Cytokinin regulates root meristem activity via modulation of the polar auxin transport

Abstract

Plant development is governed by signaling molecules called phytohormones. Typically, in certain developmental processes more than 1 hormone is implicated and, thus, coordination of their overlapping activities is crucial for correct plant development. However, molecular mechanisms underlying the hormonal crosstalk are only poorly understood. Multiple hormones including cytokinin and auxin have been implicated in the regulation of root development. Here we dissect the roles of cytokinin in modulating growth of the primary root. We show that cytokinin effect on root elongation occurs through ethylene signaling whereas cytokinin effect on the root meristem size involves ethylene-independent modulation of transport-dependent asymmetric auxin distribution. Exogenous or endogenous modification of cytokinin levels and cytokinin signaling lead to specific changes in transcription of several auxin efflux carrier genes from the PIN family having a direct impact on auxin efflux from cultured cells and on auxin distribution in the root apex. We propose a novel model for cytokinin action in regulating root growth: Cytokinin influences cell-to-cell auxin transport by modification of expression of several auxin transport components and thus modulates auxin distribution important for regulation of activity and size of the root meristem.

Plant hormones play a crucial role in regulating plant development and the flexible shaping of the plant architecture in response to variable environmental conditions. The final developmental and physiological output of the hormonal signaling in plants is the typical result of combined actions of several hormonal pathways. However, our knowledge of the mechanisms involved in the hormonal crosstalk is still poor.

In the regulation of root development, several hormonal pathways are involved, with auxin and cytokinin being the principal players. The whole process of root organogenesis, starting with the initiation of the root pole in embryos (1), positioning and formation of stem cell niche (2, 3), maintenance of mitotic activity in proximal meristem (4–6), and rapid elongation and differentiation of cells leaving the root meristem (7) has been demonstrated to be controlled by auxin. In this context, the differential auxin distribution between cells is crucial (3, 8, 9). The auxin gradients or local auxin maxima can be generated by auxin metabolic reactions, mainly by local auxin biosynthesis (6, 10, 11) and intercellular auxin transport dependent on the coordinated action of influx carriers of the AUX/LAX family (12), PIN efflux carriers (13, 14), and members of the multidrug-resistant/P-glycoprotein (MDR/PGP) subfamily B of ATP-binding cassette (ABCB) proteins (15, 16). Accordingly, interference with the polar auxin transport disrupts the auxin distribution and results in dramatic patterning defects in the root meristem (2, 3, 9).

Besides auxin, cytokinin (CK) is also involved in root organogenesis. Increase in CK levels by exogenous application or overexpression of the bacterial ISOPENTENYLTRANSFERASE (IPT) gene (17, 18) inhibits the root growth and reduces the meristem size. Accordingly, decreased endogenous CK levels via overexpression of the CYTOKININ OXIDASE/DEHYDROGENASE (CKX) family genes have an opposite effect, i.e., enhanced meristem and root growth (19).

CK seems to play an important role already in the early events of root specification, because lack of the ARR7, and ARR15 components of the CK signaling pathway causes defects in the establishment of the root stem cell niche during embryogenesis (20). Postembryonically in the root meristem, CK does not seem to interfere with specification of the quiescent center (QC) and stem cell function, or with the overall division rate, but affects mainly the meristematic cell differentiation rate, resulting in shortening of the meristematic zone (4, 5). In addition, CK regulates elongation of cells leaving the root meristem (4).

Although the interaction between auxin and CKs in control of organogenesis has been known for years (the first information related to this phenomenon appeared decades ago) (21), the molecular mechanisms underlying the mutual coordination of the auxin and cytokinin action and the possible crosstalk of their pathways in regulating root growth are so far not known. We reveal a unique mechanism of auxin–cytokinin interaction and show that CK regulates the cell-to-cell auxin transport by modulating transcription of several PIN auxin efflux carriers. We propose a model for regulation of the auxin–cytokinin balance that is critical for root organogenesis. By modulating the auxin transport CK might control the auxin levels in root meristem cells and, thus, the ratio between auxin and CK.

