Auxin is a key regulator of plant growth and development, but the causal relationship between hormone transport and root responses remains unresolved. Here we describe auxin uptake, together with early steps in signaling, in Arabidopsis root hairs. Using intracellular microelectrodes we show membrane depolarization, in response to IAA in a concentration- and pH-dependent manner. This depolarization is strongly impaired in aux1 mutants, indicating that AUX1 is the major transporter for auxin uptake in root hairs. Local intracellular auxin application triggers Ca2+ signals that propagate as long-distance waves between root cells and modulate their auxin responses. AUX1-mediated IAA transport, as well as IAA- triggered calcium signals, are blocked by treatment with the SCFTIR1/AFB - inhibitor auxinole. Further, they are strongly reduced in the tir1afb2afb3 and the cngc14 mutant. Our study reveals that the AUX1 transporter, the SCFTIR1/AFB receptor and the CNGC14 Ca2+ channel, mediate fast auxin signaling in roots.
Among all phytohormones auxin is unique for exhibiting polar transport1. The major form of auxin in higher plants, indole-3-acetic acid (IAA), is primarily synthesized in developing aerial tissues such as the shoot apex2,3. Early studies documented that auxin is mobilized, from its site of synthesis in the shoot apex, toward the root, by polar transport4,5. In the root, auxin accumulates at the tip and is then transported shoot-wards via the lateral root cap and epidermis employing a ‘reverse fountain’ transport mechanism6,7.
The polar transport of auxin, from cell to cell, is mediated by specialized influx and efflux carriers8,9,10. PIN-FORMED (PIN) and P-GLYCOPROTEIN (PGP) gene family members encode the major auxin efflux carriers, whereas the AUXIN1/LIKE-AUX1 (AUX/LAX) gene family encodes the major auxin influx carriers8,11. Depolarization of the cell membrane represents one of the earliest responses evoked by IAA, which was first recognized in oat coleoptiles and root hairs of corn12,13. However, the molecular mechanisms underlying auxin-induced early membrane responses have remained elusive.
To study early steps in auxin signaling, we selected root hairs as our single-cell model. We employed Arabidopsis mutants defective for auxin transport and responses, in combination with membrane potential-, ion flux-, and calcium measurements. We report that early auxin-induced transient depolarizations represent AUX1-mediated uptake. We show that both the membrane electrical signal and a traveling Ca2+ wave require SCFTIR1/AFB receptor-based activation of plasma membrane calcium channels.
Auxin depolarizes root hairs in a dose-dependent manner
Changes in membrane potential are often the earliest responses of plant cells to biological signals14. The membrane potential of a cell reflects the ensemble activity of ion channels, pumps, and electrogenic carriers. To monitor the membrane potential response to auxin, we used bulging root-hair cells of 3- to 5-day-old A. thaliana seedlings, grown on agarose plates. Under microscopic inspection, single-barreled microelectrodes were impaled at the tip of the root hairs (Fig. 1a), which assures that the tip of the electrode gets located in the cytosol15. With electrodes filled with 0.3 M KCl and a bath solution containing 1 mM CaCl2; 0.1 mM KCl and 5 mM MES/BTP, pH 5.5 we measured an average plasma membrane potential of −161 mV ± 1 mV. To challenge the root-hair with auxin, we positioned a pressure-operated application pipette at a distance of ~150 µm (Fig. 1a) and applied back pressure during 1 s steps. A range of apparent IAA concentrations from 0.01 to 10 µM IAA was used (see Methods). Peak depolarizations of up to 70 mV with 10 µM auxin in the pipette were recorded (Fig. 1b). The apparent IAA concentration required for a half-maximal depolarization was 0.30 µM. The same value was found for the half-maximal velocity of the initial depolarization triggered by IAA (Fig. 1c). Hence, following auxin treatment, the membrane potential rapidly depolarized in a dose-dependent manner.
