Letter | Published:

Inhibition of cell expansion by rapid ABP1-mediated auxin effect on microtubules

Nature volume 516, pages 9093 (04 December 2014) | Download Citation


The prominent and evolutionarily ancient role of the plant hormone auxin is the regulation of cell expansion1. Cell expansion requires ordered arrangement of the cytoskeleton2 but molecular mechanisms underlying its regulation by signalling molecules including auxin are unknown. Here we show in the model plant Arabidopsis thaliana that in elongating cells exogenous application of auxin or redistribution of endogenous auxin induces very rapid microtubule re-orientation from transverse to longitudinal, coherent with the inhibition of cell expansion. This fast auxin effect requires auxin binding protein 1 (ABP1) and involves a contribution of downstream signalling components such as ROP6 GTPase, ROP-interactive protein RIC1 and the microtubule-severing protein katanin. These components are required for rapid auxin- and ABP1-mediated re-orientation of microtubules to regulate cell elongation in roots and dark-grown hypocotyls as well as asymmetric growth during gravitropic responses.


Auxin is crucial for diverse developmental processes and growth responses3. One major effect of auxin is cell expansion1, which relies on the coordinated activities of cellular processes involving the cytoskeleton2. When cells elongate, cortical microtubules are arranged perpendicularly to the axis of cell elongation (transverse microtubules), whereas a longitudinal alignment accompanies growth inhibition2. The dynamic nature of microtubules provides the flexibility to rearrange them into different arrays4, enabling growth changes downstream of different signals such as gravity5 or light6. Many of these signalling pathways converge on auxin7; therefore its action upstream of microtubules translates different signals into growth responses1. Nonetheless, whether auxin acts directly on microtubule arrangement and by which mechanism remain unclear.

Microtubules co-align approximately perpendicularly to the elongation axis in roots and dark-grown hypocotyls4. The transition zone of primary root and the elongation zone of etiolated hypocotyl (Extended Data Fig. 1a) are controlled by auxin to determine their respective growth rates7. We visualized cortical microtubules in transgenic lines expressing microtubule-associated protein 4 fused with green fluorescent protein (MAP4–GFP)8 or α-tubulin 6 fused with red fluorescent protein (TUA6–RFP)9, and classified cells on the basis of prevalent microtubule arrangement into four groups (Fig. 1a). In root, microtubules were mainly transverse and underwent visible realignment within 10 min after application of the synthetic auxin naphthalene-1-acetic acid (NAA), leading to partial longitudinal re-orientation after 1 h (Fig. 1a). Comparable effects were observed irrespective of the microtubule reporter following treatment with the natural auxin indole-3-acetic acid (IAA) (Extended Data Fig. 1b–f). The same effects were observed in etiolated hypocotyls (Fig. 1b) although at a higher auxin concentration, consistent with known auxin response maxima of aerial tissues at higher doses10. Re-orientation of microtubules is not always homogenous, as revealed by the deviated angle of individual microtubules. Transverse microtubules (90 ± 30°) decreased at the expense of increasingly oblique and longitudinal microtubules (0–60°/120–180°) following auxin treatment (Extended Data Fig. 1c, f).

Figure 1: Auxin induces microtubule re-orientation.
Figure 1

a, b, MAP4–GFP visualization of microtubule orientation in roots (a) and etiolated hypocotyls (b) by time-lapse imaging following 100 nM NAA or 10 µM IAA treatment, respectively. The cartoon illustrates the four categories of microtubule orientation. c, EB1b–GFP visualization of microtubule trajectories at the upper side (US) and lower side (LS) of 90° gravistimulated roots. EB1b trajectories were quantified as transverse (90 ± 30°) or longitudinal (0–60°/120–180°) microtubules. In all panels, average values are shown, error bars are s.e.m. and Student’s t-test was calculated for transverse microtubules (*P < 0.05, **P < 0.001). Scale bars, 5 μm (a, c) and 10 µm (b).

Treatment with the weak auxin analogue11 of 2-NAA showed a very weak effect on microtubule rearrangement whereas acidic pH led to massive disruption and random orientations of microtubules (Extended Data Fig. 1g, h). Both treatments confirm the specificity of active auxins on microtubule orientation.

In roots, gravistimulation induces asymmetric auxin redistribution: lower levels at the upper side correlate with cell elongation and higher levels at the lower side with inhibition of cell expansion12. We assessed the effect of endogenous auxin redistribution on microtubule arrangement by tracking trajectories of end binding 1b (EB1b), a protein that preferentially accumulates at the growing plus ends of microtubules13. After 90° root re-orientation, transverse microtubules were maintained in the upper-side cells whereas at the lower side, microtubules longitudinally re-oriented within 10 min, preceding growth inhibition (Fig. 1c). Auxin distribution reported by the auxin response reporter DII-VENUS14 during gravitropism confirmed higher auxin response at the lower side than the upper side (Extended Data Fig. 1i–k). Thus auxin application or endogenous auxin redistribution promotes longitudinal microtubule orientation, correlating with auxin inhibiting cell elongation.