Results and Discussion

The CK Effect on the Root Meristem Does Not Interfere with Ethylene-Mediated Processes.

Root growth depends on the production of new cells, their differentiation, and elongation. New cells are produced in the mitotically active meristem zone, whereas their differentiation and elongation occur in the more proximal part of the root tip. The plant hormone CK regulates the root meristem activity. Increase of CK levels either by exogenous application of CK or overexpression of ISOPENTENYLTRANSFERASE (IPT), a gene involved in CK biosynthesis, reduces the size of root meristems and overall root growth (5) (supporting information (SI) Fig. S1C). On the contrary, plants overexpressing CYTOKININ OXIDASE/DEHYDROGENASE (CKX) causing enhanced degradation of CK have longer root meristems (19) (Fig. S1C).

Because CK stimulates ethylene biosynthesis (22) and ethylene itself strongly affects root growth (23, 24), we first examined which CK effects on the root growth are mediated by ethylene. To address this issue, either CK-induced ethylene production was eliminated with the ethylene biosynthesis inhibitor 2-aminoethoxyvinylglycin (AVG) (25) or the ethylene response was prevented genetically by mutations in the ethylene signaling pathway, thus exhibiting typical ethylene-insensitive root growth (26). Significantly, reduction of ethylene synthesis by AVG, and interference with the ethylene signaling in etr1–3 and ein2 mutants led to CK-insensitive cell elongation and overall root growth (data not shown and Fig. 1 B and D) but did not interfere with the CK effect on the root meristem (Fig. 1 A and C). To analyze the CK effect on the cell size and cell division activity of the root meristem more in detail, we used the CycB1;1::GUS_DB reporter, marking cells in the G2 stage of the cell cycle (27). CK alone and at simultaneously inhibited ethylene biosynthesis dramatically reduced the zone of the CycB1;1::GUS_DB expression, suggesting that the CK treatment might interfere with the balance between mitotic activity and differentiation of cells in the proximal meristematic zone (Fig. 2A, Fig. S2 B and D) (5), and further confirming that the CK effect on the root meristem does not depend on ethylene.

Fig. 1.

Ethylene-independent root meristem size modulation by CK. (A) Reduction of ethylene biosynthesis by AVG without interference with CK effect on the root meristem size (no difference between CK-treated and CK + AVG-treated root meristems; Student's t test, P > 0.05). (B) CK-insensitive root growth at reduced ethylene biosynthesis conditions (*statistically significant difference in root length between CK and CK + AVG-treated seedlings; Student's t test P < 0.05). (C) Reduced root meristem in the ethylene signaling mutants etr1–3 and ein2 after CK treatment (*statistically significant difference in the root meristem size between CK-treated and -nontreated seedlings; Student's t test, P < 0.05). (D) CK-resistant root growth of etr1–3 and ein2. (E) CK-induced ectopic expression of DR5::GFP reporter in outer layers of the root meristem (lateral root cap and epidermis; arrowheads). Reduction of ethylene biosynthesis by AVG diminishes the DR5 ectopic expression. CK-reduced DR5 expression in QC (asterisks). (F) Quantification of DR5::GFP expression by image analysis in QC of roots treated with CK and CK + AVG; expression of DR5::GFP is significantly reduced after CK treatment under conditions of reduced ethylene biosynthesis (*statistically significant difference between AVG and CK + AVG-treated roots, Student's t test, P < 0.05). Six-day-old seedlings grown on media containing 100 nM BA, 200 nM AVG, (if not marked differently). RM, root meristem; RL, root length; MS, Murashige and Skoog medium only; CK represented by N⁶-benzyladenine (BA); col, control Columbia seedlings. Error bars represent standard deviation (SD).

Fig. 2.

