Connective auxin transport contributes to strigolactone-mediated shoot branching control independent of the transcription factor BRC1

Abstract

The shoot systems of plants are built by the action of the primary shoot apical meristem, established during embryogenesis. In the axil of each leaf produced by the primary meristem, secondary axillary shoot apical meristems are established. The dynamic regulation of the activity of these axillary meristems gives shoot systems their extraordinary plasticity of form. The ability of plants to activate or repress these axillary meristems appropriately requires communication between meristems that is environmentally sensitive. The transport network of the plant hormone auxin has long been implicated as a central player in this tuneable communication system, with other systemically mobile hormones, such as strigolactone and cytokinin, acting in part by modulating auxin transport. Until recently, the polar auxin transport stream, which provides a high conductance auxin transport route down stems dominated by the auxin export protein PIN-FORMED1 (PIN1), has been the focus for understanding long range auxin transport in the shoot. However, recently additional auxin exporters with important roles in the shoot have been identified, including PIN3, PIN4 and PIN7. These proteins contribute to a wider less polar stem auxin transport regime, which we have termed connective auxin transport (CAT), because of its role in communication across the shoot system. Here we present a genetic analysis of the role of CAT in shoot branching. We demonstrate that in Arabidopsis, CAT plays an important role in strigolactone-mediated shoot branching control, with the triple pin3pin4pin7 mutant able to suppress partially the highly branched phenotype of strigolactone deficient mutants. In contrast, the branchy phenotype of mutants lacking the axillary meristem-expressed transcription factor, BRANCHED1 (BRC1) is unaffected by pin3pin4pin7. We further demonstrate that mutation in the ABCB19 auxin export protein, which like PIN3 PIN4 and PIN7 is widely expressed in stems, has very different effects, implicating ABCB19 in auxin loading at axillary bud apices.

Fig 1. PIN347 are required for full bud activation in strigolactone mutants.

(A) Number of primary branches at terminal flowering for the genotypes indicated. (B) Representative plants of the genotypes in (A) at 6 weeks post-germination. Bar = 50 mm. In (A) The boxes span the first to third quartile and the line represents the median. The whiskers indicate the variability outside the upper and lower quartiles. Tukey’s HSD test was carried out after obtaining the least-square means for a linear model fitting the data and different letters indicate statistically significant differences at p < 0.05, n = 12–20.

Fig 2. Loss of PIN347 reduces stem auxin transport in strigolactone mutants without affecting PIN1.

(A) Bulk stem auxin transport in basal inflorescence stem internodes of 6-week old plants of the genotypes indicated. Transport was determined as basal accumulation of radiolabeled auxin, quantified as counts per minute (CPM) after 6-hours incubation in 1 μM ¹⁴C-IAA. (B) Quantification of PIN1-GFP in arbitrary units (A.U) at the basal plasma membrane of xylem parenchyma cells in longitudinal hand sections through basal inflorescence internodes of 6-week old plants of the genotypes indicated, homozygous for PIN1::PIN1-GFP. (C) Quantification of PIN3-GFP, PIN4-GFP and PIN7-GFP in arbitrary units (A.U.) at the basal plasma membrane of xylem parenchyma cells in longitudinal hand sections through basal inflorescence internodes of 6-week old max2 plants. The boxes span the first to third quartile and the line represents the median. The whiskers indicate the variability outside the upper and lower quartiles and outliers are indicated by individual points. Tukey’s HSD test were carried out after obtaining the least-square means for a linear model fitting the data and different letters indicate statistically significant differences at p < 0.05. For (A), n = 19–24. For B and C statistical analyses were carried our using the mean of 5 membranes from 4–8 plants per line.

Fig 3. PIN347 contribute to strigolactone-mediated bud-bud competition.

