The role of Arabidopsis Actin-Related Protein 3 in amyloplast sedimentation and polar auxin transport in root gravitropism.

Gravitropism is vital for shaping directional plant growth in response to the forces of gravity. Signals perceived in the gravity-sensing cells can be converted into biochemical signals and transmitted. Sedimentation of amyloplasts in the columella cells triggers asymmetric auxin redistribution in root tips, leading to downward root growth. The actin cytoskeleton is thought to play an important role in root gravitropism, although the molecular mechanism has not been resolved. DISTORTED1 (DIS1) encodes the ARP3 subunit of the Arabidopsis Actin-Related Protein 2/3 (ARP2/3) complex, and the ARP3/DIS1 mutant dis1-1 showed delayed root curvature after gravity stimulation. Microrheological analysis revealed that the high apparent viscosity within dis1-1 central columella cells is closely associated with abnormal movement trajectories of amyloplasts. Analysis using a sensitive auxin input reporter DII-VENUS showed that asymmetric auxin redistribution was reduced in the root tips of dis1-1, and the actin-disrupting drug Latrunculin B increased the asymmetric auxin redistribution. An uptake assay using the membrane-selective dye FM4-64 indicated that endocytosis was decelerated in dis1-1 root epidermal cells. Treatment and wash-out with Brefeldin A, which inhibits protein transport from the endoplasmic reticulum to the Golgi apparatus, showed that cycling of the auxin-transporter PIN-FORMED (PIN) proteins to the plasma membrane was also suppressed in dis1-1 roots. The results reveal that ARP3/DIS1 acts in root gravitropism by affecting amyloplast sedimentation and PIN-mediated polar auxin transport through regulation of PIN protein trafficking.

Introduction

Plants can sense changes in their postion relative to the gravity vector and use this to redirect the growth of their organs for optimal growth and development. Plant gravitropism consists of three major phases: gravity sensing, gravity signal transduction, and gravitropic response (Blancaflor and Masson, 2003). In higher plants, the sensing site in roots is believed to be in the columella cells of the root caps, whereas in shoots gravity sensing occurs in the endodermal cells (Morita, 2010). Amyloplast sedimentation/movement in the gravity-sensing cells is important for gravity perception. According to the starch-statolith hypothesis, the sedimentation of amyloplasts (statoliths) in gravity-sensing cells (statocytes) can trigger the conversion of gravitational potential energy into biochemical signals (Kiss, 2000; Leitz ; Hashiguchi ; Toyota ). The Cholodny–Went theory proposed that asymmetric auxin redistribution between the upper and lower side of root or shoot triggers differential growth, resulting in the downward growth of roots and upright growth of shoots. Additional studies have supported the notion that asymmetric auxin redistribution is important for root gravitropism (Swarup ; Vanneste and Friml, 2009; Band ).

Actin filaments not only provide mechanical support for cells but also are involved in a variety of biological events (Pollard, 2007). It has been proposed that the actin cytoskeleton is a major regulator of gravitropism (Blancaflor, 2002); however, studies using actin cytoskeleton inhibitors have produced contradictory results regarding their effects on gravitropism (Yamamoto and Kiss, 2002; Hou , 2004; Palmieri and Kiss, 2005; Mancuso ). Recently, genetic studies have begun to uncover the molecular mechanisms behind how the actin cytoskeleton plays a role in gravitropism. In Arabidopsis (Arabidopsis thaliana) root tips, the central columella (CC) cells have a finer and less robust network of filamentous actin (F-actin) arrays in contrast to the prominent actin bundles in peripheral columella (PC) or lateral root cap (LRC) cells (Blancaflor, 2013). By contrast, the endodermal cells in inflorescence stems contain a network of distinct F-actin bundles (Saito ; Zhang ). The non-homogeneous structures in statocytes arising from intracellular components such as cytoskeletons and endomembranes have been shown to significantly affect the complex movements of amyloplasts (Saito ; Nakamura ). Treatment with the actin-disrupting drug Latrunculin B (Lat B) can increase the sedimentation of amyloplasts in the columella cells and promote root curvature in Arabidopsis (Hou ). ALTERED RESPONSE TO GRAVITY1 (ARG1), which encodes a DnaJ-like protein, potentially interacts with the actin cytoskeleton and is required for hypocotyl gravitropism through the regulation of amyloplast movement (Sedbrook ; Shiva Kumar ). An Arabidopsis E3 ligase SHOOT GRAVITROPISM9 (SGR9) localizes to endodermal amyloplasts and promotes detachment of amyloplasts from actin bundles, allowing the amyloplasts to sedimentate during shoot gravity sensing (Nakamura ). These observations demonstrate the role of the actin cytoskeleton in gravitropism by affecting amyloplast movement. Recently, the introduction of microrheological analysis has revealed the relationship between actin organization and amyloplast sedimentation in the columella cells (Zheng ).

The role of auxin in linking gravity sensing to response has been well established (Sato ). Asymmetric auxin redistribution between the upper and lower sides of gravity-stimulated roots causes differential growth in the root elongation zone, resulting in root curvature. It has been proposed that the actin cytoskeleton also plays an important role in the growth response phase of gravitropism by regulating auxin transport (Blancaflor, 2013). Auxin transport is mainly mediated by auxin transporters, including PIN-FORMED (PIN) group proteins (Vanneste and Friml, 2009). ARG1 and ARG1-LIKE2 (ARL2) are required for PIN3 relocalization and asymmetrical redistribution of auxin upon gravity stimulation (Harrison and Masson, 2008). Arabidopsis SPIKE1 (SPK1), which belongs to the conserved DHR2-Dock family of Rho guanine nucleotide exchange factors (Qiu ; Basu ), is required for RHO-LIKE GTPASE FROM PLANTS 6 (ROP6) activation and inhibits PIN2 internalization through the stabilization of actin filaments in roots, modulating auxin redistribution during gravitropic responses (Lin ). ROP6 and its downstream ROP-INTERACTIVE CRIB MOTIF-CONTAINING PROTEIN1 (RIC1) are required for auxin-mediated root gravitropism through regulating endocytosis and internalization of PIN1 and PIN2 (Chen ). These observations reveal the impacts of the actin cytoskeleton and its signaling pathway on auxin transport during gravitropism.