CK regulation of root growth in a manner similar to 2,4-D. (A) Increased zone of CycB1;1::GUS_DB reporter expression by lower (200 nM) NAA concentration. Treatment with CK (100 nM BA) or 2,4-D (100 nM) reduced expression of CycB1;1::GUS_DB reporter. (B) Root growth and root meristem size on NAA concentration gradients. Low NAA concentration with strong inhibitory effect on root growth still has a stimulatory effect on the root meristem size (meristems at 200 nM statistically significant, Student's t test, P < 0.05). (C) Gradual reduction of size of the root meristem and inhibition of root growth on 2,4-D concentration gradient (meristems at 40 nM and higher concentration statistically significant, Student's t test, P < 0.05). (D) Gradual reduction of the root meristem size and root growth by increasing CK (BA) concentrations (meristems at 1 nM and higher concentration statistically significant, Student's t test, P < 0.05). Six-day-old seedlings grown on media supplemented with hormones. RM, root meristem; RL, root length. Error bars represent SD.

Spatial patterns of the auxin response based on auxin gradients are important factors in the regulating of a large number of plant developmental processes, including dividing root cells (3, 9). Using the auxin response reporters DR5::GUS and DR5::GFP (28), we observed that the CK treatment induces ectopic DR5 activity in the outer layers of the root meristem and the elongation zone (Fig. 1E, data not shown). A similar change in the pattern of the DR5 reporter expression has been demonstrated previously to be caused by ethylene (23, 24, 29). Indeed, reduced ethylene biosynthesis by AVG completely eliminated this ectopic signal, indicating that CK upregulates DR5::GFP expression in the elongation zone through ethylene (Fig. 1E). Remarkably, besides induction of the DR5 activity in the epidermis of the root tip, we observed that CK attenuated the signal in the QC and columella cells and these effects became more obvious when the ethylene biosynthesis was inhibited by AVG (Fig. 1 E and F).

In summary, our results dissected the ethylene-dependent and -independent roles of CK on root growth: CK regulates the root meristem size independently of ethylene, while its effect on the cell elongation and overall root growth is mediated through ethylene (23, 24, 29). Ethylene-independent reduction of the auxin response in the QC and columella cells suggests a direct interaction between the CK and the auxin pathways, independent of ethylene.

CK and Nontransportable Auxin Show Similar Effects on the Root Meristem.

The differential auxin distribution in the root meristem requires an active auxin transport (9). Furthermore, some mutants defective in the activity or polar localization of PIN auxin efflux carriers (2, 30) are dramatically reduced in the root meristem size, resembling the CK effect. To address whether the CK effect on the auxin distribution and root meristem activity is related to the auxin transport, we compared the effects of 2 different auxins—1-naphthaleneacetic acid (NAA), well transportable by the polar transport system, and 2,4-dichlorophenoxyacetic acid (2,4-D), which tends to accumulate in cells because of its low affinity to the auxin efflux machinery (31). Our results revealed dramatic differences between these 2 auxins. NAA stimulated the mitotic activity in the proximal root meristem, as visualized by the CycB1;1::GUS_DB reporter (Fig. 2A; Fig. S2A), resulting in an enlarged root meristem at lower (200 nM) concentrations, but reduced root growth already at 50 nM concentration (Fig. 2B). In contrast to NAA, 2,4-D treatment led to continuous reduction in the root meristem size and the root length at concentrations higher than 20 nM and reached the same degree of reduction of root growth at the concentration of 200 nM as did NAA at 800 nM (Fig. 2 A and C; Fig. S2C). When compared to auxins, CK mimics in many aspects the 2,4-D mode of action. It reduced the root meristem size, and it dramatically decreased root elongation already at 30 nM (Fig. 2 A and D; Fig. S2B). Moreover, low NAA concentration did not overcome CK inhibitory effect on root meristem size (Fig. S2E) suggesting that CK effect on root meristem size is not mediated exclusively by controlling overall auxin levels. Our results demonstrate that auxins, and CK, modulate the root meristem size, but that they differ in concentration range at which they can act either in a stimulatory or an inhibitory mode. Importantly, the CK effect on the root meristem size resembles that of an inefficiently transported 2,4-D, hinting that the auxin transport can be a downstream target of the CK action in regulating this process.

