et al., 2011), but is certainly fast enough to be caused by phloem transport (Morris et al., 1973), which is not a proposed target of strigolactones. In contrast, strigolactones have been shown to affect PIN localization (Shinohara et al., 2013), which directly affect the polar auxin transport pathway (Wiśniewska et al., 2006). As NPA appeared to act at least within 2 h (Fig. 4A; see also Thomson et al. (1973); Sussman and Goldsmith (1981)) and probably over distance (Supplemental Fig. S1B-D), we sought to gauge more carefully how quickly NPA may systemically affect auxin transport in shoots in vivo in pea, particularly over a long distance (e.g., 30 cm). Stems were treated with 1% NPA in lanolin about 30 cm from the shoot tip and, in order to capture any changes over a short period, shoot tips were then treated at 2 and 4 h with 3 H-IAA (Fig. 5A). A significant difference in radiolabel from 3 H-IAA was observed in NPA-treated, compared with control plants, at positions including near the peak front (4 cm; P<0.005), indicating that NPA has a systemic effect as early as 2 h over the distance of about 30 cm (Fig. 5A). In order to test whether NPA itself moves systemically in shoots, we added 3 H-NPA to the 1% NPA lanolin mixture. Based on the radiolabel recovered from the 3 H-NPA, up to 0.26 μg of NPA was delivered to the shoot tip within 6 h (Fig. 5B). This equates to at least 40 μm in a 20 mg shoot tip, which is well within the range of the concentrations typically used in vitro to affect auxin transport (Kleine-Vehn et al., 2006; Petrášek et al., 2006; Sauer et al., 2006; Blakeslee et al., 2007). These findings of the rapid and systemic effects of NPA validate its use in the investigation of the role of auxin transport in axillary bud outgrowth. Auxin transport is not correlated with shoot branching We included strigolactone-deficient plants in the auxin transport experiments above (Fig. 4) to test whether they are affected in auxin transport or in response to NPA. In 13

14 this experiment, as in Brewer et al. (2009), there was no enhancement of auxin transport in strigolactone-deficient mutant control plants compared with wild-type control plants (Fig. 4A; Supplemental Fig. S1). This serves to highlight that differences in auxin transport are not always observed in strigolactone mutants of 14

15 pea (e.g., (Beveridge et al., 2000; Brewer et al., 2009)). If strigolactones were to act primarily via auxin transport to regulate shoot branching, we would expect to observe differences in auxin transport that correlate with the timing of bud outgrowth. While buds remained small in wild-type plants, branching occurred in strigolactonedeficient mutant plants over the course of this experiment (Fig. 4B) and over a period during which we consistently observed no enhancement of auxin transport in mutant control plants compared to wild type. Previous studies in pea have also shown that, unlike strigolactone, which is highly effective on small buds, 1 mm NPA treated in solution directly to small buds of a size comparable to inhibited intact wild-type plants (~1 mm) has little or no effect on bud outgrowth after decapitation (Mason et al., 2014) or in strigolactone mutant plants (Brewer et al., 2009). In contrast, little branches of 1 cm in length also show a small inhibitory growth reduction in response to NPA (Brewer et al., 2009). As pointed out by Waldie et al. (2014) this reduction of growth by NPA was similar to the GR24 response, but this similarity is restricted to this advanced stage of bud growth and to a period during which GR24 is less able to inhibit bud growth (Brewer et al., 2009; Dun et al., 2013). In contrast with Arabidopsis, GR24 had no effect on the auxin transport in pea (Brewer et al., 2009). In Arabidopsis, GR24 causes small effects on 3 H-IAA transport compared with NPA (Crawford et al., 2010) and low levels of NPA can reduce bud growth measured at 4 weeks (e.g., 1 µm NPA; (Bennett et al., 2006)). According to Waldie et al. (2014), this may be attributed to the different molecular effects of GR24 and NPA on auxin transport. However, this does not explain how the lack of a measurable effect of GR24 on 3 H-IAA transport in pea can coincide with significant effects on growth via auxin canalization or why NPA is poorly active in inhibiting smaller axillary buds of pea (Brewer et al., 2009). Here, despite demonstrating substantially reduced auxin transport over an extended period, there was still no effect of 0.1% NPA on axillary bud outgrowth in the strigolactone-deficient mutant (Fig. 4B). In wild type, branching was promoted below the NPA treatment site (Fig. 4B), consistent with other studies showing bud growth promotion by inhibition of auxin transport (Snyder, 1949; Tamas, 1987; Prasad et al., 1989; Tamas et al., 1989). The likely cause of this bud outgrowth may be reduced auxin content as reported by (Tamas, 1987; Morris et al., 2005), which would then enhance cytokinin and reduce strigolactone content, which would act to promote bud 15