Next we addressed the mechanism by which auxin influences microtubule orientation. The signalling pathway of the nuclear-localized auxin co-receptors transport inhibitor response 1/auxin-related F-box (TIR1/AFB)–auxin/IAA (AUX/IAA) regulates gene transcription and mediates many plant developmental effects3. On the other hand, the ABP1 pathway regulates transcriptional15 and non-transcriptional responses such as interdigitation of pavement cells16 or clathrin-dependent endocytosis17.

In root cells of tir1-1 afb1-1 afb2-1 afb3-1 (ref. 18), we observed normal transverse microtubules; however, they were less sensitive to auxin treatment (Extended Data Fig. 2a, b). Functional inactivation of ABP1 in the conditional SS12S and SS12K lines19,20 resulted in microtubule orientation defects with increasing time of ABP1 inactivation (Extended Data Fig. 2c–f), whereas abp1-5, harbouring a point mutation in the auxin binding pocket susceptible to impair auxin binding16, did not show altered microtubule arrangement (Fig. 2a). Nonetheless, in both cases a severe reduction in microtubule re-orientation was observed in response to auxin (Fig. 2a). Short-term ABP1 inactivation (8 h) in dark-grown hypocotyls also impaired microtubule responsiveness to auxin without affecting the microtubule organization per se (Fig. 2b). This observation, together with a similar effect of abp1-5 mutation, confirms that the microtubule insensitivity to auxin does not result from pre-existing microtubule alteration in these lines. Growth of ABP1-inactivated roots was previously reported to be auxin resistant20, strengthening the correlation between the effect of auxin on microtubule orientation and inhibition of cell elongation.

Figure 2: ABP1 is required for auxin regulation of microtubule re-orientation.
Figure 2

a, b, MAP4–GFP visualization of microtubule re-orientation in wild-type (WT), abp1-5, SS12S/K root (a) and hypocotyl (b) induced with ethanol vapours for 48 h (a) and 8 h (b) and following 60 min treatment with dimethyl sulphoxide (DMSO), 100 nM NAA (a) or 10 μM IAA (b). The ratio of transverse microtubules in DMSO versus NAA treatment is indicated above the charts (a, b). In all panels, average values are shown and error bars are s.e.m. and Student’s t-test was calculated for transverse microtubules (**P < 0.001). Scale bars, 5 µm (a) and 10 µm (b).

We also explored whether ABP1 function is required for the microtubule rearrangement and differential growth response in root gravitropism. Following 90° root re-orientation, ABP1-inactivated lines showed much weaker microtubule rearrangement at the lower side than wild type (WT) (Extended Data Fig. 3a); however, gravity-induced asymmetric auxin distribution (monitored by DII-VENUS) was also less pronounced in ABP1-inactivation lines (Extended Data Fig. 3b, c compared with Extended Data Fig. 1i–k). In line with these observations, ABP1-deficient lines showed defects in gravitropic response (Extended Data Fig. 3d).

Given the mutual impact of the TIR1 and ABP1 pathways15, it is difficult to distinguish between direct and indirect effects; however, because short-term ABP1 inactivation strongly impairs auxin-mediated microtubule re-orientation, this favours the ABP1 pathway as the primary mechanism. To gain further mechanistic insights into the effect of auxin on microtubule and transcriptional compared with fast, non-transcriptional responses, we studied the kinetics of auxin’s effect on microtubule re-orientations by following EB1b movement at 15 s intervals, and its trajectories. NAA did not influence the speed of EB1b movement (Extended Data Fig. 4a, b) but gradually increased oblique and longitudinal EB1b tracks (0–60°/120–180°) after 75 s (Fig. 3a and Supplementary Videos 1 and 2). To provide quantitative measures of microtubule rearrangements, we developed an image analysis tool that automatically assigned EB1b trajectories to transverse (depicted as blue area) or longitudinal (red area) directions, which revealed changes in EB1b track orientations as soon as 30 s after auxin treatment (Extended Data Fig. 4c). Both types of kinetic analysis revealed a very fast responsiveness of microtubules to auxin, making transcriptional regulation in this process unlikely. This is consistent with the lack of interference on auxin-induced microtubule rearrangement by blocking transcription with cordecypin (Extended Data Fig. 4d). Involvement of the ABP1-dependent response pathway is also supported by the inhibition of the TIR1/AFB pathway with α-(phenylethyl-2-one)-IAA (PEO-IAA)21, which did not prevent the effect of IAA on microtubules (Extended Data Fig. 4e).