CK Reduces Auxin Efflux in Tobacco BY-2 Suspension Cells.

We tested whether CK has a direct effect on the auxin efflux—the rate-limiting process in the intercellular auxin transport. The established assay for auxin efflux quantification from cells relies on measuring the accumulation of radioactively labeled NAA ([³H]NAA) in cultured tobacco BY-2 cells (14). In contrast to Arabidopsis cell cultures, in BY-2 culture each cell is in direct contact with the medium, so that it can render more precise quantitative data. NAA is well conveyed into the tobacco cells by passive diffusion and it is well transported out by the efflux carriers; therefore, NAA accumulation inside cells is a good measure for the auxin efflux activity (14, 31).

We tested the CK effect on the auxin efflux with BY-2 cells expressing under the endogenous promoter the Arabidopsis PIN1 auxin efflux carrier tagged with RFP. In this assay, CK does not affect short term the auxin efflux (within 30 min, data not shown). However, longer CK pretreatments (24 h) led to an increase of the [³H]NAA accumulation that indicates a decrease in NAA efflux activity (Fig. 3B). Treatment with the auxin efflux inhibitor 1-naphthylphthalamic acid (NPA) has very similar, but immediate, effect on the NAA accumulation in PIN1:GFP expressing cell culture (ref. 16; data not shown), further confirming that CK inhibits the auxin efflux, possibly by an indirect mechanism. We tested whether CK has an effect on PIN expression in BY-2 cells. Analysis of PIN1:RFP signal revealed that the CK treatment leads to a decreased PIN1:RFP expression (Fig. 3A; Fig. S1D). These data suggest that CK might modulate the auxin efflux via regulation of the PIN expression.

Fig. 3.

Interference of CK with expression of auxin efflux carriers. (A) Reduction of PIN1:RFP expression by CK in tobacco BY-2 cells. (B) Increased auxin accumulation in the BY-2 cells after pretreatment with CK (5 μM BA, 24 h). (C) Quantitative RT-PCR expression analysis of PIN genes. Six hours after CK treatment, expression of PIN genes is modulated. ARR5 used as positive control induced by CK (44). (D) Expression of PIN1:GFP is reduced by CK treatment. (E) Downregulation of PIN2:GFP expression by CK effect. (F) Downregulation of PIN3:GFP expression after long-term (24 h) CK treatment. (G) Stimulation of PIN7:GFP expression by CK treatment. Six-day-old seedlings incubated in control or with 5 μM BA supplemented media. MS, Murashige and Skoog medium only; CK represented by BA. Error bars represent SD.

CK Modulates Expression of the Auxin Efflux Carriers of the PIN Family.

The next question was whether endogenous or exogenous manipulation of CK levels in planta affects the expression of the auxin efflux carriers from the PIN family. To distinguish whether the observed changes in expression are CK-specific or related to CK-induced ethylene synthesis, all experiments were carried out in parallel in the presence of AVG as inhibitor and 1-aminocyclopropane-1-carboxylate (ACC) as precursor of the ethylene biosynthesis (25), respectively. Analyses using transcriptional and translational reporters, and real-time Q-RT PCR revealed that CK modulates the expression of several PIN genes in a manner specific for each particular auxin efflux carrier.