AUX1 mediates proton-coupled auxin transport in root hairs
Given that the auxin influx carrier AUX1 belongs to the amino acid permease family of proton-driven transporters16,17 and IAA is derived from the amino acid tryptophan, we tested whether proton currents are accompanying IAA-triggered depolarization. Using pH sensitive extracellular electrodes that scan between two positions near the root hairs18, we monitored proton efflux in the absence of IAA, which indicates that the membrane potential is dominated by the activity of the plasma membrane proton pump (Fig. 1d). Upon IAA application, the H+-flux reversed its direction and an influx of H+ was measured. Consistent with IAA triggering H+-uptake because of the symport of IAA with protons into root hairs, we measured no, or only a minor, IAA-evoked depolarization at weakly alkaline and neutral pH, whereas the magnitude and velocity of the depolarization increased at more acidic pH values (Fig. 1e). The proton concentration required for a half-maximal IAA-dependent depolarization was 0.91 µM (pH 6; Fig. 1f). These data indicate that auxin and protons cross the cell membrane via H+/IAA symport driven by the transmembrane voltage- and proton gradient.
Next, we quantified the H+/IAA symport membrane phenomenon in response to IAA in root hairs of wild-type, versus a selection of auxin influx carrier (aux1) mutant alleles (Supplementary Fig. 1). All aux1 mutants exhibited a strongly reduced IAA-dependent depolarization (Fig. 2a). The partial loss-of function allele aux1-2 reduced IAA-induced depolarization by 50%, whereas the loss-of function allele aux1-T exhibited an up to 80% decrease (Fig. 2a). Proton flux recordings also revealed that aux1 mutants exhibited decreased H+/IAA responses, with null alleles wav5-33 and aux1-T essentially lacking an IAA-dependent H+-influx (Fig. 2a). Comparing H+/IAA responses between wild-type and wav5-33 revealed that AUX1 contributes more than 80% to the total root-hair IAA uptake capacity (Fig. 2b; Supplementary Fig. 2a). AUX1 has a strong preference for naturally occurring IAA as substrate, based on a comparison of the root-hair depolarization of wild-type and wav5-33 mutants in response to the auxin analogs IAA, 5F-IAA, 1-NAA, 2-NAA, 2,4-D, and BA (Supplementary Fig. 2b). Hence, AUX1 is responsible for the proton-coupled IAA uptake into root hairs.
AUX1-driven root-hair depolarization is sensitive to [P]ext
AUX1 is expressed in root epidermal atrichoblasts19 and, to a lesser extent, in root hairs (trichoblasts)20,21. On the basis of membrane potential recordings we observed higher auxin transport activity in atrichoblasts compared to root hairs (Supplementary Fig. 3a), suggesting that higher transcript numbers translate into more active AUX1 transport protein. AUX1 expression and abundance is further induced during germination, as well as phosphate (P) starvation22. We therefore assumed that low phosphate supply would feedback on auxin transport capacity and performed membrane potential recordings on root hairs of Arabidopsis seedlings grown on phosphate concentrations ranging from 300 µM (standard MS medium) down to 0.3 µM (equivalent to low soil P levels). Resting membrane potentials were similar for both wild-type (163.1 mV ± 1.5 mV) and wav5-33 (163.8 mV ± 1.8) mutants irrespective of phosphate supply (Supplementary Fig. 3b). However, in wild-type root hairs fast membrane potential depolarization in response to a 1 s IAA pulse (0.3 µM) increased with decreasing phosphate supply and highest depolarization rates were obtained with seedlings grown in the presence of 0.3 µM phosphate (Fig. 2c, d). In contrast, the phosphate effect on membrane depolarization was absent in wav5-33 mutants (Fig. 2c, d). The observation that the IAA-induced instantaneous proton fluxes were higher at low phosphate conditions, suggests that this reflects differences in AUX1 activity (Supplementary Fig. 3c). Hence, our measurements reveal that AUX1 provides a key component of the auxin-dependent root-hair response to low external P.
Auxin triggers elevation of cytosolic calcium in root hairs
When root hairs were exposed to IAA pulses of 1 s duration, membrane depolarizations were evoked at pipette concentrations as low as 10 nM. Increased auxin concentrations enhanced the velocity, as well as the amplitude of the membrane potential changes (Fig. 1b, c). At IAA doses above 50 nM, the membrane potential changes were transient in nature. Such transients consist of five phases: (i) an initial delay that precedes a (ii) period of acceleration followed by a phase in which (iii) the peak value is reached and (iv) a repolarization phase leading to (v) a steady state. Although the hormone was applied at the external root-hair side for only 1 s, the depolarization transient lasted for minutes and the plateau phase (v) even longer. This behavior points to an auxin response more complex than just AUX1-mediated H+/IAA co-transport. Transient depolarizations followed by a prolonged depolarized phase have also been observed in response to the phytohormone ABA as well as microbe- or danger associated molecular pattern’s (MAMPs, DAMPs)23,24,25,26. With the latter stimuli, depolarization increases cytosolic calcium concentration, and is likely to activate plasma membrane anion channels leading to an enhanced membrane depolarization27,28.