Figure 3: The effect of auxin on fast responsiveness of microtubule rearrangement is dependent on ABP1.
Figure 3

a, b, Projections of EB1b–GFP in WT (a) or SS12S (b) roots (left) and quantification (right) from every 15 s acquisition during 10 min (Supplementary Videos 1, 2, 3 and 5) following DMSO or 100 nM NAA application (n = 10). Blue and red strips represent transverse (90 ± 30°) and oblique/longitudinal (0–60°/120–180°) directions, respectively. In all panels, average values are shown and error bars are s.e.m. determined by Student’s t-test (*P < 0.05, **P < 0.001). Scale bars, 5 μm.

To investigate further the rapid ABP1-dependent effect on microtubules, we introduced EB1b–GFP in ABP1-inactivated lines. Trajectories of EB1b following ABP1 inactivation showed more oblique and longitudinal microtubules than WT (Fig. 3b, Extended Data Fig. 4f–i and Supplementary Videos 3 and 4). After NAA treatment, no consistent switch of EB1b trajectories to longitudinal directions, but only a few stochastic changes, were observed (Supplementary Videos 5 and 6, Fig. 3b and Extended Data Fig. 4f–i compared with WT in Fig. 3a and Extended Data Fig. 4c). By a complementary approach, we analysed the effect of auxin on EB1b trajectories in inducible ABP1 gain-of-function lines (XVE  ABP1-OE). Overexpression of ABP1–GFP in the presence of oestradiol increased the overall amount of ABP1 (Extended Data Fig. 5a–c), leading to an enhanced microtubule re-orientation in response to auxin (Extended Data Fig. 5d). In contrast, abp1-5 exhibited delayed microtubule re-orientation in response to auxin (Extended Data Fig. 5d). Overall these results strongly suggest that the fast, non-transcriptional effect of auxin on microtubule re-orientation is mediated primarily by ABP1-dependent signalling.

Next we addressed the downstream mechanism by which ABP1 mediates the effect of auxin on microtubule arrangement. Although auxin induces calcium transients22, the manipulation of exogenous calcium had very different effects on microtubule arrangements compared with auxin (Extended Data Fig. 6). Then we tested the downstream components of the ABP1 pathway: the ROP6 GTPase, its effector RIC1 (ref. 16) and its downstream component microtubule-severing protein katanin (KTN1)23,24. We analysed microtubules in rop6-1, ric1-1 and ktn1 mutants. Compared with WT, roots of rop6-1 and ric1-1 showed almost normal transverse microtubules but were much less auxin responsive (Fig. 4a and Extended Data Fig. 7a). Double mutants (SS12S ric1-1, SS12K ric1-1), with ABP1 inactivation, exhibited the SS12K/S root phenotype15 but ric1-1 microtubule arrangement (Fig. 4a and Extended Data Fig. 7a–d), consistent with the reported action of RIC1 downstream of ABP1 in early responses25. The ktn1 mutant exhibited a severe microtubule phenotype, with completely random microtubules in roots compromising further analysis. Microtubules were less disturbed in dark-grown ktn1 hypocotyls, allowing investigation of the response to auxin and the genetic interaction with ABP1 (Extended Data Fig. 7e). Rapid auxin-induced re-orientation of microtubules was impaired in ktn1 (Fig. 4b). Inactivation of ABP1 in ktn1 (SS12K ktn1) resulted in a microtubule pattern similar to SS12K and conferred insensitivity to auxin (Fig. 4b). These data suggest that KTN1 is required for microtubule re-orientation in response to auxin but that other microtubule-associated components might be involved as well. The present data and the crucial roles of RIC1 and KTN1 in the control of microtubule architecture and crossover6,24 support the idea that auxin-dependent ABP1 signalling might act through Rho GTPases and RIC effectors on critical targets such as KTN1 for guiding microtubule orientation.

Figure 4: Auxin–ABP1 control microtubule arrangement through downstream ROP6–RIC1 and involvement of KTN1.
Figure 4

a, Microtubule orientation and quantification in roots of WT, rop6-1, ric1-1, SS12S/K ric1-1 following 60 min of DMSO or 100 nM NAA application. b, Microtubule orientation and quantification in 24 h ethanol-induced hypocotyls of WT, SS12K, ktn1 and SS12K ktn1 following 60 min of DMSO or 10 μM IAA application. The ratio of transverse microtubules in DMSO versus NAA/IAA treatment is indicated above the charts (a, b). In all panels, average values are shown and error bars are s.e.m. and Student’s t-test was calculated for transverse microtubules (**P < 0.001). Scale bars, 5 µm (a) and 10 µm (b).