CK negatively regulates PIN1 expression. CK treatment strongly reduced the PIN1:GFP or PIN1 (visualized by antibody staining) signal in root tips (Figs. 3D and 4A; Fig. S3A). Similarly, endogenously increased CK levels, by dexamethasone-induced expression of the IPT gene reduced the PIN1 signal (Fig. S1A). In contrast, the decrease of CK levels by overexpression of the CKX3 gene (19) resulted in upregulation of the PIN1 signal (Fig. S1B). A more detailed time-lapse analysis revealed that shortly after CK application, PIN1:GFP expression is temporally enhanced and then gradually decreases (Fig. S3E). However, the negative effect of CK on PIN1 expression is complicated by the CK-induced ethylene production and subsequent induction of the PIN1 expression by ethylene (29). Accordingly, both ACC and CK treatments upregulated the PIN1::GUS expression (Fig. S4A) and the CK-induced PIN1 expression was completely abolished by simultaneous application of AVG (Fig. S4A). Similarly to PIN1, expression of PIN4 is also negatively controlled by CK. Expression of both PIN4::GUS and PIN4:GFP reporters was strongly downregulated by CK (Fig. S4E, data not shown) counteracting the positive effect of ethylene on the PIN4 transcription (29) (Fig. S4E). In contrast to PIN1 and PIN4, the expression of PIN2:GFP was relatively resistant to the CK treatment and an occasionally decreased expression was observed only after longer (over 24 h) treatment with CK (Fig. 3E; Fig. S3 C and E). Because PIN2 expression is strongly stimulated by ethylene (29), the inhibitory effect of CK was more pronounced when the ethylene biosynthesis was inhibited (Fig. S4B). In the case of PIN3, short-term CK treatment upregulated and longer treatments reduced the PIN3:GFP expression (Fig. 3F; Fig. S3 B and E) and cotreatments with AVG and ACC did not indicate any strong influence of ethylene (Fig. S4C and S3B). In contrast to other PIN genes, expression of PIN7:GFP and PIN7::GUS was clearly upregulated by CK with no significant influence of ethylene (Fig. 3G and Figs. S3 D and E and S4D). The real time Q-RT PCR experiments in general corroborated reporter-based observation and confirmed that CK has a negative effect on transcription of PIN1, PIN2, and PIN3 and a positive effect on PIN7 (Fig. 3C). Nonetheless, the time dynamic and extent of response differs for each particular PIN gene. Higher CK concentrations (10 μM BA) even after short-term treatments show stronger effects on PIN transcription (Fig. S4F).

Fig. 4.

Dependence of CK-mediated inhibition of PIN1 expression on CK perception. (A) Reduced expression of PIN1 in wild-type, but not in the cre1–12 x ahk2–2 double-mutant root meristem after CK treatment. (B) CK-resistant root meristem of cytokinin receptor mutants cre1–12, cre1–12 x ahk2–2, cre1–12 x ahk3–3, and CK-sensitive root meristem in ahk2–2, ahk3–3, and ahk2–2 x ahk3–3 mutants (*significantly different from control, Student's t test P < 0.05). Six-day-old seedlings (A) incubated for 20 h in control media or supplemented with 1 μM BA, (B) grown on control media or supplemented with 0.1 μM BA, respectively. Red signal, immunolocalization of PIN1 by PIN1-specific antibodies. Control, wild type, CK represented by BA, error bars represent SD.

In summary, our results show that CK in a concentration-dependent manner differentially regulates transcription of PIN auxin efflux carriers in roots. Despite the fact that the effects of CK on PIN transcription are obvious and can account for CK effects on root meristem activity, there might also be additional CK-dependent posttranscriptional regulations involved.

CK Effect on PIN1 Transcription Requires Histidine Kinase-Based CK Signaling.

We investigated the requirement of the CK signaling for CK-dependent regulation of PIN transcription and root meristem growth. To corroborate whether the CK effect on the PIN1 expression does not depend on ethylene, we treated seedlings with CK and performed immunodetection of PIN1 in an etr1–3 mutant that is defective in ethylene perception (26). In this experimental setup, CK reduced the expression of PIN1 in the etr1–3 mutant similarly to control roots (Fig. S5A), confirming that ethylene biosynthesis and perception are not required for the CK effect on root meristem growth.

It is known that CK signals through the histidine kinase family of receptors (32). To investigate the involvement of the CK perception in the CK effect on root meristem, we analyzed CK sensitivity of single and multiple receptor mutants in terms of root meristem size. While root meristems of the single CK receptor mutants ahk2–2 and ahk3–3 showed an almost normal CK sensitivity, cre1–12 and its multiple mutant combinations (cre1–12 × ahk2–2 and cre1–12 × ahk3–3) displayed reduced CK sensitivity (Fig. 4B). These results demonstrate that the CK control over the root meristem requires a functional CK perception pathway.