To test whether IAA triggers a calcium influx, we performed measurements with scanning Ca2+-selective extracellular electrodes. IAA provoked a transient influx of Ca2+, with similar kinetics as observed for the simultaneously measured H+-fluxes (Fig. 3a). However, in contrast to H+-fluxes, inward Ca2+-fluxes were transient in nature. To test if auxin-induced transient calcium influx leads to elevation of cytosolic Ca2+ levels ([Ca2+]cyt), we studied Arabidopsis seedlings expressing the genetically encoded calcium reporters R-GECO1 and YC3.629,30, of which R-GECO1 shows high Ca2+-dependent changes in fluorescence intensity30. A 1 s pulse of 10 µM IAA caused a transient rise of [Ca2+]cyt that matched the Ca2+-influx measured with ion-selective electrodes (Fig. 3b–d). Simultaneous membrane potential recordings and Ca2+ imaging revealed that the cytosolic calcium level returned to its resting level after 4–5 min, whereas membrane depolarization continued beyond this period (Fig. 3c). This suggests that the rise in [Ca2+]cyt is not directly linked to the prolonged IAA-dependent membrane depolarization. This is further supported by different affinities for IAA transport and signaling (Supplementary Fig. 4a, b). The kinetics of both processes however matched each other during the initial phase of the IAA response (Supplementary Fig. 4c) indicating that auxin transport activity and cytosolic calcium changes are tightly coupled. This finding is further supported by the observation that calcium influx is impaired in aux1 mutants (Supplementary Fig. 4d) and the finding that the auxin analogs 2-NAA and BA failed to efficiently trigger both responses (Fig. 3e).
Calcium signals depend on auxin perception and CNGC14
When root hairs were pretreated with the potent SCFTIR1/AFB auxin receptor inhibitor auxinole31, the IAA-triggered depolarization rate was reduced to 20% of control cells. Moreover, auxinole completely repressed the transient increase in [Ca2+]cyt (Fig. 3f), despite of the fact that auxinole shifts the root-hair resting membrane potential towards depolarized values (Supplementary Fig. 4e, f). This indicated that the IAA activated SCFTIR1/AF receptor complex is required to trigger the calcium transient and feeds back on AUX1 transport activity. To provide genetic evidence for the role of SCFTIR1/AFB we studied AUX1 activity and Ca2+ signaling in tir1, as well as the tir1afb2afb3 triple mutant (Fig. 4; Supplementary Fig. 4g, h). Both the IAA-induced membrane potential depolarization, as well as calcium influx, were essentially unaffected in the tir1 mutant, but were reduced to levels observed for auxinole treated root hairs in the triple mutant tir1afb2afb3 (Fig. 4a, b). This suggests that the auxin receptor complex, including TIR1, AFB2, and/or AFB3, control AUX1 transport activity, as well as fast auxin signaling. Well in agreement with our hypothesis that functional auxin receptors are tightly linked to auxin transport and Ca2+ signaling, the IAA-triggered H+- and Ca2+-influx were severely impaired in the tir1afb2afb3 triple mutant when compared to wild-type, or tir1 (Fig. 4b; Supplementary Fig. 4g, h). The transcript abundance for AUX1, as well as for the cyclic nucleotide-gated channel CNGC14 (see below), was unaffected in the receptor mutants, or by auxinole treatment (Supplementary Fig. 4i). This indicates that the SCFTIR1/AFB receptor modulates this response through a posttranslational mechanism.