The organization and function of the actin cytoskeleton are regulated by diverse actin-binding proteins, including profilin, actin-depolymerizing factor, formin, and the Actin-Related Protein 2/3 (ARP2/3) complex (Staiger and Blanchoin, 2006). The ARP2/3 complex produces branched filaments both to push forward the leading edge of motile cells and for endocytosis (Pollard and Borisy, 2003). In Arabidopsis, mutations in the ARP2/3 complex usually lead to distorted trichomes and cause epidermal cell adhesion defects (Le ; Mathur ; El-Assal ). Arabidopsis ARP2/3 complex subunits have also been reported to be involved in stomatal movement (Jiang ; Li , 2014) and salt stress (Zhao ). Moreover, it has been reported that Arabidopsis ARP3/DISTORTED1 (DIS1) and ARPC2A/DISTORTED2 (DIS2) have different roles in gravitropism and phototropism (Reboulet ); however, the mechanism by which DIS1 is involved in gravitropism remains unclear. It has been hypothesized that DIS1 may affect translocation of auxin transporters or other actin-associated proteins involved in gravitropism (Reboulet ). Here we report that ARP3/DIS1 takes part both in amyloplast sedimentation by affecting local apparent viscosity in the central columella cells and in asymmetric auxin redistribution across the root tips through the modulation of PIN cycling.

Materials and methods

Plant materials and growth conditions

Arabidopsis thaliana ecotype Columbia (Col) was used as the wild type in this study. The dis1-1 and dis2-1 mutants were kindly provided by Daniel B. Szymanski (Purdue University, West Lafayette, Indiana, USA). pPIN2::PIN2-GFP, pPIN3::PIN3-GFP, pPIN7::PIN7-GFP, pDR5::GFP, pDR5::GUS, and p35S::DII-VENUS-N7 have been described previously (Le ). Arabidopsis seeds were surface-sterilized in an aqueous solution of 30% (w/v) hydrogen peroxide and 85% (v/v) ethanol at a ratio of 1:4 (v/v). The seeds were then sown onto half-strength Murashige and Skoog (MS) medium supplemented with agar (0.8%, w/v) and sucrose (1%, w/v) and kept in the dark for 2 d at 4 °C. Seedlings were grown in a vertical orientation in growth chambers under a 16-h light / 8-h dark cycle at 22±2 °C.

Vector construction and plant transformation

To generate the construct for complementation of dis1-1, a 4112-bp genomic sequence of DIS1 was amplified by PCR with specific primers (forward primer 5′-TTGAGCTCTTTATTACC TTGAAAACAGGTCATA-3′ and reverse primer 5′-TTGGTACCC TAGAACTTAAGCTCTTGAGTGGAA-3′; Sac I and Kpn I restriction sites underlined, respectively). The PCR product was verified by DNA sequencing and cloned into the pCAMBIA1300 vector (CAMBIA). The construct was then transformed into dis1-1 mutants. Transgenic plants were selected on half-strength MS medium containing 25 mg l−1 hygromycin. The T2 transgenic plants were used for root gravitropic experiments.

Curvature and growth analyses

To examine root gravity responses, 4-d-old seedlings grown on vertical plates were rotated 90° and photographs were taken at selected time points after reorientation. The root growth rate was calculated using the increase in length from the start time point at reorientation. To examine hypocotyl gravity responses, 3-d-old seedlings grown vertically on half-strength MS medium in darkness were reoriented at 90° for 24h. The curvature angles and root growth were measured using ImageJ (http://rsb.info.nih.gov/ij/).

Observation of amyloplast sedimentation

To image amyloplast movement in the central S2 columella cells (Leitz ), roots were prepared and mounted on a rotatable stage of a horizontally oriented BX51 microscope (Olympus, Japan). Differential interference images were captured at 1-s intervals for 600s after a 90° reorientation (Wang ).

Apparent local viscosity

Using the Stokes–Einstein relation, the apparent local viscosity η in each subcellular region was measured using the short-time diffusion coefficient D = k B T/2Rη , where k B is the Boltzmann constant, T is the temperature, and R is the amyloplast radius (Crocker ; Levine and Lubensky, 2000). We analyzed the Brownian motion and collective motion by measuring the short-time Brownian diffusion, the intermediate-time caged sub-diffusion, and the long-time normal diffusion of each amyloplast during sedimentation.

Actin labeling

Imaging of actin filaments in Arabidopsis roots was performed as previously described (Le ) with slight modifications. Briefly, the roots were incubated in PME buffer (50mM PIPES, 5mM MgSO4, and 5mM EGTA, pH6.9) containing 300 µM m-maleimidobenzoyl-N-hydroxysuccinimide ester for 30min. The roots were then incubated in PME buffer containing 2% paraformaldehyde in PME buffer for 1h. The roots were washed three times with PME buffer and then treated with 0.1% Y-23 for 10min. Samples were further washed with PME buffer and incubated in actin-staining buffer (PME, 1% glycerol, and 0.3M mannitol and 0.1 µM Alexa Fluor 488 phalloidin) (Invitrogen, USA) at 4 ºC in the dark overnight. Images were captured using a FV1000-MPE confocal laser-scanning microscope (Olympus, Japan). To quantitatively evaluate the actin bundling in gravity-sensing cells, the skewness of the actin fluorescence intensity distribution was measured as previously described using the Skewness plug-in on ImageJ (Higaki, ).