We tested whether CK receptors are similarly involved in the regulation of the PIN1 expression. Immunolocalization of PIN1 revealed that PIN1 expression was reduced following CK treatment of the ahk2–2 single mutant, but exhibited increased resistance to CK and remained unchanged in cre1–12 x ahk2–2 double mutant (Fig. 4A; data not shown). A similar set of experiments using PIN1:GFP revealed that CK reduces PIN1 expression with lower efficiency in cre1–12 background (Fig. S5B). Thus, the CK effect on PIN1 transcription correlates well with the CK-sensitive or CK-resistant root meristem of particular CK receptor mutant combinations. In conclusion, our data demonstrated that CK controls meristem size and that the PIN1 expression requires functional CK, but not ethylene perception.

Model for Auxin–CK Interaction in Root Meristems.

The interaction of auxin and CK plays an important role in the regulation of plant development and organogenesis (20, 33, 34). Early experiments on tobacco pith tissue cell cultures revealed that auxin and CK are essential hormones for maintenance of cell proliferation, but also for the regeneration of plant organs (21). Amazingly, the decision of whether a cell culture stays in the proliferating status or new organs, such as shoots or roots, are formed depends on the concentration ratio between these 2 plant hormones (21). Another interesting aspect of the auxin–CK crosstalk is that their interaction mode strongly depends on the developmental context: they act synergistically to stimulate the mitotic activity of cells in cell cultures (21), but antagonistically in root and shoot branching (for review, refs. 35 and 36). When focused at the apical root meristem, the role for both auxin and CK is well defined. Auxin is an important regulator of stem cell niche establishment (2, 3), cell proliferation occurring in the meristematic zone (4–6), and differentiation and rapid elongation of cells leaving the meristematic zone (7). Recently, significant progress on the role of CK has been made. CK levels have been shown to be critical for maintenance of the root meristem size and the site of CK action is primarily the regulation of the ratio between cell division and differentiation of cells leaving the meristem. In addition, a role for the CK signaling pathway has been suggested in stem cell niche establishment during embryogenesis (20).

Because auxin and CK closely overlap in the regulation of root meristems the question arises “how are their overlapping activities coordinated, and what are the molecular mechanisms behind their interaction?” Müller and Sheen (20) revealed that auxin regulates transcription of ARR7 and ARR15, negative regulators of the CK signaling pathway, suggesting that auxin might control output of the CK signaling pathway through modulation of transcription of one of its critical components. Here, we show a novel mechanism of CK–auxin interaction involved in the control of the root meristem size. CK by modifying the expression of several PIN genes might regulate cell-to-cell auxin transport, and thus the actual level of auxin in the cells, providing the amount of signal for downstream auxin signaling.

Materials and Methods

Plant Materials and Growth Conditions.

Seeds were chloral-gas sterilized, plated (0.5 × MS medium with 1% sucrose), stored for 2 days at 4 °C in the dark, and then grown under a 16-h photoperiod at 20 °C.

The following transgenic lines were characterized elsewhere: 35S::CKX2, 35S::CKX3 (37); etr1–3 and ein2–1 (26); pOp::ipt/Lh6r (38), CycB1;1::GUS_DB (39); DR5rev::GFP; PIN1,2,3,4,7::GUS; PIN1:GFP (8, and references herein); PIN2:GFP (40), PIN3:GFP, PIN4:GFP, PIN7:GFP (2), cre1–12, ahk2–2, ahk3–3 and their double-mutant combinations (41). PIN1::PIN1:mRFP1 gene construct was obtained from PIN1::PIN1:GFP (8) by replacing GFP coding sequence with mRFP1. Tobacco BY-2 cells (Nicotiana tabacum L. cv. Bright Yellow 2) were transformed by cocultivation with Agrobacterium tumefaciens strain C58C1 carrying PIN1::PIN1:mRFP1 according to ref. 42 and cultured as described (14).