The effects of auxin on cell expansion are not mechanistically well understood; however, it is clear that sustainable growth control requires fast, non-transcriptional auxin effects and regulation of transcription3. Here we show that auxin signalling targets immediate changes in microtubule orientation. Auxin, by a non-transcriptional effect requiring ABP1 and downstream signalling, influences microtubule re-orientations within minutes, leading to changes in microtubules from transverse to longitudinal orientation to inhibit cell elongation. It is possible that branching of the ABP1 signalling for microtubule realignment occurs at the level of Rho GTPases, their RIC effectors and downstream targets such as KTN1. It remains unclear how this newly identified effect of ABP1 signalling on microtubules is related to other actions of the ABP1 pathway, such as inhibition of clathrin-mediated endocytosis, in particular given the differences in efficiencies of synthetic and natural auxins on both processes26. The relationship between microtubules and endocytosis in plants is still unclear; however, given the observations from animals that clathrin controls microtubule acetylation27, auxin-mediated defects in clathrin organization might cause the mis-modification of microtubules and thus disturb their crossover. On the other hand, Rho GTPases acting downstream of ABP1 might influence clathrin and microtubule functions via distinct effectors. Our observations on the rapid regulation of microtubule arrangement by the ABP1-mediated auxin signalling pathway provide insight into the long-sought molecular mechanism by which a major plant hormone exerts its fast effect on plant cell growth.


Material and growth conditions

Seeds were sown on 0.8% agar containing 1/2 Murashige and Skoog medium with sucrose for root experiments and 1% agar containing 1/2 Murashige and Skoog medium without sucrose for hypocotyl experiments. For root analyses, seedlings were grown under 16 h light/8 h dark photoperiod; for hypocotyl analyses, seedlings were grown in darkness at 22 °C for 4 days. Ethanol induction of conditional lines for ABP1 (ref. 19) (named SS12S or SS12K) was performed by exposure of the seedlings to ethanol vapour for various times as indicated for the different assays. We routinely used 48 h of exposure to vapours generated from 500 µl of 5% ethanol in light-grown root and 8 h of exposure to vapours from 500 µl of 8% ethanol for dark-grown hypocotyls (except particular induction times annotated in the figure legends). In each single experiment, WT and ABP1 conditional lines were always grown on the same plate and exposed to ethanol in identical conditions. The ric1-1 and rop6-1 lines were in the Wassilewskji (WS) background16, the other lines were derived from the Columbia (Col-0) background. The T-DNA insertion mutant of KTN1 (SAIL_343_D12) was provided from the Arabidopsis Information Resource. Offspring of the double mutants were analysed by PCR amplification as described previously17,25. For the generation of inducible overexpressing ABP1 lines (XVE  ABP1-OE), ABP1–GFP was cloned by inserting GFP into ABP1 after the glycine 120 by primer extension PCR with two glycines flanking the GFP coding sequence17; then the fragment of full-length ABP1 genomic DNA with GFP was cloned into the Gateway vector pMDC7B using Gateway cloning technology (http://www.invitrogen.com). At least three independent lines were used for the analysis (H.R. et al., unpublished observations). MAP4–GFP8, TUA6–RFP9 and EB1b–GFP13 were used as microtubule markers, and DII-VENUS was used as an early auxin response sensor14.

Confocal microscopy observation

For observations of microtubule orientation in root, cells in the transition zone of the primary root were visualized by vertical Zeiss LSM 700 confocal laser scanning microscope (with a ×63 objective using hair gel as immersion medium). For observations in dark-grown hypocotyls, cells in the elongation zone were visualized by a Nipkow Spinning Disk confocal system (Yokogawa CXU-X1-A1) mounted on a Nikon Eclipse Ti E inverted microscope (with a ×40 oil immersion objective). For dark-grown hypocotyls, all manipulations were performed under green light to avoid any light effects on microtubule re-orientation before confocal imaging. For all the visualization of auxin-treated seedlings, 1 min manipulation time was needed to apply auxin before imaging. The videos of EB1b trajectory were performed by spinning disc confocal system with ×63 water immersion objective. Videos were acquired with 300 ms exposure time every 500 ms for 10 min. The settings of excitation and detection were as follows: for GFP, 488 nm, 505–550 nm; for VENUS, 514 nm, 527 nm; for RFP, 587 nm, 610 nm. All the images in a single experiment were captured with the same setting. In experiments where rapid microtubule re-orientation was imaged, the seedlings were gently placed in chamber slides, covered by 0.8% agar containing 1/2 Murashige and Skoog medium or placed on liquid 1/2 Murashige and Skoog medium glass slides, and then placed on a vertical Zeiss LSM 700 confocal laser scanning microscope along their original direction of growth. When gravistimulation started, slides were rotated with a rotatable stage by 90°, and epidermal cells at the upper and lower layers were individually visualized. For live imaging on DII-VENUS, immediately after the beginning of re-orientation, the seedlings were scanned every 10 min for 1 h to follow the evolution of the DII-VENUS signal. Additionally, time-lapse visualization was performed as the time indicated (single prime (′), minute; double prime (′′), second) in figures and all the experiments were repeated at least three times.