Latrunculin B treatment

Lat B treatment was conducted as previously described (Hou ). After seedlings grown in a vertical direction for 4 d, 200nM Lat B solution was added to the Petri dishes. After 1h treatment, the Lat B solution was removed and the Petri dishes were kept vertically for an additional 30min before the dishes were reoriented by 90°. Photographs were then taken at selected time points and the curvature of roots was measured using ImageJ.

DII-VENUS fluorescence intensity measurement

Four-day-old DII-VENUS (a sensitive auxin input reporter) seedlings grown vertically on the surface of half-strength MS medium were transferred onto a new plate. After another 2h vertical growth, the plates were rotated 90° from the original direction, then the seedlings were mounted and the fluorescence was imaged at selected time points using a FV1000-MPE confocal laser scanning microscope (Olympus, Japan). DII-VENUS fluorescence intensity was analyzed using ImageJ as previously described (Wang ).

FM4-64 staining and confocal microscope observation

Four-day-old seedlings were incubated in half-strength MS liquid medium containing 5 μg ml–1 of the membrane-selective dye FM4-64 (Invitrogen, USA) for 10min, followed by washing three times with half-strength MS liquid medium. After incubation in half-strength liquid MS medium for 20min at room temperature, the seedlings were mounted and the fluorescence was imaged using a FV1000-MPE confocal laser scanning microscope (Olympus, Japan).

Brefeldin A (BFA) treatment

To monitor cycling of PIN proteins, 4-d-old seedlings were incubated in half-strength MS liquid medium containing 50 μM BFA for 2h, followed by 1h and 2h of washing with half-strength MS liquid medium. The seedlings were then mounted and images were captured at selected time points using a FV1000-MPE confocal laser-scanning microscope (Olympus, Japan).

Results

Intracellular environment of root gravity-sensing cells revealed by microrheological analysis

To investigate the role of the ARP2/3 complex in root gravitropism, the ARP3/DIS1 mutant dis1-1 and ARPC2A/DIS2 mutant dis2-1 were used for root gravitropic analysis. Four-day-old seedlings grown vertically were reoriented by 90° and root curvature was measured at selected time points. As shown in Fig. 1A, the dis1-1 mutants showed reduced root curvature compared with wild-type plants after 90° reorientation of the roots. Consistent with previous work, dis2-1 showed similar root curvature to the wild-type plants (Reboulet ). The wild-type plants and dis1-1 mutants showed similar root growth rates after gravity stimulation, indicating that root growth was not impaired in dis1-1 mutants (Fig. 1B). In contrast to the different gravitropic responses found in dis1-1 and dis2-1 roots, these two mutants showed similar increases in hypocotyl curvature after 90° reorientation for 24h in darkness, indicating the different regulatory mechanisms of ARP3/DIS1 and ARPC2A/DIS2 in response to gravity stimulation between roots and shoots (Supplementary Fig. S1 at JXB online).

Fig. 1.

The ARP2/3 complex subunit DIS1 mutant dis1-1 displayed reduced root curvature and decreased amyloplast movement compared with wild-type plants. (A) Root curvature of Col, dis1-1, and dis2-1 after 90° reorientation. Four-day-old seedlings were rotated by 90° to test the gravitropic response, with angles of root curvature measured at the indicated time points. Values are means ±SD (n=30–50 seedlings). Asterisks indicate significant differences relative to Col (Student’s t-test, *P<0.05, **P<0.01). (B) Root growth rates (mm h–1) of wild-type plants and dis1-1 mutants. Values are means ±SD (n=35–45 seedlings). (C) Root curvature of Col, dis1-1, and two complementation lines. Values are means ±SD (n=20–50 seedlings). Asterisks indicate significant differences relative to Col (Student’s t-test, **P<0.01). (D) Time-lapse images of amyloplast sedimentation in the central columella cells of Col, dis1-1, and dis2-1 after 90° reorientation. Black arrows at top-right indicate the direction of the gravity vector before (solid line) and after (dashed) reorientation. Scale bars are 2 µm.

To test whether root gravity defects in dis1-1 mutants result from ARP3/DIS1 disruption, the genomic DNA sequence of ARP3/DIS1 was introduced into the dis1-1 mutants. The transgenic complementation lines can rescue dis1-1 trichome and root gravity defects, indicating that ARP3/DIS1 takes part in root gravitropism (Fig. 1C and Supplementary Fig. S2).