To test whether H+/IAA import is directly coupled to calcium signaling (Fig. 3), we injected IAA into the root-hair cells directly. Doubled-barreled microelectrodes were placed in the root-hair cytoplasm, the membrane potential was monitored via the first barrel and on demand IAA could be iontophoretically injected via the second microcapillary. Upon auxin injection the R-GECO signal reported an IAA-dependent Ca2+ transient (Fig. 5a, b). The physiologically inactive 2-NAA failed to elicit the Ca2+ response and in the presence of auxinole the Ca2+ response was inhibited (Fig. 5c, d). The rise in cytoplasmic calcium evoked by IAA injected into the root-hair, however, was not superimposed by a fast membrane potential depolarization (Supplementary Fig. 5a). This shows that AUX1-driven H+ import induces the fast depolarization, whereas cytosolic IAA only elicits the calcium signal, possibly via the SCFTIR1/AFB complex. Current injection of IAA caused a hyperpolarization of the plasma membrane of approximately −10 mV, followed by a slow depolarization of about 18 mV (Fig. 5c; Supplementary Fig. 5a). We hypothesized that the depolarization relates to calcium influx from the apoplast (Fig. 4a), possibly via the previously identified cyclic nucleotide-gated channel CNGC1428. Ion flux recordings and cytosolic auxin injection revealed that in line with cytoplasmic calcium imaging data28, the IAA-triggered Ca2+-influx was completely absent in the cngc14 mutant (Fig. 5e; Supplementary Fig. 5b, c). Surprisingly, the IAA-induced membrane potential depolarization as well as proton influx were absent in the cngc14 mutant, too (Fig. 5e, f)28. Compared to wild-type, however, AUX1 as well as TIR1, AFB2, or AFB3 expression appeared unaltered in cngc14 (Supplementary Fig. 5d). This indicates that CNGC14 regulates AUX1 activity at a post-transcriptional level. Further, this demonstrates that IAA is perceived by the SCFTIR1/AFB auxin receptor complex and evokes a CNGC14-dependent Ca2+ signal. Transcriptional analyses revealed that auxin-inducible expression of IAA19 was strongly impaired in the tir1afb2afb3 triple mutant, as well as in cngc14 when compared to wild-type (Supplementary Fig. 5e). This observation is in line with an important role for CNGC14 in regulating AUX1-mediated auxin import.
Auxin signaling in root cells is modulated by calcium waves
We observed that the elevation of [Ca2+]cyt in root hairs not only occurred in those cells injected with IAA, but also in those that are symplastically connected to it (Fig. 5a, b). The calcium signal propagated as a transversal wave from the IAA stimulated root-hair towards the center of the root and beyond that to epidermal cells at the opposite side of the root. This Ca2+ wave traveled at a speed of 0.5 cm h−1 and traversed the entire root profile within two minutes (Fig. 5b). The Ca2+ signal underwent amplification while spreading through the root, as secondary Ca2+ signals (ROI2) were about twofold higher when compared to those at the site of origin (ROI1, Fig. 5b, d). As seen for the local depolarization response, the inactive auxin 2-NAA failed to elicit Ca2+ waves, whereas the auxin response inhibitor auxinole blocked auxin-induced Ca2+ signaling (Fig. 5d).
Upon reaching the root stele, the Ca2+ wave propagated tipwards with a constant speed of 1.56 cm h−1 (Fig. 6a, b; Supplementary Fig. 6a). Auxin transport also can occur at these velocities32. Because of the constant velocity of propagation however, the Ca2+ signals most likely resemble Ca2+ waves such as those that have been observed in a variety of eukaryotes33,34,35. Such a Ca2+ wave may modulate auxin signaling in cells, at distance from the IAA injected one. To challenge this hypothesis we took advantage of the Aux/IAA-based auxin-signaling sensor DII-VENUS36. Upon injection of IAA into a single root-hair cell, the DII-VENUS signal in cells of the elongation zone (ROI in Fig. 6c) started to decay within minutes, to reach a new steady state within 25 min (Fig. 6d). We investigated the possible causal relationship between calcium and auxin signaling by repeating these experiments in the presence of the Ca2+ channel blocker lanthanum. A concentration of 128 µM La3+ was chosen, as it entirely repressed a second IAA-dependent depolarization that occurs in root hairs exposed to two consecutive hormone pulses in the absence of IAA (Supplementary Fig. 6b). When La3+ was applied at this concentration, it inhibited both IAA-triggered Ca2+ signaling and DII-VENUS degradation at the root tip (Supplementary Fig. 6c; Fig. 6d). Hence, we conclude that AUX1-regulated local auxin signaling in root hairs triggers a Ca2+ wave that controls auxin action, both at short and long distances.