16 outgrowth. As discussed below, it is likely that auxin transport was also reduced in these wild-type branches, again not consistent with the auxin-canalization model. To address the issue of whether auxin transport is actually required for bud outgrowth, we analyzed the effects of 1% NPA, which as we have shown above reduces auxin transport down to <0.4% of untreated wild type and below the level thought to be required for auxin canalization-driven branching inhibition (i.e., 30% of wild type; (Shinohara et al., 2013; Waldie et al., 2014)). In order to test whether a ring of 1% NPA on the main stem (e.g., Fig. 4) only affects auxin transport in the main stem and not in axillary branches, we promoted bud outgrowth in wild type by decapitation and simultaneously added the NPA in lanolin to the stem at a position three internodes below. In so doing, we were able to evaluate auxin transport in large branches that quickly developed above and below the NPA treatment site (Fig. 5C). 3 H-IAA added to the tips of the decapitation-induced branches, that were at least 4 cm in length, indicated that auxin transport was severely suppressed in these branches by the stem NPA treatment compared to control treatment. Little radiolabel from 3 H-IAA was recovered beyond the apical most 1 cm of branches above or below the NPA site, compared with branches of control decapitated plants, which transported large quantities of radiolabel (Fig. 5C). Again this study confirms that branches can grow out in plants with dramatically reduced polar auxin transport both within the stem and within the growing branch itself. Strigolactone treatment inhibits bud outgrowth regardless of auxin transport status If strigolactones regulate bud outgrowth by affecting auxin transport from axillary buds (Shinohara et al., 2013), then strigolactones should not be able to inhibit branching under dramatically reduced polar auxin transport. Alternatively, if strigolactones act more downstream or independently of auxin on axillary bud outgrowth, then they ought to be able to inhibit branching even when polar auxin transport is suppressed. Moreover, according to the direct action model, strigolactones should inhibit bud outgrowth regardless of auxin content or transport. Thus, we designed two experiments where branching would occur during auxin transport inhibition by 1% NPA. This included intact plants grown under growth promoting conditions of supplemented light (Fig. 6A) and decapitated plants with 16

17 reduced auxin content (Fig. 6B). We treated buds of these NPA-treated plants directly with the synthetic strigolactone, GR24, and observed a significant reduction in growth in both intact and decapitated plants (Fig. 6). About a 50% inhibition by GR24 occurred regardless of control or NPA treatment (Fig. 6B). Given the 17

18 magnitude of reduction in auxin transport by the 1% NPA (~0.4% of wild type), this full inhibitory effect of GR24 is unlikely due to altered polar auxin transport (Figs. 4 and 5). This is consistent with the finding above where strigolactone deficient mutants continue to branch when polar auxin transport is suppressed (Fig. 4B). It is also consistent with strigolactones acting in a more direct mechanism such as via regulation of BRC1, independent of auxin transport. Auxin-repressed branching requires strigolactones Having discovered that polar auxin transport is not required for strigolactone inhibition of bud outgrowth, we next tested the other important aspect that contributes to the auxin sink in the stem, auxin level (Shinohara et al., 2013). Auxin is produced predominantly in young leaves of growing apical shoots and then transported in the polar auxin transport stream Snow (1937). Following the canalization hypothesis, enhanced auxin content in the stem should suppress auxin flow from buds due to decreasing the auxin sink strength of the main stem (Fig. 1; (Shinohara et al., 2013)). Consequently, both wild-type and strigolactone mutant plants should therefore show some branching inhibition under high stem auxin content. Alternatively, if the enhanced auxin content acts only via enhanced strigolactones, the wild type should respond to enhanced auxin content, but a strigolactone mutant would not. We therefore took advantage of the dominant, gainof-function yucca mutant, yuc1d, which has enhanced auxin biosynthesis due to overproduction of a flavin monooxygenase (Zhao et al., 2001), to examine the effect of enhanced auxin levels in Arabidopsis. We compared the branching phenotype of yuc1d and the strigolactone deficient more axillary growth3 (max3) yuc1d double with wild type and strigolactone deficient max3. Branching was significantly suppressed in yuc1d compared with wild type (Fig. 7). In contrast, the double mutant combination appeared identical in increased branching to max3 single mutants (Fig. 7). This indicates that virtually all the branching repression induced by increased auxin production in yuc1d acts through strigolactones and therefore that there is little or no effect of decreasing the auxin sink strength of the main stem. 18