It has been reported that mutation in ARP3/DIS1 induced the formation of disorganized, thick actin bundles in developing trichome branches (Le ). Sedimentation of amyloplasts is correlated with intracellular components such as actin filaments during gravity sensing (Blancalfor, 2013). To test the influence of actin organization on amyloplast movement in CC cells of wild-type and dis1-1 plants, the dynamic movements of amyloplasts in the CC cells were captured using time-lapse imaging after 90° reorientation. As shown in Fig. 1D, amyloplasts are initially located at the bottom of the CC cells. After reorientation for 400s, most of the amyloplasts reached the new bottom side of the CC cells. Most amyloplasts in the dis1-1 and dis2-1 mutants, however, stayed in the middle of the CC cells (Fig. 1D), indicating that the intracellular environment of the gravity-sensing cells in these mutants might be different from that in the wild-type plants. We then labeled actin filaments in the root cells using Alexa Fluor 488 phalloidin dyes (Le ). Differing from the formation of actin filaments/bundles in PC and LRC cells, only diffuse fluorescent signals were observed in the CC cells of wild-type plants, consistent with results reported previously (Hou ) (Fig. 2A). By contrast, the dis1-1 and dis2-1 mutants displayed thick actin bundles surrounding the amyloplasts in the CC cells, as well in the PC and LRC cells (Fig. 2B, C). To quantitatively evaluate the bundling of actin, the skewness of the actin fluorescence intensity distribution in CC cells was measured (Higaki, ). The significantly increased values of skewness in the dis mutants revealed that the formation of actin bundles may alter the amyloplast kinetics in the CC cells (Fig. 2D).

Fig. 2.

The dis1-1 mutant showed different actin cytoskeleton organization and local apparent viscosity in the central columella cells compared with wild-type plants. (A-C) Organization of the actin cytoskeleton in the central columella cells of plants of Col (A), dis1-1 (B), and dis2-1 (C). Actin filaments in fixed root tips were labeled with Alexa Fluor phalloidin dyes and viewed with a confocal microscope. Inset: enlarged CC cell. The cell outline is indicated by dashed-lines. Scale bars are 10 µm. (D) Microfilament bundling (skewness) was measured in the CC cells of Col, dis1-1, and dis2-1 plants. Values are means ±SD (n=5–9 seedlings). Different letters indicate significantly different means (one-way ANOVA test). (E, F) Movement trajectories of amyloplasts in the central columella cells after 90° reorientation and apparent local viscosity in the corresponding cells in wild-type (E) and dis1-1 (F) plants. Polygonal lines in different shades denote different amyloplasts. (This figure is available in color at JXB online.)

Microrheology has been rapidly developed as a powerful method to explore the relationships between local mechanical responses and local structures in inhomogeneous fluids (Wirtz, 2009). The amyloplasts in the columella cells can be used as native microprobes to analyze the inhomogenous intracellular environment. Recently, we implemented a novel method for measurement of diffusive dynamics and in planta microrheological analysis of amyloplasts by multi-particle tracking in the CC cells of Arabidopsis root caps (Zheng ). We found that actin organization dominated the intracellular environment of CC cells and highlighted the spatial heterogeneity and the cage-confinement of amyloplasts characterized by the local apparent viscosity, η (Zheng ). Here, we plotted the spatial coupling of the movement trajectories of each amyloplast and the local viscosity in corresponding sub-regions of columella cells (Fig. 2E, F). In the wild-type plants, amyloplast trajectories showed a frequent small-step rattling motion with an occasional large-step chain-like motion. The confined Brownian motion within cages coincides with high local viscosity (dark gray) regions, while the co-operative out-of-cage motion of several amyloplasts emerges in low local viscosity (light gray) regions (Fig. 2E). By contrast, in the dis1-1 mutant, the amyloplasts rattled randomly within separate cages but did not undergo a co-operative cage escape (Fig. 2F). This indicates a stronger cage confinement that can be characterized by the greatly increased local viscosity and its spatial fluctuation. In each type of CC cells, the compact and loose trajectories of amyloplasts respectively correspond to the higher and lower local viscosity in that sub-region (Fig. 2E, F). Taken together, this microrheological analysis indicates that the actin cytoskeleton functions in affecting amyloplast movements through regulating local viscosity in the CC cells.

ARP3/DIS1 is also required for gravity signal transduction

In addition to its role in gravity sensing, we then questioned whether ARP3/DIS1 functions in gravity signal transduction. The formation of starch-filled amyloplasts, the statoliths, is very important for gravity sensing (Sack, 1997). PHOSPHOGLYCERATE/BISPHOSPHOGLYCERATE MUTASE (PGM) is involved in starch biosynthesis. The starchless pgm mutant exhibits a delayed gravitropic response in roots (Caspar and Pickard, 1989; Kiss ). To investigate whether dis1-1 takes part in both gravity sensing and gravity signal transduction, dis1-1 pgm double-mutants were generated to examine their root gravitropic responses. As shown in Fig. 3A and B, dis1-1 and pgm single-mutant roots showed similar root gravitropic defects, whereas dis1-1 pgm double-mutants displayed much stronger root gravitropic defects than the single-mutants. When treated with Lat B, the Col and dis1-1 seedlings showed a similar enhanced bending response, indicating that breaking down of the actin network in the root caps can rescue root gravitropic defects in the dis1-1 mutant (Fig. 3C). This enhanced gravitropic response in Lat B-treated roots was reduced in both pgm and dis1-1 pgm mutants (Fig. 3C). Together with amyloplast movement, this indicates that ARP3/DIS1 takes part in both gravity sensing and gravity signal transduction phases.

Fig. 3.

Root curvature of Col, dis1-1, pgm, and dis1-1 pgm seedlings. (A) Images of root curvature of untreated and Lat B-treated Col, dis1-1, pgm, and dis1-1 pgm seedlings after gravity stimulation. Four-day-old seedlings were rotated by 90° to test their gravitropic response. Images were taken 6h and 24h after reorientation. The black arrow at the top-left indicates the direction of the gravity vector after reorientation. Scale bars are 0.5cm. (B, C) Quantification of root gravitropic response in Col, dis1-1, pgm, and dis1-1 pgm seedlings before and after treatment with 200nM Lat B for 1h. Four-day-old seedlings were rotated by 90° to test their gravitropic response. Values are means ±SD (n=20–40 seedlings). Asterisks indicate significant differences relative to Col (Student’s t-test, **P<0.01). (This figure is available in color at JXB online.)