To study the very earliest auxin responses in Arabidopsis root hairs, we employed electrophysiological techniques. This revealed that IAA depolarizes the plasma membrane in an AUX1-dependent manner. On the basis of a stoichiometry of at least two H+ per IAA− and a chemiosmotic potential ranging from 180 to 300 mV, our data indicate that AUX1-mediated transport could theoretically accumulate auxin to 105 fold higher concentrations inside root hairs, as compared to the cell walls. The in planta affinity of AUX1-dependent IAA transport determined in this study (0.3 µM) is higher compared to those obtained in heterologous expression systems37,38, but in close agreement with values found for auxin-dependent degradation of Aux/IAAs mediated by TIR139. Consistent with this major role, mutants lacking AUX1 are profoundly impaired in root development and nutrient controlled root adaptive responses40,41,42. The observed difference in AUX1-dependent membrane depolarization under PO4 limiting conditions, suggest AUX1 is important to auxin-dependent adaptive root responses. Phosphate-starved roots were also characterized by increased auxin levels and auxin responses43,44. This is related to the enhanced expression of the auxin receptor TIR144 and reveals coupling of auxin transport and perception, to fine-tune root auxin responses.
Auxin-regulated developmental processes are usually described to be dependent on transcriptional changes that occur over a timescale of several minutes to hours45. Our ion flux studies revealed that application of local auxin pulses virtually immediately result in inwardly directed proton fluxes in accordance with experiments employing genetically encoded pH sensors46. These proton fluxes are largely reduced in aux1 mutants, consistent with the auxin-induced rise in extracellular pH being dominated by AUX1-mediated 2H+/IAA− transport activity. Our studies reveal that IAA triggers cytoplasmic Ca2+ transients, which require SCFTIR1/AFB-based auxin signaling. Auxinole suppresses the auxin-induced Ca2+ response almost completely and functional studies employing the tir1afb2afb3 triple mutant suggest a prominent role for SCFTIR1/AFB in fast auxin signaling (Fig. 4). In response to 1 s IAA pulses, the membrane potential depolarizes at the same velocity as the transient Ca2+ elevation is initiated (Supplementary Fig. 4c). Again, this can only be explained by a very short auxin-signaling pathway, in which IAA entry via AUX1 binds to SCFTIR1/AFB, which in turn regulates opening of plasma membrane calcium channels. We found that loss-of-function mutants of the cyclic nucleotide-gated channel CNGC14 lack an IAA-triggered Ca2+, as well as an AUX1-mediated H+-influx. In line with previous studies28,47, our findings support a model in which CNGC14 acts as the bona fide auxin-activated Ca2+-permeable channel at the plasma membrane (Fig. 7). This fast, membrane-delimited auxin signaling mechanism is unlikely to require downstream components, such as nuclear localized AUX/IAA proteins that regulate gene transcription48. This situation is reminiscent of fast and slow ABA signaling events49. In analogy to fast ABA signaling, one would expect that the auxin-activated SCFTIR1/AFB complex regulates a protein phosphatase, or kinase, to open the CNGC14 channel. The impact of CNGC14 on [Ca2+]cyt probably is crucial for AUX1 activity, since AUX1-mediated IAA transport is inhibited in the presence of the Ca2+ channel blocker La3+, as well as in cngc14 (Supplementary Fig. 6, 7). This suggests a negative, Ca2+-dependent, feedback control circuit within a functional signaling pathway consisting of AUX1, SCFTIR1/AFB, and CNGC14.
