![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 16, Issue 4, 1913-1927, April 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Plant Sciences, Tel Aviv University, Tel Aviv 69978, Israel;
Department of Ornamental Horticulture, Volcani Center, Bet Dagan 50250, Israel
Submitted July 7, 2004;
Revised January 28, 2005;
Accepted January 30, 2005
Monitoring Editor: Anne Ridley
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
). Rhos are subdivided into four subfamilies designated Rho, Racs, Cdc42, and a plant-specific group designated RACs or Rops (Hall, 1998; Winge et al., 2000; Yang, 2002). In animal cells, Rhos, Racs, and Cdc42 differentially regulate the actin cytoskeleton (Hall, 1998). Similarly, plant RACs are shown to regulate actin organization (Fu et al., 2001, 2002; Molendijk et al., 2001; Jones et al., 2002; Yang, 2002; Chen et al., 2003; Cheung et al., 2003).
Rhos seem to regulate exocytosis and endocytosis events such as pinocytosis, endocytosis of clathrin-coated pits, and localization of the multiprotein vesicle-tethering complex, the exocyst (Ridley et al., 1992; Lamaze et al., 1996; Di Cesare et al., 2000; Donaldson and Jackson, 2000; Malecz et al., 2000; Guo et al., 2001; Etienne-Manneville and Hall, 2002). Vesicle transport can be divided into five major steps: budding from a source membrane, targeting of the vesicle to specific regions, priming, docking at the target membrane, and fusion of the vesicles with the target membrane (Pfeffer, 1994, 2001; Jurgens and Geldner, 2002). In yeast, Cdc42 and Rho1 have been shown to regulate homotypic vesicle docking during vacuole formation in an actin-dependent manner (Eitzen et al., 2001. 2002; Muller et al., 2001; Eitzen, 2003), whereas Rho3 and Cdc42 regulate vesicle docking late in exocytosis during polar growth in budding yeast independent of their role in actin polarization (Adamo et al., 1999, 2001). In addition, it is well established that actin cytoskeleton function is crucial for endocytosis (Engqvist-Goldstein and Drubin, 2003).
Like other members of the Ras superfamily of small GTPases, Rhos exist in either GTP-bound active state or GDP-bound inactive state. Rhos have an intrinsic GTPase activity that is enhanced via interaction with GTPase-activating proteins. Activation of the Rhos occurs via interaction with GDP/GTP exchange factors (GEFs). Conserved dominant mutations abolishing the GTPase activity render Rhos constitutive active. Other conserved mutations preventing the GDP/GTP exchange are thought to cause irreversible interactions between the mutant Rhos and GEFs, converting the former dominant negative mutants (Hall, 1998; Winge et al., 2000; Yang, 2002).
The plant-specific Rho subfamily, designated either RACs or Rops, is subdivided into two major subgroups called type-I and type-II (Winge et al., 2000; Yang, 2002; Christensen et al., 2003). All type-I RACs are putatively prenylated, whereas the type-II are palmitoylated but not prenylated (Lavy et al., 2002). Cell shape in plants is regulated by polar growth in one direction or diffused growth in different directions, and RACs participate in the regulation of this process. Expression of constitutive active RACs induced reorganization of actin, implicated in changes in polar growth patterns, such as swelling of root hairs and pollen tubes and the misshaping of leaf epidermis cells concomitant with actin cytoskeleton organization (Fu et al., 2001, 2002; Molendijk et al., 2001; Jones et al., 2002; Yang, 2002; Chen et al., 2003; Cheung et al., 2003).
Arabidopsis spike1 mutants display loss of lobing of leaf epidermal cells and disorganized actin cytoskeleton (Qiu et al., 2002). The SPIKE1 protein shows homology to recently identified group of mammalian Rac-GEFs and was suggested to function as a RAC-GEF in plants (Molendijk et al., 2004). This provides a link between RAC activation, actin cytoskeleton organization, and cell shape. The Arabidopsis mutants dis1 and wrm have altered trichome shape and actin cytoskeleton organization (Le et al., 2003). The DIS1 and WRM genes encode components of the evolutionary conserved Arp2/3 complex, which regulates actin nucleation. The function of the Arp2/3 complex is regulated by the RAC and Cdc42-regulated WAVE complex, the components of which were recently identified in plants (Basu et al., 2004; Brembu et al., 2004; Zimmermann et al., 2004). Accordingly, WAVE mutants display similar phenotypes to dis1 and wrm (Hulskamp et al., 1994; Basu et al., 2004; Brembu et al., 2004; Zimmermann et al., 2004), and the Arabidopsis homologue of the WAVE complex SRA1 protein, PIROGI, interacted with Arabidopsis At-RAC4 in yeast two-hybrid assays (Basu et al., 2004). Collectively, these data indicate that plant RACs regulate actin cytoskeleton organization by evolutionarily conserved mechanisms. The induced changes in organization of the actin cytoskeleton are associated with altered cell shape.