Asymmetric auxin redistribution in dis1-1 root tips is delayed during the gravitropic response

During gravity signal transduction, asymmetric auxin redistribution between the upper and lower side of root tips can cause differential root growth and lead to root curvature (Band ). To test whether DIS1 regulates asymmetric redistribution of auxin after the gravitropic response, a sensitive auxin input reporter, DII-VENUS, was introduced to monitor the speed and magnitude of changes in auxin distribution during the root gravitropic response (Brunoud ). Gravity-induced rapid auxin redistribution to the lower side of the root tips occurred within minutes of a 90° gravity stimulation. In the root tips, cells within the lateral root cap mediate the creation of shootward auxin fluxes (Ottenschläger ; Swarup ). As shown in Fig. 4A and B, the DII-VENUS signal was reduced in LRC cells on the lower side of root tip after a 90° gravity stimulation for 30min. This asymmetric auxin redistribution between the upper and lower sides of the root tip was reduced in the dis1-1 mutant (Fig. 4D, E). When treated with Lat B, both wild-type and dis1-1 plants showed increased asymmetric auxin redistribution in the root tips (Fig. 4C, F). Auxin asymmetry was quantitatively analyzed by measuring the DII-VENUS signal ratios between the upper and the lower sides of LRC cells adjacent to the columella cells. In wild-type roots, the DII-VENUS ratios continued to increase after 90° gravity stimulation and were approximately two-fold higher after 30min. However, increases in the DII-VENUS ratios were significantly smaller in dis1-1 plants. Disruption of the actin cytoskeleton with Lat B significantly increased DII-VENUS ratios in wild-type plants and dis1-1 mutants relative to untreated plants (Fig. 4G). We also monitored the expression patterns of an auxin activity reporter, pDR5::GFP, in wild-type and dis1-1 root tips after gravity stimulation. Analysis of the pDR5::GFP expression pattern showed that stronger GFP signals were found on the lower side of root tips in the wild-type compared with the dis1-1 mutant after 4h of gravity stimulation (Fig. 4H–K). Analysis of pDR5::GUS expression patterns during gravity stimulation also showed similar results to pDR5::GFP (Supplementary Fig. S3).

Fig. 4.

The dis1-1 mutant showed decelerated asymmetric auxin distribution between the upper and lower side of the root tip following a 90° gravity stimulation. (A–F) Representative heat map images of DII-VENUS fluorescence show the asymmetric distribution of DII-VENUS fluorescence in the root tips of wild-type and dis1-1 plants after 90° reorientation at the indicated time points. Cell outlines were visualized after staining with propidium iodide. The black arrow at the top-left indicates direction of the gravity vector after reorientation. The bar on the lower-right indicates the signal intensity range from high (H) to low (L). Scale bars are 10 µm. (G) The fold-changes in DII-VENUS ratios between wild-type and dis1-1 following a 90° gravity stimulation at the indicated time points. Values are means ±SD (n=3–9 seedlings). Different letters indicate significantly different means (one-way ANOVA test). (H–K) Heat map images of pDR5::GFP show the asymmetric distribution of pDR5::GFP in root tips of wild-type and dis1-1 plants after 90° reorientation at the indicated time points. The arrows indicate the distribution of pDR5::GFP expression on the lower side of the root caps. The direction of the gravity vector after reorientation is indicated at the top-left. The bar on the lower-right indicates the signal intensity range from high (H) to low (L). Scale bars are 10 µm. (This figure is available in color at JXB online.)

In addition, we tested whether exogenous auxin (IAA or NAA) can rescue dis1-1 root gravitropic defects. Three-day-old seedlings were transferred to half-strength MS medium containing IAA (1 or 10nM) or NAA (1 or 10nM) for 18h. The root curvature of wild-type and dis1-1 plants was then measured at 4h and 24h after 90° reorientation. As shown in Supplementary Fig. S4, neither IAA nor NAA could rescue the reduced root curvature in the dis1-1 mutants. These findings showed that the delayed root gravitropic response in dis1-1 is not caused by decreased overall auxin accumulation, and that ARP3/DIS1 may regulate the root gravitropic response by affecting polar auxin transport.

Vesicle trafficking is defective in dis1-1 mutants

The ARP2/3 complex has previously been shown to be important for actin filament assembly and is needed for cell motility, vesicle trafficking, and endocytosis (Rotty ). It has been reported that in Arabidopsis PIN proteins undergo constitutive endocytic recycling between the plasma membrane and the endosomal compartments (Kleine-Vehn ). It was hypothesized that ARP3/DIS1 might regulate auxin transport by affecting vesicle trafficking (Reboulet ). FM4-64 is a water-soluble marker that is widely used to study endocytosis, vesicle trafficking, and organelle organization in living eukaryotic cells (Bolte ). We therefore used FM4-64 to monitor endocytosis in root epidermal cells of wild-type, dis1-1, and dis2-1 plants. As shown in Fig. 5A, after 30min of staining, the FM4-64 dye was internalized and substantial numbers of punctuated fluorescent vesicles were detected in the cytosol of wild-type root epidermal cells. Conversely, only a few fluorescent vesicles were observed in the root epidermal cells of dis1-1 mutants, indicating that endocytosis is defective in the dis1-1 mutant (Fig. 5B). The dis2-1 mutants showed a similar result to the wild-type plants after FM4-64 staining (Fig. 5C). Quantification of FM4-64 uptake showed a significantly decreased uptake of FM4-64 in dis1-1 mutants compared with wild-type plants and dis2-1 mutants (Fig. 5D). These results suggest that ARP3/DIS1 positively regulates endocytosis.