ACC stock solution (Sigma-Aldrich) was prepared as 10 mM, AVG (Fluka) as 10 mM, 2,4-D (Duchefa) as 1 mM, dissolved with water, NAA (Duchefa) as 10 mM, dexamethasone (Sigma-Aldrich) as 5 mM, dissolved with dimethyl sulfoxide, and N⁶-benzyladenine (BA) (Sigma-Aldrich) as 50 mM, dissolved with 1 N sodium hydroxide. Dexamethasone induction of IPT in pOp::ipt/Lh6r was done as decribed (16, 17).

Phenotypical Analysis, Microscopy and Statistics.

Roots and root meristem regions were measured as published (29) with the following modification: root meristem length was assessed as the distance between the QC and the first elongating cell, the experimental image data sets were then randomly measured using the Altap Salamander Renamer (www.altap.cz) and ImageJ programs (National Institutes of Health, http://rsb.info.nih.gov/ij).

Automated whole mount protein immunolocalization was done as described (43); each sample was processed in 3 technical replicates (each ≈10 seedlings) and all experiments were repeated at least twice. The bright field and GFP fluorescence photographs were obtained by a Zeiss Axiophot microscope equipped with an Axiocam HR CCD camera. For the confocal laser scanning microscopy, a Leica TCS SP2 AOBS was used. GFP fluorescence of membrane-localized proteins was quantified as published (29). Images presented were processed in Adobe Photoshop.

Histochemical GUS stainings and root tissue clearing was done as published (43). Data were statistically evaluated with NCSS 2007 (www.ncss.com). Equal variances of values were verified by Levene test; a Mann-Whitney nonparametric test was performed simultaneously with Student's t test.

PIN1:mRFP1 Expression Analysis in BY-2 Cells.

For treatment with CK, stationary 7-day-old cells were inoculated into fresh medium containing (5 μM) BA after 24 h of cultivation. Cells were then incubated for another 24 h and auxin accumulation assays or confocal microscopy was performed. For observations of PIN1::PIN1:mRFP1 in BY-2 cells, a Zeiss LSM 5-DUO confocal microscope with a 40× C-Apochromat objective (N.A. = 1.2 W) was used.

Auxin Accumulation Assays.

[³H]NAA accumulation into the cells was measured in 0.5-mLiter cell suspension aliquots as described (14).

Quantitative RT-PCR.

RNA was extracted with the RNeasy kit (Qiagen) from root samples (last 2 mm of the root tip). A DNase treatment with RQ1 RNase-Free DNase (Promega) was carried out for 30 min at 37 °C. Poly(dT) cDNA was prepared from 1 μg total RNA with SuperScript III Reverse Transcriptase (Invitrogen) and quantified with a LightCycler 480 (Roche) with the LightCycler 480 SYBR GREEN I Master (Roche) according to the manufacturer's instructions. PCR was carried out in 384-well optical reaction plates heated for 10 min to 95 °C to activate hot start TaqDNA polymerase (Roche), followed by 40 cycles of denaturation for 60 s at 95 °C and annealing/extension for 60 s at 58 °C. Targets were quantified with specific primer pairs designed with the Beacon Designer 4.0 (Premier Biosoft International, Palo Alto, CA). Expression levels were normalized to ACTIN2 expression levels. All RT-PCR experiments were done at least in triplicates. The statistical significance was evaluated by the t test. The following primers were used: ACTIN2 (TTGACTACGAGCAGGAGATGG and ACAAACGAGGGCTGGAACAAG), PIN1 (TACTCCGAGACCTTCCAACTACG and TCCACCGCCACCACTTCC), PIN2 (CCTCGCCGCACTCTTTCTTTGG and CCGTACATCGCCCTAAGCAATGG), PIN3 (GAGGGAGAAGGAAGAAAGGGAAC and CTTGGCTTGTAATGTTGGCATCAG), PIN7 (CGGCTGATATTGATAATGGTGTGG and GCAATGCAGCTTGAACAATGG), and ARR5 (ACACTTCTTCATTAGCATCACCG and CTCCTTCTTCAAGACATCTATCGA).