Maintenance of a tip-focused cytoplasmic calcium gradient is essential for root-hair growth50. Consistent with such a model, the R-GECO reporter revealed an auxin-induced Ca2+ signal at the root-hair tip, but this was transient (Fig. 6b). Hence, a prolonged and elevated auxin source (for example, under low external PO4 conditions) may be necessary to maintain the tip-focused Ca2+ gradient and promote hair elongation. During root-hair growth, the tip-focused Ca2+ gradient is known to coordinate secretion, endocytosis and actin dynamics51. In contrast, the role of the observed auxin-inducible intercellular calcium wave is less clear. One possibility is that the calcium wave modulates auxin transport45. Consistent with such a model, we observed changes in the abundance of the auxin sensor DII-VENUS (Fig. 6). Blocking Ca2+ signaling via La3+ inhibits DII-VENUS degradation and thus distal auxin signaling (Fig. 6d). This calcium wave may therefore help to modulate auxin responses in surrounding root epidermal cells to coordinate hair elongation and improve foraging for PO4. However, further studies are required to test these intriguing mechanistic possibilities. Nevertheless, the spatial and temporal resolution of our electrophysiological-based auxin transport measurements, combined with imaging dynamic changes in auxin and Ca2+ abundance within and between root-hair cells, provides insight into the auxin-dependent fast signaling events taking place in epidermal cells, at the interface between the root system and nutrient sources in the soil.
Plant growth conditions
Seeds of Arabidopsis thaliana were sterilized for 5 min by application of 6% NaOCl, supplemented with 0.05% Triton-X 100. Three to six washing steps with deionized water removed the sterilizing solution. Single seeds were placed in a row on the surface of small petri-dishes (Ø 35 mm) filled with 1 ml of plant growth medium (0.12% Murashige and Skoog basal salt mixture incl. MES, Duchefa; 0.5% sucrose; 1% agarose, pH 5.8 with TRIS). The petri-dishes were placed vertically in a growth chamber (KBWF 720, Binder) with controlled environmental conditions (12 h day vs. 12 h night; 21 °C at day vs. 16 °C at night; 120 µmol photons m−2 s−1) 3 to 5 days before experiments. For phosphate starvation experiments, ¼-strength MS-media (5.2 mM NH4NO3; 4.7 mM KNO3; 0.6 mM CaCl2; 0.2 mM MgSO4; 25.1 µM H3BO3; 25 µM Na2EDTA; 25 µM FeSO4; 19 µM MnSO4; 13.3 µM ZnSO4; 1.7 µM KI; 386 nM Na2MoO4; 30 nM CoCl2; 25 nM CuSO4; 312 µM–0 µM KCl ; 312 µM–0.3 µM KH2PO4; 0.5 % sucrose; 2.35 mM MES; pH 5.8 with TRIS) were manually prepared and plants were grown for 4 days on those media.
Membrane potential recordings and auxin application
Sterile grown seedlings were exposed to standard bath solution (0.1 mM KCl, 1 mM CaCl2, 5 mM MES/BTP pH 5.5; in the case of La3+-containing bath solutions ionic strength was kept constant and the sum of La3+ and Ca2+ concentrations was 1 mM). The response of the membrane potential of root epidermal cells to apoplastic and cytosolic auxin application was measured by sharp single- and double-barreled microelectrodes impaled through the tip of bulging root-hair cells. Fabrication of single- and double-barreled microelectrodes from borosilicate glass capillaries and membrane potential recordings were as described by Wang et al.15 Back-pressure-operated application pipettes for apoplastic auxin application were made of single-barreled microelectrodes, of which the tips were broken off, to obtain an opening of 20–40 µm. The application pipettes were filled with bath solution containing the required concentration of auxin and back-pressure pulses of 1 s were applied through a Picospritzer II microinjection dispense system (General Valve) operating at 30 psi. A range of IAA concentrations from 0.01 to 10 µM IAA was used, but the hormone concentration at the root cell plasma membrane will have been lower. The application pipette was operated at a distance of 150 µm of the root and experiments with the fluorescent dye Lucifer yellow CH (Fluka/Sigma) revealed an average fourfold dilution solutes in the pipette over the distance of 150 µm. The tip of single and double-barreled microelectrodes, used for iontophoretic injection of auxin into the cytosol, was filled with bath solution containing 6.66 mM 3-IAA (Sigma-Aldrich), or 3.33 mM 2-NAA (Merck), 0.5 mM Lucifer yellow and 0.83 mM HEPES/TRIS (pH 7), whereas the remaining of the electrode was filled with 300 mM KCl. For cytosolic injection a current of −1 nA was applied.