19 Discussion Here we discover that auxin transport does not correlate well with shoot branching. Strigolactone mutants of pea branch out during a period where there is no difference in polar auxin transport between strigolactone deficient plants and wild-type plants. 19

20 In addition, strigolactones can inhibit branching in vitro or in vivo without an apical auxin supply. Moreover, buds can be grown from a state of suppressed growth through to a well-developed branch with severely diminished polar auxin transport. All of these findings can be explained in terms of strigolactones acting directly on bud outgrowth (Fig. 1). However, these findings are not consistent with auxin canalization being required for bud outgrowth as the auxin transport/canalization hypothesis has been reported to require an auxin source and a directional flow of auxin (Shinohara et al., 2013; Waldie et al., 2014). Furthermore, consistent with the direct action model, enhanced auxin biosynthesis in the yuc1d mutant, and any resultant change in stem auxin content, requires strigolactones to inhibit branching. The new branches on NPA-treated plants showed some twisting and poor leaf expansion, but otherwise grew remarkably well (Figs. 4-6; Supplemental Fig. S2). This is consistent with established roles of auxin transport in tropic shoot growth and is reminiscent of the phenotypes of mutants that affect ABCB transporters (e.g., (Blakeslee et al., 2007)). The weak effect on branch morphology may be due to the relatively short length of treatment and prior organogenesis of axillary buds. Inhibited axillary buds already have several leaf primordia, small leaves and internodes, and contain vasculature, all laid down during initial bud formation (Pate, 1975; Stafstrom and Sarup, 2000; Kang et al., 2003). In contrast to strigolactones, NPA sometimes promotes bud outgrowth (e.g., Fig. 4B), sometimes reduces branch elongation (e.g., Fig. 6B and Brewer et al. (2009)), and sometimes has little effect (e.g., Morris et al. (2005); Mason et al. (2014)). This might have been explained by the canalization model where different states of auxin transport predict different branching outcomes (Shinohara et al., 2014). However, this model also predicts that greater than 30 percent of wild type is required for auxin canalization and bud outgrowth. The considerable growth of buds into branches during repression of auxin transport by 1% NPA (Figs. 5C and 6) demonstrates that auxin transport is not required for bud outgrowth. In contrast, the outgrowth of axillary buds and their inhibition by strigolactone under these conditions is consistent with the direct action model. The growth of branches during severely diminished auxin transport occurs with various tissue and vasculature development (Supplemental Fig. S2). This is consistent with previous findings. In the early studies of auxin transport and 20

21 phyllotaxis, which described the development of a pin-like structure in NPA treated plants, plants were grown for much longer periods on very high NPA concentrations (15-50 μm in media), and were able to develop some leaves and inflorescence stems with vascular tissues (Okada et al., 1991; Gälweiler et al., 1998; Mattsson et al., 1999). Some vasculature forms even in the presence of up to 160 μm NPA (Mattsson et al., 1999). These data and modeling experiments have led others to conclude that auxin canalization during vascular development can occur without polar auxin transport (Rolland-Lagan and Prusinkiewicz, 2005). In contrast with the findings here, other studies have shown that NPA partially inhibited bud outgrowth in Arabidopsis and rice strigolactone deficient mutants (Bennett et al., 2006; Lin et al., 2009). However, those results were obtained from plants exposed to continual, long-term NPA treatments in the growth medium that affected overall plant growth, and it is likely that NPA and/or associated suppressed auxin transport may have had secondary effects that affect branching. In this study, we have affected auxin transport specifically during bud release and outgrowth, and observed no reduced branching in pea strigolactone deficient mutants compared to control treated plants (Fig. 4B). In this and other studies where an auxin transport inhibitor has been supplied in a ring around the stem in wild-type plants, increased, rather than decreased, branching has been observed below the site of application ((Snyder, 1949; Prasad et al., 1989); Figs. 4B and 6A). This can be best interpreted in terms of effects of NPA on hormone levels below the site of application (Morris et al., 2005). Another important piece of evidence in support of the auxin transport/canalization model, and which needs interpreting in terms of the direct action model, is that a very low dose of strigolactone can promote branching in the transport inhibitor resistant3 (tir3) mutant (Shinohara et al., 2013). The highly pleiotropic tir3 mutant is hypersensitive to strigolactone-inhibition of root growth (Shinohara et al., 2013), which could increase available sucrose for shoot and axillary bud growth as well as decrease cytokinin contents, both of which could explain this enhanced shoot branching via direct action (Fig. 1). Moreover, the function of the TIR3 protein, otherwise known as BIG, is not fully established at the molecular level, despite being well known as affected in auxin transport (Leyser, 2010). In addition, the weakest 21