(A) Violin plots of the relative growth index (RGI) of 2-node explants of the genotypes indicated 10 days post decapitation, with (orange) or without (grey) 5 μM GR24 supplied basally. The RGI is the proportion of branch length in the longest branch. Black dots indicate the median value and the area of each plot represents the probability distribution of the values, n = 20–23. (B) Primary branch number of plants of the genotypes indicated grown axenically on ATS with (orange) or without (grey) 5 μM GR24 in long days, scored after 8 weeks, n = 14–21. (C) Dry weight of the plants represented in B. For B and C the boxes span the first to third quartile and the line represents the median. The whiskers indicate the variability outside the upper and lower quartiles. For A-C Tukey’s HSD tests were carried out after obtaining the least-square means for a linear model fitting the data and different letters indicate statistically significant results at p < 0.05.

Fig 4. ABCB19 contributes to stem auxin transport.

(A) Bulk stem auxin transport through 15 mm basal inflorescence internodes of 6-week old Col-0 and abcb19 plants. Transport was determined as basal accumulation of radiolabeled auxin, quantified as counts per minute (CPM) in the basal 5 mm of stem after 18 hours of incubation of the apical end 1 μM ¹⁴C-IAA. (B) Progression of a 10-minute pulse of 5 μM ¹⁴C-IAA applied to the apical end of 24 mm long basal inflorescence internodes of 6-week old Col-0 and abcb19 plants assessed at 30, 60 and 90 minutes after application of the pulse (left to right), shown as mean CPM in 2 mm sections of the stems. For A the boxes span the first to third quartile and the line represents the median. The whiskers indicate the variability outside the upper and lower quartiles and outliers are indicated by individual points. Tukey’s HSD test was carried out after obtaining the least-square means for a linear model fitting the data and different letters indicate statistically significant differences at p < 0.05, n = 24. For B the error bars represent the 95% confidence interval of the mean, n = 8 for each point. No significant differences were detected at any time point (unadjusted Wilcoxon rank-sum tests).

Fig 5. ABCB19 is required for full bud activation in strigolactone mutants.

(A) Number of primary branches at terminal flowering for the genotypes indicated. The boxes span the first to third quartile and the line represents the median. The whiskers indicate the variability outside the upper and lower quartiles and outliers are indicated by individual points, n = 20–24. (B) Representative plants of the genotypes in A at 6 weeks post-germination. Bar = 50 mm. (C) Violin plot of the relative growth index (RGI) of 2-node explants 10 days post decapitation with (orange) or without (grey) 5 μM GR24 supplied basally. The RGI is the proportion of branch length in the longest branch. Black dots indicate the median value and the area of each plot represents the probability distribution of the values, n = 21–24. (D) Primary branch number of 8-week old plants grown axenically in long day conditions on ATS with (orange) or without (grey) 5 μM GR24. The boxes span the first to third quartile and the line represents the median. The whiskers indicate the variability outside the upper and lower quartiles, n = 20–21. (E) Representative maximum projection image showing expression of ABCB19::ABCB19-GFP in longitudinal hand sections through basal inflorescence internodes of 6-week old wild-type stems, n = 22. (F) Representative maximum projection image showing expression of ABCB19::ABCB19-GFP in longitudinal hand sections through basal inflorescence internodes of 6-week old d14 mutant stems, n = 16. For A, C and D Tukey’s HSD tests were carried out after obtaining the least-square means for a linear model fitting the data and different letters indicate statistically significant differences at p < 0.05. For E and F the scale bar is 50 μm.

Fig 6. PIN347 and ABCB19 have contrasting effects on bud activation kinetics.