Chemical treatment

NAA (Sigma), IAA (lab 3) and PEO-IAA21 were dissolved in DMSO; IAA (Sigma, lab 1, 2) and 17-β-oestradiol (Sigma) were dissolved in ethanol; and cordycepin (Sigma) was dissolved in double-distilled H2O. For oestradiol induction, 3-day-old seedlings were transferred to 2 µM 17-β-oestradiol containing 1/2 Murashige and Skoog solid medium and were grown vertically for 12 h or 48 h before observation or protein/RNA extraction. Non-induced XVE  ABP1-OE was used as a control. For the experiments with short-term chemical treatments, 4-day-old seedlings were gently placed on chemical-containing plates according to their original direction of growth. For combined treatments with PEO-IAA or cordycepin and 1 µM IAA, 4-day-old seedlings grown vertically on 1/2 Murashige and Skoog medium were first transferred to plates containing 10 µM PEO-IAA or 400 µM cordycepin for 30 min pre-treatment (vertically grown), then transferred on the same medium plus 1 µM IAA. For the treatment with different concentrations of CaCl2, 1/2 Murashige and Skoog medium with vitamins but without CaCl2 was used compared with the same medium with added 1 or 10 mM CaCl2 or with a standard 1/2 Murashige and Skoog medium (1.5 mM CaCl2). For short-term treatments, seedlings were mounted on 0.8% agar 1/2 Murashige and Skoog medium chamber slides or liquid 1/2 Murashige and Skoog medium glass slides containing the indicated concentration of auxin, then were immediately imaged.

Gravitropic response

Four-day-old vertically grown seedlings, under light conditions, were re-oriented by 90°, and the angles deviating from the original vertical growing direction of primary roots (defined as 0°) were tracked every 30 min until 8 h.

Quantification methods

1. To quantify the percentage of different cell types, the number of cells at a certain microtubule orientation type was calculated as a percentage of the total measured cells in each root, and different cell types were divided into transverse, oblique, random and longitudinal groups on the basis of prevalent microtubule arrangement (Fig. 1a). At least 15 different roots were analysed and at least eight cells were quantified in each root. For hypocotyls, cells were classified into a microtubule orientation group depending on the overall angles of microtubules for each cell (0–22.5°/157.5–180° for longitudinal, 22.5–67.5°/112.5–157.5° for oblique and 67.5–112.5° for transverse). Angles of microtubules were determined both manually and by using the plug-in of the ImageJ FibrilTool28. At least ten different hypocotyls were analysed and 100 hypocotyl cells were quantified.

2. To quantify the oriented angles of microtubules in root cells, the number of microtubules at a certain angle range was calculated as a percentage of the total microtubules measured. The longitudinal direction parallel to and going along the growth axis was defined as 0°, the transverse direction perpendicular to the growth axis was defined as 90°, the longitudinal direction parallel to but going opposite the growth axis was defined as 180° and angles deviating from the longitudinal direction (0°) were measured. Different microtubules angles were divided into two categories: transverse direction (90 ± 30°), oblique and longitudinal direction (0–60°/120–180°). The measurement of deviated angles of microtubules was processed by ImageJ combined with Matlab. At least eight cells of different roots were randomly selected, and in each cell at least 100 microtubules were quantified.

3. For the ‘time-stack’ images of EB1b trajectory (Fig. 1c), the video of EB1b trajectory was taken every 15 s per picture in a vertical Zeiss LSM 700 confocal laser scanning microscope for 10 min in total, every 3 s per picture in a spinning disc for 10 min in total. The maximum intensity projections of all frames were stacked by ImageJ (Image > Stack > Z-project). Then, the EB1b trajectory was quantified as described previously (method 2).

4. To quantify the rate of EB1b movement, the method has previously been described in details29. In brief, the manual quantification of the EB1b trajectory (Fig. 3a, b) was performed according to the directions of each single EB1b track followed with strips. As described before, the longitudinal direction parallel to and going along the growth axis was defined as 0° and the transverse direction perpendicular to the growth axis was defined as 90°. Blue strips represent the transverse direction (90 ± 30°), red strips represent the oblique and longitudinal direction (0–60°/120–180°). The deviated angle of each EB1b strip from the transverse direction was measured by ImageJ. At least ten cells of different roots were randomly selected, and in each cell at least 100 EB1b tracks were quantified.

5. Additionally, automated quantification of the directionality of EB1b trajectory used the following steps. To determine the direction of EB1b trajectory within the cells, it is necessary to eliminate the apparent motion caused by the growth of the seedling. In a first step image, stabilization was therefore performed by manually locating a prominent feature, for example the boundary between two specific cells in all frames of the video. Translating each frame accordingly then kept the prominent feature stationary. This proved more accurate than various automated image registration methods.