Fig. 5.

DIS1 deficiency caused decelerated vesicle trafficking. (A–C) Vesicle trafficking was suppressed in dis1-1 compared with Col and dis2-1 plants. Roots of 4-d-old seedlings stained with FM4-64 (5 µg ml–1, 30min) were observed under a confocal microscope. Scale bars are 10 µm. (D) Relative FM4-64 internalization fluorescence intensity in Col, dis1-1, and dis2-1 cells. Values are means ±SD (n=20–22 cells). Different letters indicate significantly different means (one-way ANOVA test). (This figure is available in color at JXB online.)

ARP3/DIS1 mutation causes decelerated PIN cycling

During gravity signal transduction, polar auxin transport is affected by PIN proteins (Křeček ). The actin cytoskeleton takes part in auxin transporter endocytosis and cycling by regulating vesicular trafficking (Zhu and Geisler, 2015). In Arabidopsis, PIN3 and PIN7 have been shown to localize to the columella cells and to exhibit constitutive intracellular cycling between the plasma membrane and endosomal compartments (Kleine-Vehn ). The fungal toxin BFA, a vesicle transport inhibitor that can inhibit protein trafficking in the endomembrane system to form BFA compartments, has been used to investigate PIN trafficking (Nebenführ ; Chen ; Lin ). To determine whether DIS1 regulates auxin transport by affecting PIN3 and PIN7 cycling in the columella cells, pPIN3::PIN3-GFP and pPIN7::PIN7-GFP lines were introduced into the dis1-1 mutant background. After 2h of BFA treatment, PIN3-GFP and PIN7-GFP aggregated into BFA bodies in the columella cells of both wild-type plants and the dis1-1 mutants (Fig. 6B, E, H, K). For PIN3, as BFA has a stronger effect on the intracellular PIN3 trafficking in gravity-stimulated roots than on non-stimulated roots (Kleine-Vehn ), there was no obvious difference in accumulated PIN3-GFP in BFA bodies between the wild-type and the dis1-1 mutant. After 2h of BFA wash-out, normal plasma membrane localization of PIN3-GFP or PIN7-GFP was recovered in the wild-type plants (Fig. 6C, I). However, aggregated GFP fluorescence remained in the dis1-1 mutants, indicating that PIN3 and PIN7 recycling to the plasma membrane in the columella cells is regulated by ARP3/DIS1 (Fig. 6F, L).

Fig. 6.

Intracellular cycling of PIN3 and PIN7 is decelerated in the dis1-1 mutants. (A–F) Heat map images of pPIN3::PIN3-GFP fluorescence in 4-d-old root cells of wild-type (A–C) and dis1-1 (D–F) seedlings after treatment with 50 µM BFA and subsequent wash-out. The arrows indicate BFA bodies. The bar on the lower-right indicates the signal intensity range from high (H) to low (L). Scale bars are 10 µm. (G–L) Heat map images of pPIN7::PIN7-GFP fluorescence in 4-d-old root cells of wild-type (G–I) and dis1-1 (J–L) seedlings after treatment with 50 µM BFA and subsequent wash-out. The arrows indicate BFA bodies. The bar on the lower-right indicates the signal intensity range from high (H) to low (L). Scale bars are 10 µm. (This figure is available in color at JXB online.)

PIN2 is localized to the apical end of epidermal cells and the basal end of cortical cells in the root tips and is critical for the root gravitropic response by regulating auxin redistribution. Dynamic changes in PIN2 proteins are also important for auxin flows during the root gravitropic response (Chen ; Müller ; Rahman ; Lin ; Sassi ; Rigó ). Polar localization and expression of PIN2 were not affected in the dis1-1 mutants compared with wild-type plants under normal growth conditions (Fig. 7A, E). After 1h of BFA treatment, the accumulation of PIN2-containing BFA bodies increased in the dis1-1 mutants compared with wild-type plants (Fig. 7B, F, I). This effect was amplified after 2h of BFA treatment (Fig. 7C, G, I). After 2h of BFA wash-out, almost all the BFA bodies disappeared and normal polar localization of PIN2-GFP was detected in the epidermal cells of wild-type plants (Fig. 7D). Conversely, small numbers of BFA bodies still accumulated in the epidermal cells of dis1-1 plants, indicating that accumulated PIN2-GFP bodies were not recovered effectively (Fig. 7H). As for the dis2-1 mutant, there was no difference in the accumulation of PIN2-containing BFA bodies between wild-type plants and the dis2-1 mutants (Supplementary Fig. S5), indicating that ARP3/DIS1 and ARPC2/DIS2 may play different roles in regulating PIN protein cycling. When the actin cytoskeleton organization was analyzed in the epidermal cells of the root transition zone where PIN2 is localized, no obvious differences were found among Col, dis1-1, and dis2-1 (Supplementary Fig. S6).

Fig. 7.