22 concentration at which max4 mutants respond to strigolactone for branching inhibition is the same concentration which is toxic to tir3, causing extremely stunted overall plant growth (Shinohara et al., 2013). If strigolactone instead functions almost entirely via direct action to regulate bud outgrowth, then why does strigolactone promote the cellular internalization of PIN auxin transporters (Shinohara et al., 2013) and reduce expression of auxin signaling and transport genes in the main stem (Bennett et al., 2006; Hayward et al., 2009; Waters et al., 2012)? Strigolactones have various functions in leaves, stems and roots, and some of these have been attributed to the effects of strigolactones on auxin transport (reviewed in (Brewer et al., 2013)). In addition, strigolactones were recently shown to alter tiller branch gravitropism in rice by inhibiting auxin biosynthesis (Sang et al., 2014). It is therefore possible that strigolactone regulation of PINs relates to these functions and not to the role of strigolactones in inhibiting bud outgrowth. Moreover, we have long proposed that strigolactone depletion enhances auxin levels and/or transport as part of a long-distance feedback mechanism (Beveridge et al., 1997; Beveridge, 2000; Beveridge et al., 2000; Foo et al., 2005; Brewer et al., 2009; Hayward et al., 2009). New flow of auxin from stimulated buds and branches presumably increases strigolactones levels and suppress cytokinin levels in the stem (Fig. 1 (Dun et al., 2009)) forming part of the competition system between buds and branches. This competition would be further amplified or accelerated under reduced strigolactone levels in buds and branches either by increased auxin biosynthesis/signaling or increased PINs at the cell membranes. Future studies should therefore explore the role of auxin export from buds in terms of within plant competition, as whilst this may not be a cause of bud outgrowth, it may be a mechanism that enhances competition among growing shoots (Snow, 1937; Crawford et al., 2010; Shinohara et al., 2013; Mason et al., 2014; Waldie et al., 2014). Materials and Methods Plant material, growth conditions and branching assays For garden pea (Pisum sativum) experiments, the lines used were wild-type cultivar 22

23 Torsdag (L107; Figs. 2, 3, 5 and 6; Supplemental Figs. S1 and S2) and mutants ramosus1-2t (rms1-2t; Fig. 2; Supplemental Fig. S1), rms5-3 (BL298; Fig. 3) in Torsdag and wild-type Parvus and mutant rms1-1 (WL5237) (Fig. 4; Supplemental Fig. S1). Unless otherwise stated, pea plants were grown at two per two-litre pot under glasshouse conditions (24 C day/18 C night, photoperiod extended to 18 h by incandescent lighting), as described by Ferguson and Beveridge (2009). Pots contained potting mix (7:2:1 pine bark fines:peat blend:sand (Fig. 4; Supplemental Fig. S1); or 7:3 0-5 mm composted pine bark:coco peat, with 1 kg/m 3 Yates Flowtrace, 1 kg/m 3 iron sulphate heptahydrate, 0.4 kg/m 3 superphosphate, 0.03 kg/m 3 copper sulphate and 1 kg/m 3 gypsum (Figs. 5 and 6)) with 2 g of Osmocote (Scotts, Europe); Flowfeed EX7 (Grow Force, Australia) was supplied weekly. Nodes were numbered acropetally from the first scale leaf as node 1, and lengths of lateral branches and buds were recorded using digital calipers. For the Arabidopsis thaliana experiment, Columbia-0 (Col-0), max3-11, yuc1d and double mutant plants were germinated and grown on UC potting mix supplemented with Flowfeed EX7 nutrients in an Arabidopsis Chamber (Percival) (22 C day/18 C night, 16 h day length). The number of rosette branches longer than 5 mm was counted when the plants were d old. A Student s t-test were used to test for statistical significance using GraphPad Prism Version 6.01 ( In vitro excised stem segment assays The classical excised stem segment assay (Chatfield et al., 2000) was adapted for pea. The agar growth medium contained 2.3 g/l Murashige and Skoog Basal Salt Mixture (MS salts; PhytoTechnology Laboratories M404), 0.5 g/l 2-(Nmorpholino)ethanesulfonic acid (MES; Sigma M2933) and 10 g/l agar (Sigma A1296), adjusted to ph 5.7, sterilized and poured into 10 x 10 cm plastic square petri dishes. A 2 cm trough was carefully cut removing agar across the middle of the dish. The remaining agar parts were then injected with hormone or control solution to give the required final concentrations and left to rest for 72 h. In experiments where auxin was supplied apically, the apical agar segments contained 0.56 µl/ml ethanol, with or without naphthaleneacetic acid (NAA). In experiments where strigolactone was 23