The temporal resolution of the videos was generally not sufficient to track individual particles over multiple frames. Thus particle tracking velocimetry resulted in a very sparse vector field. To obtain a more complete vector field that better represented the transport direction of all reporter proteins, the Horn–Schunck method of estimating optical flow was implemented and subsequently used for the analysis. In this method a regularization parameter enforces a smooth vector field which then assigns a vector of motion (speed and direction) to each pixel and frame of the video, filling in the missing flow information from the motion of the neighbouring pixels. The regularization parameter was adjusted to match the visual assessment of the flow. The resulting vector field was afterwards averaged over several frames.

The dominant direction of motion at each pixel was calculated only if the speed reached a certain threshold, which was determined by comparison with the maximum intensity projection of the video where regions with no or little protein transport appeared darker. The values chosen for the regularization parameter and the minimum speed were kept constant throughout the analysis of all videos.

The direction of motion of the remaining pixels, for which the speed was above the threshold value, was sorted into two classes: blue indicates a direction of transport of 90° ± 30°, red corresponds to 0–60°/120–180°. Finally, the areal fraction of the two classes was calculated.

6. To quantify the fluorescence, DII-VENUS fluorescence was pseudo-coloured, and the epidermal and cortical cells of similar developmental stages with symmetrical areas in roots were selected for quantification. The intensity of nuclei was extracted using the ROI tool of Fiji software (http://fiji.sc/wiki/index.php/Fiji), and the sum of fluorescence in the upper- and lower-side cells of each seedling was individually quantified14.

Quantitative PCR with reverse transcription

Whole RNA of seedlings was extracted using an RNeasy Mini Kit (Qiagen) and complementary DNA was synthesized using an iScript cDNA Synthesis Kit (Bio-rad). Quantitative PCR with reverse transcription (qRT–PCR) used LightCycler 480 SYBR Green I Master (Roche) following the recommendations of the manufacturer. qRT–PCR was performed in 384-well optical reaction plates by using Perkin Elmer Janus Robot and Roche Lightcycler 480 with heat for 10 min to 95 °C to activate hot-start Taq DNA polymerase, followed by 40 cycles of denaturation for 60 s at 95 °C and annealing extension for 60 s at 58 °C. Expression levels were normalized to the expression levels of ACTIN8. Specific primers of ABP1 (F: TCGTCGTCTTTTCCGTCGCG; R: TTGGCAAGCCATTGATGGGACA) and ScFv (F: TTACTGGATGCACTGGGTGA; R: AAGACTGACAGGCAGGGAGA) were used for gene expression as previously reported19,20. Two biological repeats were analysed in triplicate. qRT–PCR relative quantification was performed on Lightcycler 480 software combined with the local website (http://qpcr.ista.local).

Protein analysis

Proteins were extracted from WT and XVE  ABP1-OE 10-day-old seedlings induced or not with oestradiol by grinding at 4 °C with a mortar in the extraction buffer (50 mM Tris-HCl, pH 7.5, 1 mM EDTA, 100 μM MgCl2, 5 mM sodium ascorbate, 500 mM sucrose, phosphatase and protease inhibitors). Samples were spun at 5,000g for 10 min at 4 °C to remove cell debris. Supernatants were centrifuged at 50,000g for 60 min at 4 °C to pellet the total membrane fraction. Pellets were re-suspended in microsomal buffer (25 mM Tris-HCl pH 6.8, 0.5 mM EDTA, 0.1 mM MgCl2, 330 mM sucrose, 10% glycerol, protease inhibitor cocktail) for the following SDS–PAGE analysis. Protein loading was controlled by Coomassie brilliant blue staining. Protein gel blot used the mAb12 mouse monoclonal antibody30 to detect ABP1 protein. Protein amount was specified by GelQuant.NET software and normalized according to sample loading. Two biological repeats were analysed in duplicate.


For all quantitative data, error bars indicate s.e.m. The number of analysed samples is indicated as n from at least three biological replicates, and statistical analyses used Student’s t-test where * or ** correspond to P < 0.05 or 0.001, respectively.


  1. 1.

    Cellular responses to auxin: division versus expansion. Cold Spring Harb. Perspect. Biol. 2, a001446 (2010)

  2. 2.

    & Microtubules, MAPs and plant directional cell expansion. Trends Plant Sci. 13, 303–310 (2008)

  3. 3.

    & Mechanism of auxin-regulated gene expression in plants. Annu. Rev. Genet. 43, 265–285 (2009)

  4. 4.

    & Cortical microtubule arrays in the Arabidopsis seedling. Curr. Opin. Plant Biol. 11, 94–98 (2008)

  5. 5.

    The cytoskeleton and gravitropism in higher plants. J. Plant Growth Regul. 21, 120–136 (2002)

  6. 6.

    et al. A mechanism for reorientation of cortical microtubule arrays driven by microtubule severing. Science 342, 1245533 (2013)

  7. 7.