PIN2 cycling is decelerated in dis1-1 mutants compared with wild-type plants. (A–H) Internalization of PIN2 between Col and dis1-1 plants after treatment with 50 µM BFA and subsequent wash-out. The arrows indicate BFA bodies. Scale bars are 10 µm. (I) Quantification of BFA bodies in Col and dis1-1 plants. Values are means ±SD (n=40–90 cells). Different letters indicate significantly different means (one-way ANOVA test). (J–M) Observation of fluorescence of pPIN2::PIN2-GFP in 4-d-old root cells of wild-type (J, K) and dis1-1 (L, M) seedlings following the gravity stimulation. The arrow on the lower-right indicates direction of the gravity vector after reorientation. Scale bars are 10 µm. (O–R) Quantification of pPIN2::PIN2-GFP fluorescence intensity at the plasma membrane in the upper and lower epidermis of roots of wild-type plants (O, P) and dis1-1 mutants (Q, R). The pPIN2::PIN2-GFP fluorescence intensity is calculated along the lines shown in (K) and (M). (This figure is available in color at JXB online.)

Next, we examined the redistribution of PIN2-GFP in wild-type and dis1-1 plants after gravity stimulation. After 90° reorientation of root tips for 4h, PIN2-GFP signals in wild-type plants were higher on the lower side of the root tip than on the upper side, with the difference being pronounced in the epidermal cells (Fig. 7J, K, O, P). By contrast, PIN2-GFP signals in dis1-1 roots were not significantly different between the upper and lower sides of the epidermal cells (Fig. 7L, M, Q, R). These results indicate that ARP3/DIS1 participates in PIN2 internalization and recycling between the plasma membrane and endosomal compartments in the root tips.

Discussion

The actin cytoskeleton has been proposed to be an important component of gravity sensing and signal transduction, with pharmacological and genetic evidence beginning to reveal the roles of the actin cytoskeleton in root gravitropism (Blancaflor, 2013). In this study, we provide evidence that the ARP2/3 complex subunit ARP3/DIS1 is involved in the root gravity response by affecting both amyloplast sedimentation and PIN-mediated polar auxin transport.

Microrheological analysis provides new insights by revealing the role of the actin cytoskeleton in gravity sensing

Although the behavior of amyloplast sedimentation in gravity-sensing cells has been investigated intensively (Sack ; Yoder ; Saito ; Leitz ; Nakamura ; Toyota ), the effects of intracellular components, including vacuoles and the actin cytoskeleton, on amyloplast movement have not been characterized in detail (Saito ; Nakamura ). In this study, pharmacological treatments that disrupt the actin filaments induced a rapid and free diffusive sedimentation of amyloplasts, while the actin bundles that formed in actin mutants caused restrained sedimentation of amyloplasts (Fig. 1D and Fig. 2). Consistent with these results, pharmacological treatments that disrupt the actin filaments have previously been shown to induce an enhanced gravity sensitivity in roots, hypocotyls, and inflorescence stems (Yamamoto and Kiss, 2002; Hou , 2004; Nakamura ). Thus, in contrast to the formation of actin bundles in other cells, the fine actin filament network in the root CC cells may provide a suitable intracellular environment for the unimpeded sedimentation of amyloplasts (Blancaflor, 2013).

Furthermore, our microrheological analysis revealed the compact and loose trajectories of amyloplasts corresponded to higher and lower local apparent viscosity in different subregions of the central columella cells, respectively (Fig. 1D and Fig. 2)

ARP3/DIS1 plays an important role during root gravity sensing and signal transduction

Sedimentation of amyloplasts in the columella cells provides the means for converting the gravitational potential energy into a biochemical signal (Leitz ). Sedimentation of amyloplasts onto the lower side of the columella cells can trigger the formation of the lateral auxin gradient. In our study, the dis1-1 pgm double-mutants showed strong gravitropic defects in roots compared with single-mutants, supporting the proposed role for ARP3/DIS1 in gravity signal transduction in addition to gravity perception (Fig. 3). Asymmetric auxin redistribution between the upper and lower side of the root tips was reduced in dis1-1 mutants relative to that in wild-type plants after gravity stimulation (Fig. 4), and analysis showed that ARP3/DIS1 mediates vesicle trafficking during gravity signal transduction (Fig. 5). These results further confirm the role of APR3/DIS1 in root signal transduction through regulating auxin polar transport.

ARP3/DIS1 is involved in PIN-mediated auxin transport

Cross-talk between polar auxin transport and the actin cytoskeleton in plant development has been the subject of discussion (Zhu and Geisler, 2015). Polar auxin transport is mediated by specific auxin influx and efflux carriers, with the auxin efflux carrier PIN family reported to be important for polar auxin transport during gravitropism (Swarup ; Křeček ).

PIN proteins redirect the auxin flows, representing an important response mechanism to gravity stimulation (Kleine-Vehn ; Baster ). The abundance and localization of PIN proteins at the plasma membrane are finely regulated and controlled by transcriptional regulation (Blilou ; Cui ; Garay-Arroyo ; Wang ), phosphorylation regulation (Sukumar ; Huang ; Ganguly ), or degradation (Kleine-Vehn ; Sassi ). PIN protein trafficking between the plasma membrane and intracellular compartments is also crucial for root gravitropism (Lin ; Mei ). It has been reported that actin filaments regulate vesicle trafficking and cycling of PINs, such as PIN1 (Geldner ), PIN2 (Chen ; Lin ), and PIN3 (Friml ; Harrison and Masson, 2008). Our results provide evidence that ARP3/DIS1 is also important for PIN trafficking and cycling during root gravitropism. dis1-1 mutants showed reduced FM4-64 dye uptake, indicating that ARP3/DIS1 may function in PIN internalization (Fig. 5). BFA treatment and wash-out experiments indicated that the cycling of PIN2, PIN3, and PIN7 is dependent on the function of ARP3/DIS1 (Figs 6 and 7). These results indicated that ARP3/DIS1-mediated actin organization also participates in asymmetric auxin redistribution by regulating PIN internalization and recycling.