24 supplied basally, basal agar segments contained 1.11 µl/ml acetone, with or without GR24. For plant material, peas were grown at 4 or 5 per pot (6 x 6 x 9 cm) in a peat-sand blend potting mix (Green Fingers B2 Potting Mix; for 10 days in a growth cabinet (22 C, 18 h photoperiod). Plants with a fully expanded leaf at node 4 were used. These plants, Torsdag and rms1-2t mutant, have a repressed bud at node 3, which together with 1.5 cm of stem either side of the node, were excised, had their stipules, leaflets and majority of petiole removed (leaving only ~ 10 mm of petiole), and were carefully embedded as a bridge between the agars, two per petri dish. No sterilizing of pea tissue was required due to the absence of sucrose in the medium. Enclosed petri dishes were returned to the growth cabinet conditions, lying flat. The open tube method was adapted from above to provide less agar, less humidity and more airflow. Similar agar growth medium (3 g/l MS salts, 0.5 g/l MES, 10 g/l agar, ph 5.9), containing hormone or control solutions added immediately prior, was pipetted into sterilized 2 ml screw cap tubes, 1.5 ml per tube. Pea plant tissue was used, as above, and the cut stems were carefully embedded into the agar. The tubes were arranged upright with sticky tack (Bostik Blu-Tack) in empty 10 x 10 cm plastic square petri dishes, three per dish, as shown (Fig. 3A) and kept upright in the growth cabinet. Radiolabel assays The long-term effect of NPA on auxin transport and branching (Fig. 4 and Supplemental Fig. S1) was performed as follows; 17-d-old wild-type (WT) and rms1 strigolactone deficient pea plants were supplied with a lanolin ring, with or without 0.1% NPA, to the stem in the upper end of the oldest expanding internode, as per Brewer et al. (2009). Then, at certain times after NPA treatment, 3 H-IAA (American Radiolabeled Chemicals, Inc., St Louis, MO, USA) was added to the shoot apex and allowed time to flow down through the NPA treatment site. Radiolabel was then quantified from 0.5 cm stem segments, as per Brewer et al. (2009). The total radiolabel was the sum of all segments excluding the first segment and was measured as disintegrations per minute; DPM. Data from individual segments are shown in Supplemental Fig. S1. Branch length was measured at the node below the NPA treatment site. 24