    , & Hormonal regulation of root growth: integrating local activities into global behaviour. Trends Plant Sci. 17, 326–331 (2012)

  8. 8.

    et al. A GFP-MAP4 reporter gene for visualizing cortical microtubule rearrangements in living epidermal cells. Plant Cell 10, 1927–1940 (1998)

  9. 9.

    , , & Cortical microtubule arrays undergo rotary movements in Arabidopsis hypocotyl epidermal cells. Nature Cell Biol. 9, 171–175 (2007)

  10. 10.

    , & Hormonal interactions in the control of Arabidopsis hypocotyl elongation. Plant Physiol. 124, 553–562 (2000)

  11. 11.

    et al. Defining the selectivity of processes along the auxin response chain: a study using auxin analogues. New Phytol. 200, 1034–1048 (2013)

  12. 12.

    Directional gravity sensing in gravitropism. Annu. Rev. Plant Biol. 61, 705–720 (2010)

  13. 13.

    , , & EB1 reveals mobile microtubule nucleation sites in Arabidopsis. Nature Cell Biol. 5, 967–971 (2003)

  14. 14.

    et al. A novel sensor to map auxin response and distribution at high spatio-temporal resolution. Nature 482, 103–106 (2012)

  15. 15.

    et al. Auxin-binding protein 1 is a negative regulator of the SCFTIR1/AFB pathway. Nature Commun. 4, 2496 (2013)

  16. 16.

    et al. Cell surface- and rho GTPase-based auxin signaling controls cellular interdigitation in Arabidopsis. Cell 143, 99–110 (2010)

  17. 17.

    et al. ABP1 mediates auxin inhibition of clathrin-dependent endocytosis in Arabidopsis. Cell 143, 111–121 (2010)

  18. 18.

    et al. Plant development is regulated by a family of auxin receptor F box proteins. Dev. Cell 9, 109–119 (2005)

  19. 19.

    et al. Conditional repression of AUXIN BINDING PROTEIN1 reveals that it coordinates cell division and cell expansion during postembryonic shoot development in Arabidopsis and tobacco. Plant Cell 20, 2746–2762 (2008)

  20. 20.

    et al. The AUXIN BINDING PROTEIN 1 is required for differential auxin responses mediating root growth. PLoS ONE 4, e6648 (2009)

  21. 21.

    et al. Small-molecule agonists and antagonists of F-box protein-substrate interactions in auxin perception and signaling. Proc. Natl Acad. Sci. USA 105, 5632–5637 (2008)

  22. 22.

    & A new perspective on auxin perception. J. Plant Physiol. 167, 417–422 (2010)

  23. 23.

    et al. Mechanical stress acts via katanin to amplify differences in growth rate between adjacent cells in Arabidopsis. Cell 149, 439–451 (2012)

  24. 24.

    et al. Rho GTPase signaling activates microtubule severing to promote microtubule ordering in Arabidopsis. Curr. Biol. 23, 290–297 (2013)

  25. 25.

    et al. ABP1 and ROP6 GTPase signaling regulate clathrin-mediated endocytosis in Arabidopsis roots. Curr. Biol. 22, 1326–1332 (2012)

  26. 26.

    et al. Auxin inhibits endocytosis and promotes its own efflux from cells. Nature 435, 1251–1256 (2005)

  27. 27.

    et al. αTAT1 catalyses microtubule acetylation at clathrin-coated pits. Nature 502, 567–570 (2013)

  28. 28.

    et al. FibrilTool, an ImageJ plug-in to quantify fibrillar structures in raw microscopy images. Nature Protocols 9, 457–463 (2014)

  29. 29.

    , , , & Quantification of microtubule nucleation, growth and dynamics in wound-edge cells. J. Cell Sci. 118, 4113–4122 (2005)

  30. 30.

    et al. A novel immunological approach establishes that the auxin-binding protein, Nt-abp1, is an element involved in auxin signaling at the plasma membrane. J. Biol. Chem. 274, 28314–28320 (1999)

Download references


We thank R. Dixit for performing complementary experiments, D. W. Ehrhardt and T. Hashimoto for providing the seeds of TUB6–RFP and EB1b–GFP respectively, E. Zazimalova, J. Petrasek and M. Fendrych for discussing the manuscript and J. Leung for text optimization. This work was supported by the European Research Council (project ERC-2011-StG-20101109-PSDP, to J.F.), ANR blanc AuxiWall project (ANR-11-BSV5-0007, to C.P.-R. and L.G.) and the Agency for Innovation by Science and Technology (IWT) (to H.R.). This work benefited from the facilities and expertise of the Imagif Cell Biology platform (http://www.imagif.cnrs.fr), which is supported by the Conseil Général de l’Essonne.