In addition to changes in actin organization, it was found that RACs induce Ca2+ influx at the tip in germinating pollen tubes and growing root hair cells (Li et al., 1999; Molendijk et al., 2001). Furthermore, swollen root hair expressing an activated AtRop6 had multiple Ca2+ gradients (Molendijk et al., 2001). It was suggested, that activation of Ca2+ influx by RACs is actin independent and acts to promote exocytosis (Zheng and Yang, 2000; Fu et al., 2001). In agreement, in pollen tubes, RACs physically interacted with phosphatidylinositol monophosphate kinase (PtdIns P-K). Phosphatidylinositol 4,5-bisphosphate, the product of PtdIns P-K activity, showed similar intracellular localization to RAC (Kost et al., 1999).
Similar to mammalian cells, plant RACs activate NADPH oxidase to produce reactive oxygen species (Agrawal et al., 2003). The growth of roots and root hairs in Arabidopsis depends on the function of the NADPH oxidase complex protein RHD2. Compromised RHD2 function inhibited Ca2+ uptake into cells, thereby inhibiting cell growth (Foreman et al., 2003). This suggests that activation of NADPH oxidase by RACs is required for cell growth.
It is yet unknown whether expression of activated forms of RACs in plants influence membrane trafficking. Plant RACs studied to date act at the plasma membrane and may be required to regulate membrane-trafficking processes at this site during cell growth. Continuous cycling of some membrane proteins and sterols has recently been shown crucial for proper growth and development of plants (Geldner et al., 2001, 2003; Grebe et al., 2003). The existence of embryonic, together with mature, fully differentiated cells in the same organism through its life cycle, make plants a unique model system to study interaction of cellular processes with growth and differentiation. In the present research, novel evidence for the involvement of AtRAC10 in membrane trafficking in Arabidopsis is demonstrated.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
) by using the following primers for Atrac10CA: SYP500-GTGTGACTGTTGGTGATGTTGCTGTTGGTAAAACCTG and SYP501-CAGGTTTTACCAACAGCAACATCACCAACAGTCACAC, and following primers for Atrac10DN: SYP502-GATGGTGCTGTTGGTAAAAACTGTATGCTCATCTGC and SYP503-GCAGATGAGCATACAGTTTTTACCAACAGCACCATC. The mutated plasmids were designated pSY507 and pSY508, respectively. For expression in plants, pSY507 and pSY508 were digested with HindIII to isolate a cassette comprised of the 35S promoter of Cauliflower mosaic virus (CaMV), Atrac10CA or Atrac10DN fused to the 3' end of green fluorescent protein (GFP), and the nitric-oxide synthase (NOS) transcriptional terminator. This cassette was subcloned into pCAMBIA3300 (CAMBIA) to obtain pSY509 and pSY510.
Gene fusion between yellow fluorescent protein (YFP) and the mouse talin F-actin binding domain (YFP-Tn) was created as follows. First, a linker encoding five repetitions of Gly-Ala pair was inserted upstream to the talin gene sequence by two sequential polymerase chain reaction (PCR) steps. In the first step, pYS 514 (Kost et al., 1998, #8) was used as a template together with primers SYP517-GCTGGAGCTGGAGCTGGAATCCTAGGAGCTGGC and SYP518-GGATCCGAGCTCTTAGTGCTCGTCTCG. The product of the first reaction was used as template for a second PCR reaction with primers SYP521-GGATCCGAGCTCGGAGCTGGAGCTGGAGCTGGAGCTGGAGCTGGAATCCTA and SYP518. The PCR product of the second reaction was subcloned into pGEM-T vector (Promega, Madison, WI) to create pSY514. pSY514 was digested with SacI. The resulting fragment containing 5 x Gly-Ala linker and Tn fused in frame was subcloned into SacI digested pSY54, downstream of YFP. The resulting plasmid, pSY516, had 5 x Gly-Ala linker and mTalin fused in frame to YFP. For expression in plants, pSY516 was digested with HindIII to isolate a cassette comprised of the 35S promoter of CaMV, YFP-5 x Gly-Ala linker-Tn, and NOS transcriptional terminator. This cassette was subcloned into pCAMBIA3300 (CAMBIA) to obtain pSY524.