25 Other methods are described in the text and legends except that for 3 H-NPA transport analyses, 370 kbq [2,3,4,5-3 H]-NPA ( 3 H-NPA; American Radiolabeled Chemicals, Inc., St Louis, MO, USA) in 60 μl (0.125 pmol) was added to 500 μl of 1% NPA for NPA transport experiments and the small amount of background (21.77 DPM in Fig. 4; Supplemental Fig. S1; 17 DPM in Fig. 5) was subtracted from the values before calculating mean and SEM. Radiolabeled transport samples were analyzed as per Brewer et al. (2009). Acknowledgments We thank K. Condon, R. Powell and B. Reid (The University of Queensland) for technical support, J. Mravec (University of Copenhagen) for providing the yuc1d line, S.M. Smith (The University of Tasmania), G.J. Mitcheson (University of Cambridge) and O. Leyser (University of Cambridge) for helpful discussions and J. Bertheloot (INRA, Angers), N. Leduc and S.C. Kerr (The University of Queensland) for comments on the manuscript. Figure legends Figure 1. Experimentally demonstrated relationships among key branching signals and bud outgrowth. These relationships have been used to support the direct action or auxin transport/canalization hypotheses of bud outgrowth. Auxin, which moves basipetally in the stem, inhibits cytokinin levels (Li et al., 1995; Tanaka et al., 2006) and promotes expression of strigolactone biosynthesis genes (Foo et al., 2005; Zou et al., 2006; Arite et al., 2007; Hayward et al., 2009). These hormones, strigolactone and cytokinin, in addition to sucrose (Mason et al., 2014), can move acropetally into buds and regulate branching (Dun et al., 2012). This is mediated at least partly via regulation of BRC1, which occurs rapidly (Mason et al., 2014) and/or without the need for protein synthesis (Dun et al., 2012). BRC1 encodes a transcription factor required for branching inhibition (Doebley et al., 1997; Aguilar-Martínez et al., 2007; Braun et al., 2012; Dun et al., 2012). These findings led to the direct action hypothesis of bud outgrowth. An alternative, but not necessarily mutually exclusive 25

26 hypothesis is that reduced auxin transport/canalization from axillary buds suppresses their outgrowth. In support of this, strigolactones can reduce auxin transport in the stem (Crawford et al., 2010) and enhance PIN cellular internalization (Shinohara et al., 2013). NPA used in this study inhibits the function of PIN and ABCB auxin transport proteins (Kleine-Vehn et al., 2006; Petrášek et al., 2006; Sauer et al., 2006; Blakeslee et al., 2007). Figure 2. Strigolactone applied directly can inhibit bud growth in pea in the classical isolated node, split-plate assay (A). B, Strigolactone (GR24, basal) requires auxin (NAA, apical) to inhibit pea bud growth. C, Strigolactone deficient rms1 mutant with 0 or 0.5 µm NAA (apical) and 0 or 10 µm GR24 (basal) in the media. B and C, Results are from the same experiment. D and E, Daily application of 10 µl of 1 µm GR24 alone directly to buds (Dun et al., 2013) suppressed growth in wild-type (WT) and the (E) rms1 strigolactone deficient mutant. Data are presented as mean ± SEM, provided the SEM exceeds the size of the symbol; n = 12. Figure 3. An open tube in vitro method (A) enables a basal strigolactone supply to inhibit bud outgrowth in the absence of an apical auxin supply. Segments similar to the experiment in Fig. 2 are placed in agar as shown. B and C, GR24 supplied basally reduced bud outgrowth in (B) wild-type (WT) and (C) rms5 strigolactone deficient plants. Data are presented as mean ± SEM; (B) n = 11 or 12; (C) n = 14 or 15. Figure 4. NPA (0.1%) acts profoundly on polar auxin transport for a long duration, and does not inhibit bud outgrowth. A, Total 3 H-IAA transported in the main stem of WT and rms1 plants was assayed at different intervals over several days after control or 0.1% NPA treatment. The individual data from each time point are shown in Supplemental Fig. S1. B, Bud outgrowth at the node below the NPA treatment site is shown for comparable plants. A and B, Most rms1 symbols are obscured by WT, NPA symbols. Data are all from the same experiment and are presented as mean ± SEM plotted on a log scale; (A) n = 5 or 6; (B) n = 7 to 10. The times shown are the harvest (A) or scoring time (B) from the NPA treatment at time zero. Figure 5. NPA acts rapidly and systemically to block polar auxin transport in the 26