Author information


  1. Institute of Science and Technology Austria (IST Austria), Am Campus 1, 3400 Klosterneuburg, Austria

    • Xu Chen
    • , Hongjiang Li
    • , Robert Hauschild
    • , Hana Rakusová
    • , Eva Benkova
    •  & Jiří Friml
  2. Department of Plant Systems Biology, Vlaams Instituut voor Biotechnologie (VIB), Ghent University, B-9052 Gent, Belgium

    • Xu Chen
    • , Hongjiang Li
    • , Anas Abuzeineh
    • , Hana Rakusová
    • , Eva Benkova
    •  & Jiří Friml
  3. Department of Plant Biotechnology and Genetics, Ghent University, B-9052 Gent, Belgium

    • Xu Chen
    • , Hongjiang Li
    • , Anas Abuzeineh
    • , Hana Rakusová
    • , Eva Benkova
    •  & Jiří Friml
  4. Institut des Sciences du Végétal, UPR2355 CNRS, Saclay Plant Sciences LabEx, 1 Avenue de la Terrasse, 91198 Gif sur Yvette, Cedex, France

    • Laurie Grandont
    • , Sébastien Paque
    •  & Catherine Perrot-Rechenmann


  1. Search for Xu Chen in:

  2. Search for Laurie Grandont in:

  3. Search for Hongjiang Li in:

  4. Search for Robert Hauschild in:

  5. Search for Sébastien Paque in:

  6. Search for Anas Abuzeineh in:

  7. Search for Hana Rakusová in:

  8. Search for Eva Benkova in:

  9. Search for Catherine Perrot-Rechenmann in:

  10. Search for Jiří Friml in:


X.C., L.G., C.P.-R. and J.F. conceived the study and designed experiments. X.C. performed experiments in roots, and L.G. performed experiments in hypocotyls. H.L., S.P. and A.A. assisted in microscopy and data generation. H.R. generated partial double mutants. R.H. did bioinformatics analysis. E.B. helped with discussion of the data. X.C., L.G., C.P.-R. and J.F. wrote the manuscript.

Competing interests

The authors declare no competing financial interests.

Corresponding authors

Correspondence to Catherine Perrot-Rechenmann or Jiří Friml.

Extended data

Supplementary information


  1. 1.

    The trajectories of EB1b in WT background following DMSO treatment

    EB1b-GFP seedlings were mounted on DMSO-contained 1/2 MS glass slides and imaged immediately for 10min. EB1b-GFP comets illustrate major transversal MT growth trajectories (90±30°). Corresponding to Fig. 3a.

  2. 2.

    The trajectories of EB1b in WT background following 100nM NAA treatment

    EB1b-GFP seedlings were mounted on 1/2 MS glass slides containing 100nM NAA and imaged immediately for 10min. EB1b-GFP moves mainly along 90±30° transversal direction in the beginning (0-60sec), while increasing EB1b tracks switch along oblique/longitudinal direction (0-60°/120-180°) after 75sec. Corresponding to Fig. 3a.

  3. 3.

    The trajectories of EB1b in ABP1 knockdown lines following DMSO treatment

    48h ethanol induced SS12S or SS12K seedlings expressing EB1b-GFP were mounted on DMSO-contained 1/2 MS glass slides and imaged immediately for 10min. High proportion of EB1b-GFP moves in oblique/longitudinal directions. Corresponding to Fig. 3b, Extended Data Fig. 4d.

  4. 4.

    The trajectories of EB1b in ABP1 knockdown lines following DMSO treatment

    48h ethanol induced SS12S or SS12K seedlings expressing EB1b-GFP were mounted on DMSO-contained 1/2 MS glass slides and imaged immediately for 10min. High proportion of EB1b-GFP moves in oblique/longitudinal directions. Corresponding to Fig. 3b, Extended Data Fig. 4d.

  5. 5.

    The trajectories of EB1b in SS12S or SS12K background following 100nM NAA treatment

    48h ethanol induced SS12S or SS12K seedlings expressing EB1b-GFP were mounted on 1/2 MS glass slides containing 100nM NAA and imaged immediately for 10min. Compared with WT, no consistent switch of EB1b trajectories to longitudinal directions but only few stochastic changes were observed. Corresponding to Fig. 3b, Extended Data Fig. 4d.

  6. 6.

    The trajectories of EB1b in SS12S or SS12K background following 100nM NAA treatment

    48h ethanol induced SS12S or SS12K seedlings expressing EB1b-GFP were mounted on 1/2 MS glass slides containing 100nM NAA and imaged immediately for 10min. Compared with WT, no consistent switch of EB1b trajectories to longitudinal directions but only few stochastic changes were observed. Corresponding to Fig. 3b, Extended Data Fig. 4d.

About this article

Publication history







By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.