Ion flux measurements
Ion fluxes across the plasma membrane of root epidermal cells were monitored using non-invasive ion-selective scanning microelectrodes. To avoid rupture of the root-hairs due to contact with the vibrating electrodes, the tip of the electrode was placed on the root body at the early differentiation zone between root-hairs. Prior to experiments, Arabidopsis seedlings were accustomed to the bath solution (100 µM KCl, 100 µM CaCl2, 100 µM MES/BTP pH 5.5) for at least 20 min. After stable fluxes were recorded for at least 3 min, 3-IAA was added to the bath solution to a final concentration of 10 µM. Measurements in which plants were treated with a solvent control (EtOH) showed no flux alterations (Supplementary Fig. 8). Electrode preparation, calibration and experimental set-up are described in the Supplementary Methods.
Calcium and DII-Venus imaging
Life cell imaging was performed on a Zeiss Axiokop 2FS, equipped with a CARV2 confocal imaging unit (Crest Optics). Image acquisition was carried out without the spinning disc in the light path, using a charge multiplying CCD camera (QuantEM 512SC, Photometrics), controlled by Visiview software (Visitron). The following single bandpass excitation filters were used; 430 nm (ET 430/24 nm, Chroma technology), 500 nm (filter 500/20, Semrock) and 562 nm (562/40 nm, Semrock) for LY, DII-Venus and R-GECO1, respectively. Light was reflected on the sample by a 490 nm dicroic mirror (Zeiss), a 444/521/606 dichroic beamsplitter (Brightline triple-edge beamsplitter, Semrock) and a 590 nm dichroic mirror (Zeiss). The emission signals were filtered with single bandpass filters; 520/30 nm (Semrock), 520 nm (Brightline HC 520/35, Semrock) and 628 nm (628/40 nm, Semrock). The excitation light was focused on the sample through an Achroplan ×40/0.80w objective (Zeiss) or an Achrostigmat ×10/0.25 objective (Zeiss). For analysis of imaging data the ImageJ Software (imagej.nih.gov/ij/) was used.
Quantitative real-time PCR
RNA was isolated from whole seedlings after electrophysiological experiments. To obtain an adequate amount of material for extraction approx. 5 to 10 seedlings were pooled into one sample. For RNA extraction the NucleoSpin® RNA Plant Kit (Macherey-Nagel) was used and transcript levels were ultimately analyzed after cDNA synthesis through quantitative real-time PCR using the Absolute QPCR SYBR green capillary mix (Thermo Scientific) and a Realplex Mastercycler (Eppendorf). Expression levels of individual genes were calibrated to standard curves for each transcript and subsequently normalized to 10,000 transcripts of actin (AtACT2/8). The following primers were used: AtACT2/8: AtACT2/8 forward (fw): 5′-GGTGATGGTGTGTCT-3′, AtACT2/8 reverse (rev): 5′-ACTGAGCACAATGTTAC-3′; AtAUX1: AtAUX1fw: 5′-GGATGGGCTAGTGTAAC-3′, AtAUX1rev: 5′-TGACTCGATCTCTCAAAG-3′; AtCNGC14: AtCNGC14fw: 5′-CAGCCAAGCTAAGACTCT-3′, AtCNGC14rev: 5′-GTTGAAGCCTTTGCTTTA-3′; AtTIR1: AtTIR1fw: 5′-CTTCTTGTTCCGTGAGTT-3′, AtTIR1rev: 5′-ATTCAAATTATTGGCGAC-3′; AtAFB2: AtAFB2fw: 5′-ATGATAATAACCGGATGGA-3′, AtAFB2rev: 5′-TCGGGAAAGACACACTAAC-3′; AtAFB3: AtAFB3fw: 5′-GACGTGGGTAGGTACGAAA-3′, AtAFB3rev: 5′-AAAACACATGAAGGTGCA A-3′; AtIAA19: AtIAA19fw: 5′-TTGATAAGCTCTTCGGTTT-3′, AtIAA19rev: 5′-TGCAAAATAATATACTATAAC-3′
To test significances two-tailed Student’s t-tests were applied. All data points marked as significantly different from wild-type or control experiments were so with p < 0.05.
The authors declare that all data supporting the findings of this study are available within the article and its Supplementary Information files, or from the corresponding author on request.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This work was supported by the German Research Foundation (DFG, FOR964 to R.H. and D.B.). We thank Dr. Ken-Ichiro Hayashi for providing auxinole and Pamela Korte for qPCR assistance.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.