27 main stem and axillary shoots. A, IAA transport from the shoot tip is rapidly blocked by 1% NPA treated about 30 cm below the highest expanded leaf. Plants were supplied with 3 H-IAA to the shoot-tip at 2 and 4 h after NPA or lanolin control treatment. Segments were harvested at 6 h. n = 6. B, Within 6 h NPA moves and accumulates to about 40 µm in the shoot tip 30 cm away. 1% NPA was supplemented with approximately 62 kbq 3 H-NPA per plant and was supplied as per (A). Adjacent 1 cm segments were taken either side of the treatment site. The apical bud and two lower 1 cm sections were also taken. n = 6. C, IAA transport inhibition persists systemically even though branches are able to form. Plants were decapitated below node 9 and 1% NPA added below node 6. 3 H-IAA was added to the apex of axillary shoot tips at node 7 (upper) or node 4 (lower). Branches were at least 4 cm in length (day 10). The total % transport to segments at 1-4 cm was 0.8 and 0.3% respectively for the upper and lower branches; at 1.5 cm, the fold-drop was 560 to 1100; n = 4. Controls from lanolin treated decapitated plants were combined node 7 and 4; n = 2. Segments were harvested after 2 h. A-C, Stems were lightly abraded at the treatment site before NPA or lanolin treatment in a ring around the stem. Data presented on a log scale are mean ± SEM, provided the SEM exceeded the size of the symbol. Figure 6. Strigolactone does not require auxin transport to inhibit branching. Plants were left intact under high light conditions (A) or were intact or decapitated under standard conditions (B). A and B, Plants were treated with 0% or 1% NPA midway between nodes 6 and 7 when the plants had 9 leaves expanded. 5 µl of 5 µm GR24 (Dun et al., 2009) applied to the bud at node 6 inhibited bud outgrowth compared with control-treated buds. Branches were measured on day 4. Data are mean ± SEM. A, Intact plants were encouraged to branch by supplementing the standard glasshouse light conditions with halogen lights which provided approximately an additional 140 to 200 µmol.m -2.sec -1 at the pot top and 40 cm above the pot top, respectively. Control plants, n = 7; NPA treated plants, n = 14 or 15. B, Plants were left intact or decapitated above node 9, and buds were treated with control or GR24 solution, daily. Intact control plants, n = 3; all other treatments, n = 9. A and B, n.d.; not determined. Figure 7. Reduced branching due to increased auxin biosynthesis requires 27

28 strigolactone production. The yuc1d mutant of Arabidopsis, which has increased auxin production (Zhao et al., 2001), also has decreased branching (P < 0.05, twotailed t-test; n = 7-12). This effect of decreased branching is not expressed in the strigolactone-deficient max3-11 background. 28

29 1 Supplemental Material Supplemental Figure S1. NPA acts profoundly on polar auxin ( 3 H-IAA) transport and for a long duration. (A-E) Radiolabel recovered after treatment of 3 H-IAA (measured as disintegrations per minute (DPM) and plotted on a log scale) to wildtype (WT) and rms1 strigolactone deficient plants. The 3 H-IAA was added to the shoot apex, as per Brewer et al. (2009). 17-d-old plants were supplied with a lanolin ring, with or without 0.1% NPA, to the stem in the upper end of the oldest expanding internode, as per Brewer et al. (2009). The times shown are the 3 H-IAA treatment time and the harvest time, starting from the NPA treatment at time zero. The 3 H-IAA treatment and harvest times were calculated in order to capture 3 H-IAA as was transported down through the NPA treatment site, and thus the 3 H-IAA treatment time increased as the stem grew longer. The distances shown are from immediately below the oldest unexpanded leaf. The dotted outline shows the span of the treatment site which expanded to about 1 cm. (C) The % of radiolabel transported in NPA-treated compared to control-treated plants from cm was 1.01% and 0.62% in wild-type and rms1 respectively; at 6 cm from the shoot tip, NPA-treated 1

30 plants had 0.41% and 0.30% of the radiolabel found in control-treated wild-type and rms1 plants. (A-E) As reported in other systems (Johnson and Morris, 1989; Prasad et al., 1989), the massive block in polar auxin transport occurred despite the continuing main stem growth, as evidenced here by the treatment zone increasing in distance from the shoot tip (A-E). Data are from the same experiment as Fig. 4 and are presented as mean ± SEM; (A-E) n = 5 or Supplemental Figure S2. NPA-treated wild-type plants develop branches after decapitation similarly to the lanolin control despite huge reductions in polar auxin transport. (A) Representative plants from Fig 6B and C showing the modified branch appearance of NPA treated plants. Both plants in the pot on the left were treated with 2