The roles of ARP3/DIS1 and ARPC2A/DIS2 in root gravitropism

Previous reports have indicated that the dis1-1 and dis2-1 mutants show similar defects in trichomes and hypocotyls (Le ; El-Assal ; Basu ); however, they display different responses to gravitropism and phototropism (Reboulet ). In the CC cells of root tips, dis1-1 and dis2-1 mutants show similar actin bundling and amyloplasts sedimentation, indicating that both ARP3/DIS1 and APRC2A/DIS2 may contribute to the function of ARP2/3 during the gravity sensing phase.

In Saccharomyces cerevisiae, analyses of actin nucleation activity, cell growth, and endocytosis of different p35/ARPC2 mutant alleles showed that the measured loss of the actin nucleation activity does not perfectly match the severity of cell growth and endocytosis defects. For example, the Surface III of ARPC2 is essential for endocytosis but not actin nucleation (Daugherty and Goode, 2008). It was reported that the dis2-1 mutation caused the accumulation of two mis-spliced transcripts that encode two proteins, dis2-1U and dis2-1S. The dis2-1S protein can interact weakly with ARPC4, indicating that the remaining dis2-1S protein in dis2-1 may still function in some aspects of cellular function (El-Assal ). In this study, vesicle trafficking and BFA treatments showed that vesicle trafficking and PIN2 cycling were not affected in the dis2-1 mutant, indicating that mutation in dis2-1 may have effects on actin cytoskeleton organization but not on endocytosis and vesicle trafficking (Fig. 5 and Supplementary Fig. S5).

Previous studies have reported that PIN3 and PIN7 have partially overlapping expression patterns in the columella cells and function redundantly in the gravitropic response. Roots of the pin3 mutant only show marginal defects in response to gravity stimulation. The gravitropic response defects of pin3 pin7 double-mutant seedlings were stronger than those of either single-mutant (Kleine-Vehn ). PIN2 single-mutants, however, display strong root gravitropic response defects (Rahman ). These two PINs, therefore, contribute differently to the gravitropic response. Moreover, it has been reported that light plays an essential role in PIN2 intracellular trafficking, probably by modulating the activity of the actin cytoskeleton (Laxmi ; Sassi ; Wan ). It was previously observed that dis1-1 showed more severe gravitropic defects in the dark than dis2-1 mutants (Reboulet ), suggesting the involvement of APR3/DIS1 in PIN cycling. However, it will be worth checking whether ARPC2A/DIS2 has impacts on PIN2 localization in the dark. As PIN2 cycling is not dependent on ARPC2A/DIS2, the defects in root gravitropic response of dis2-1 may be weak and, therefore, difficult to observe, in spite of the amyloplast sedimentation defects found in the dis2-1 mutants (Fig. 1D). Unlike its behavior in roots, PIN3 plays a major role in hypocotyl gravitropism. Gravity induces the translocation of PIN3 in endodermal cells and results in an on-site auxin asymmetry across hypocotyls. In contrast to the weak gravitropic phenotype in roots, pin3 hypocotyls display a pronounced defective gravitropic bending (Rakusová ). Interestingly, both the hypocotyls of dis1-1 and dis2-1 mutants displayed a similar gravitropic bending (Supplementary Fig. S1). Thus we propose that both ARP3/DIS1 and ARPC2A/DIS2 may contribute to the PIN3 translocation in shoot endodermal cells as well.

In mammals, the ARPC1 and ARPC5 subunits are each encoded by two genes. ARPC1B and ARPC5L are significantly better at promoting actin assembly than subunits with ARPC1A and ARPC5, revealing that distinctive ARP2/3 complexes (consisting of different subunit isoforms) may exert fundamentally divergent activities in higher eukaryotes (Abella ). In Arabidopsis, as well in rice, there are only one ARPC1 and one ARPC5, but ARPC2 is encoded by two genes, ARPC2A/DIS2 and ARPC2B (Le ; El-Assal ). Differing from ARPC2A/DIS2, ARPC2B has a candidate calmodulin-binding domain within its C-terminal extension (El-Assal ). Since gravistimulation can induce a transient change of cytoplasmic free calcium ion concentration, it is possible that ARPC2B is involved in root gravitropism through a calcium/calmodulin signaling pathway. One possibility is that ARPC2A/DIS2 and ARPC2B paralogs may have evolved distinct functions in regulating actin organization and endocytosis/vesicle trafficking. We then speculate that ARPC2A/DIS2 and ARPC2B may play complementary roles during gravity perception (amyloplast movement) and signal transduction (PIN protein recycling and calcium signaling), respectively.

In summary, our data indicate that the actin-related protein ARP3/DIS1 functions in root gravitropism, affecting amyloplast sedimentation in gravity perception, and mediating PIN protein cycling mainly through vesicle trafficking, thereby determining polar auxin transport in root gravity signal transduction. However, it remains to be determined if other components of the ARP2/3 complex as well the upstream WAVE/SCAR complex are involved in gravitropism and how PIN recycling is affected.

Supplementary data

Supplementary data are available at JXB online.

Figure S1. Hypocotyl gravitropic responses of Col, dis1-1, and dis2-1 plants.

Figure S2. Phenotype complementation of the dis1-1 mutant.

Figure S3. Asymmetric distribution of pDR5::GUS between the upper and lower sides of root tips after 90° reorientation.

Figure S4. Exogenous IAA or NAA did not recover defects of root gravitropic response in dis1-1 mutants.

Figure S5. PIN2 cycling in dis2-1 and wild-type plants.

Figure S6. Actin cytoskeleton organization in the root transition zone of wild-type, dis1-1, and dis2-1 plants.