(A) Primary branch number at terminal flowering of plants of the genotypes indicated grown under long day growth conditions. n = 18–24. (B) Bulk stem auxin transport through 15 mm basal inflorescence internodes of 6-week old plants of the genotypes indicated. Transport was determined as basal accumulation of radiolabeled auxin, quantified as counts per minute (CPM) in the basal 5 mm of stem after 18 hours of incubation of the apical end 1 μM ¹⁴C-IAA, n = 24. (C) Mean number of active rosette branches over time following decapitation at day 0 for Col-0 (blue), abcb19 (orange), pin347 (green) and abcb19pin347 (grey) plants. Plants were grown under short day conditions for 4 weeks, shifted to long days to induce flowering and decapitated when the inflorescences reached 10 cm. The number of active rosette branches, defined as longer than 5 mm were counted daily. Error bars represent the 95% confidence interval of the mean. Non-overlapping error bars indicate statistical differences compared to wild type, verified using non-parametric tests comparing wild type and each mutant with a threshold of p < 0.05, with Holm-Bonferroni adjustment, n = 18–19. (D) Number of primary branches at terminal flowering for plants of the genotypes indicated grown under short day conditions for 4 weeks and then shifted to long days, n = 20–24. For A, B and D the boxes span the first to third quartile and the line represents the median. The whiskers indicate the variability outside the upper and lower quartiles and outliers are indicated by individual points. Tukey’s HSD tests were carried out after obtaining the least-square means for a linear model fitting the data and different letters indicate statistically significant differences at p < 0.05. For C, Holm-Bonferroni corrected Wilcoxon rank-sum tests were used to test between genotypes and time points as indicated in the text.

Fig 7. Shoot branching in brc1brc2 is independent of PIN347 but partially dependent on ABCB19.

(A) and (B) Primary branch number at terminal flowering in plants of the genotypes indicated grown continuously under long day conditions (grey) or under short day conditions for four weeks and then shifted to long day conditions (magenta), n = 20–24. The boxes span the first to third quartile and the line represents the median. The whiskers indicate the variability outside the upper and lower quartiles. Tukey’s HSD tests were carried out after obtaining the least-square means for a linear model fitting the data and different letters indicate statistically significant differences at p < 0.05.

Fig 8. PIN347 and ABCB19 mutation have limited impact on strigolactone response in brc1brc2 mutant branching.

(A) Primary branch number of plants of the genotypes indicated grown axenically on ATS with (orange) or without (grey) 5 μM GR24 in long days, scored after 8 weeks, n = 24–28. The boxes span the first to third quartile and the line represents the median. The whiskers indicate the variability outside the upper and lower quartiles. (B) Violin plots of the relative growth index (RGI) of 2-node explants of the genotypes indicated 10 days post decapitation, with (orange) or without (grey) 5 μM GR24 supplied basally, The RGI is the proportion of branch length in the longest branch. Black dots indicate the median value and the area of each plot represents the probability distribution of the values, n = 20–24. For A and B Tukey’s HSD tests were carried out after obtaining the least-square means for a linear model fitting the data and different letters indicate statistically significant differences at p < 0.05.

Fig 9. Model for bud activation.

Cartoon of a nodal Arabidopsis stem segment with associated bud (leaf not shown). The coloured arrows indicate the flow of auxin (green) and strigolactone (red). The blue shading indicates the polar auxin transport stream (PATS), with PIN1 dominating, and the orange shading represents the connective auxin transport (CAT), with PIN3, PIN4 and PIN7 dominating. The black lines indicate hypotheses involving either promotion (arrowheads) or repression (end lines). Specifically, we have previously proposed that positive feedback between auxin flux and PIN polarisation drives canalisation of auxin transport between the bud and the main stem, and this is necessary for sustained bud activation. We have also previously proposed that strigolactone inhibits canalisation by triggering PIN1 endocytosis. Data presented here suggest that this might also be true for PIN7, but that the transport activities of PIN3, PIN4 and PIN7 are all important to allow maximal bud activation in strigolactone-defective mutants. Although it is possible that ABCB transporters are also important in the stem, the very different effects of the abcb19 mutant compared to the pin347 triple mutant lead us to propose that an important site of action is in the bud, where they may contribute to auxin loading, and therefore canalisation of auxin transport out of the bud. Consistent with this idea, the branchy phenotype of brc1 mutants is partly suppressed in the abcb19 mutant background. This is consistent with the idea that BRC1 acts at least in part through increasing bud auxin loading, either independently of upstream of ABCB19.