31 NPA and both plants in the pot on the right were treated with lanolin control (yellow). The treated bud is shown with an arrow with or without GR24 as shown. A ring of NPA on the stem does not stop bud outgrowth after decapitation. In contrast, strigolactone (GR24) reduces branching with or without NPA. The place of shoot tip removal (stump) and the sites of NPA and GR24 treatments are shown. (B) A hand longitudinal section of a node with inhibited buds from control plants stained for lignin by phloroglucinol. The main buds were removed during sectioning, but a connecting vein was still visible. A tiny accessory bud was still attached. (C) Node 11 from a plant grown in supplemented lighting as per Fig. 6A and treated with NPA in lanolin above node 6. Surprisingly well developed vasculature can be seen in the branch that had grown out to about 6 cm at that node. (D) A node from a decapitated control plant with a branch of similar length as in C. Tissues were dissected and fixed overnight in 6 parts ethanol to 1 part acetic acid, transferred to 100% ethanol, then transferred to 1% phloroglucinol in 6N HCl for 5-10 min until red color was visible and transferred back to 100% ethanol for observation and photography. Representative sections are shown from a range of 4-8 samples. Bars = 500 μm. Brewer PB, Dun EA, Ferguson BJ, Rameau C, Beveridge CA (2009) Strigolactone acts downstream of auxin to regulate bud outgrowth in pea and Arabidopsis. Plant Physiol 150: Johnson CF, Morris DA (1989) Applicability of the chemiosmotic polar diffusion theory to the transport of indol-3yl-acetic acid in the intact pea (Pisum sativum L.). Planta 178: Prasad TK, Hosokawa Z, Cline MG (1989) Effects of auxin, auxin-transport inhibitors and mineral nutrients on apical dominance in Pharbitis nil. J Plant Physiol 135:

32 Parsed Citations Aguilar-Martínez JA, Poza-Carrión C, Cubas P (2007) Arabidopsis BRANCHED1 acts as an integrator of branching signals within axillary buds. Plant Cell 19: Agusti J, Herold S, Schwarz M, Sanchez P, Ljung K, Dun EA, Brewer PB, Beveridge CA, Sieberer T, Sehr EM, Greb T (2011) Strigolactone signaling is required for auxin-dependent stimulation of secondary growth in plants. P Natl Acad Sci USA 108: Arite T, Iwata H, Ohshima K, Maekawa M, Nakajima M, Kojima M, Sakakibara H, Kyozuka J (2007) DWARF10, an RMS1/MAX4/DAD1 ortholog, controls lateral bud outgrowth in rice. Plant J 51: Balla J, Kalousek P, Reinöhl V, Friml J, Procházka S (2011) Competitive canalization of PIN-dependent auxin flow from axillary buds controls pea bud outgrowth. Plant J 65: Bangerth F (1989) Dominance among fruits/sinks and the search for a correlative signal. Physiol Plant 76: Bennett T, Sieberer T, Willett B, Booker J, Luschnig C, Leyser O (2006) The Arabidopsis MAX pathway controls shoot branching by regulating auxin transport. Curr Biol 16: Beveridge CA (2000) Long-distance signalling and a mutational analysis of branching in pea. Plant Growth Regul 32: Beveridge CA, Symons GM, Murfet IC, Ross JJ, Rameau C (1997) 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-transmissibie signal(s). Plant Physiol 115: Beveridge CA, Symons GM, Turnbull CG (2000) Auxin inhibition of decapitation-induced branching is dependent on grafttransmissible signals regulated by genes Rms1 and Rms2. Plant Physiol 123: Blakeslee JJ, Bandyopadhyay A, Lee OR, Mravec J, Titapiwatanakun B, Sauer M, Makam SN, Cheng Y, Bouchard R, Adamec J, Geisler M, Nagashima A, Sakai T, Martinoia E, Friml J, Peer WA, Murphy AS (2007) Interactions among PIN-FORMED and P- glycoprotein auxin transporters in Arabidopsis. Plant Cell 19: Braun N, de Saint Germain A, Pillot JP, Boutet-Mercey S, Dalmais M, Antoniadi I, Li X, Maia-Grondard A, Le Signor C, Bouteiller N, Luo D, Bendahmane A, Turnbull C, Rameau C (2012) The pea TCP transcription factor PsBRC1 acts downstream of strigolactones to control shoot branching. Plant Physiol 158: Brewer PB, Dun EA, Ferguson BJ, Rameau C, Beveridge CA (2009) Strigolactone acts downstream of auxin to regulate bud outgrowth in pea and Arabidopsis. Plant Physiol 150: Brewer PB, Koltai H, Beveridge CA (2013) Diverse roles of strigolactones in plant development. Mol Plant 6: 18-28