AtAAT1 (Gb# At4g21120) was cloned from an Arabidopsis flower cDNA library in two parts. Primers SYP95-GAGTGGATCCATGCTTCGGGTGGTGGTGAC and SYP806-CTTGAGCTAGCCATGGAGGCATC were used to amplify the 5' 1154 base pairs of AtAAT1 and primers SYP96-ACTCGGATCCTTAAGTTGCAGAACAAG and SYP805-GATGCCTCCATGGCTAGCTCAAG were used to amplify the 3' 631 base pairs of the gene. Both fragments were ligated into pGEM (Promega) to create pGEM-AAT1-5' and pGEMAAT1–3'. pGEMAAT1-5' was digested with BamHI NotI and pGEMAAT1–3' with XhoI and NotI. The isolated fragments were ligated at the NotI site by using T4 DNA ligase. The ligated fragment was used as template in a PCR reaction with primers SYP95 and SYP96 to amplify the full-length 1785 base pairs AtAAT1 gene. AtAAT1 was subsequently subcloned downstream of GFP gene pGFP-MRC (Rodriguez-Concepcion et al., 1999) to create plasmid pSY950. For expression in plant, pSY950 was digested with HindIII, and the resulting 35S::GFP-AtAAT1-NOS-3' cassette was cloned into pCambia2300 to create pSY952. All PCR products were sequenced to verify that no PCR and cloning-generated mutations were introduced into the sequence.
Wild-type Col-0 Arabidopsis plants were transformed using the floral dip method (Clough and Bent, 1998) with Agrobacterium tumerfaciens strain GV3101/pMp90. Analysis was carried out on several independent transgenic lines that displayed the same phenotypes.
Creation and Analysis of YFP-Tn GFP-Atrac10CA Plants
Several independent YFP-Tn transgenic lines obtained as described above were carefully screened to verify that the YFP-Tn transgene did not induce any new phenotypes. Three independent YFP-Tn tranformants were selected and introduced into the GFP-Atrac10CA background by crosses. Plants that cosegregated for the GFP and YFP markers were selected for further analysis.
Plant Growth Conditions
Seeds of wild-type and transgenic Arabidopsis plants were grown on soil under long-day conditions (16-h light/8-h dark cycle) as described previously (Lavy et al., 2002). To facilitate staining with N-[3-triethylammoniumpropyl]-4-[p-diethylaminophenylhexatrienyl] pyridinium dibromide (FM4-64) and drug treatments, plants were grown under high-humidity conditions by maintaining the plants under transparent cover. To analyze root hairs and young leaves, seeds were germinated on 0.8% agar plates composed of 1/2x Murashige & Skoog salt mixture (Invitrogen, Carlsbad, CA). After 48 h of vernalization at 4°C, the plates were transferred to a growth chamber and grown vertically under long-day conditions (16-h light/8-h dark cycle) at 23°C for 7 d. The light intensity was 100 µE m-2 s-1.
Pharmacological Agents and FM4-64 Staining
For all treatments, 0.5-cm2 pieces of fully expanded leaves (>3–4 cm in length), or whole seedling, were used. The plant material was submerged for 2 h at room temperature in distilled water supplemented with one of the following drugs diluted in dimethyl sulfoxide (DMSO): 90 µM brefeldin A (BFA) (Fluka, Buchs, Switzerland), 25 µM latrunculin B (Calbiochem, San Diego, CA), 39.4 µM cytochalasin D (Sigma-Aldrich, St. Louis, MO), and 4 µM FM4-64 dye (Molecular Probes). For control treatment, plant material was submerged in 0.25–0.5% DMSO solution containing 4 µM FM4-64 dye. Depolymerization of actin filaments was examined by submerging transgenic seedlings expressing fusion protein of GFP and the actin binding domain of mouse talin (GFP-Tn) (Kost et al., 1998) in 25 µM latrunculin-B, 39.4 µM cytochalasin-D, or 0.25% DMSO solution for 2 h at room temperature. For jasplakinolide treatments, seedlings were submerged in 1 µM jasplakinolide solution for up to 1 h before staining with FM4-64. Control plants were submerged in 0.1% methanol. Seedlings were kept in the jasplakinolide solution throughout the experiment.
Light, Fluorescence, and Confocal Microscopy
Wide-field fluorescence imaging was performed with an Axioplan-2 Imaging fluorescence microscope (Carl Zeiss, Jena, Germany) equipped with an Axio-Cam, cooled charge-coupled device camera by using either 20x dry or 63x water immersion objectives with numerical aperture (NA) values of 0.5 and 1.2, respectively. Confocal imaging was performed using either a Zeiss R510 or Leica TCS-SL confocal laser scanning microscope with 20x dry, 20x multi-immersion, 40x and 63x water objectives with NAs of 0.5, 0.7, and 1.2, respectively.
Endocytosis and Time-Lapse Microscopy. Time-lapse experiments were performed on 7-d-old seedlings that were grown on plates as described above. To minimize sample movement and maintain focus through the experiment, tissues were embedded in 1% low melting agarose supplemented with 7 µM FM4-64. Double stick tape was used as spacer, to avoid crashing of root hairs. Images were taken from young and growing root hairs at closest location to the root tips.
Visualization of Fluorescent Markers. GFP was visualized by excitation with an argon laser at 488 nm. Emission was detected with a 505- to 530-nm band-path filter (Zeiss R510) or spectral detector set between 505 and 530 nm (Leica TCS-SL). FM4-64 was visualized by excitation with an argon laser set to 514 nm. Emission was detected with either a 530- to 600-nm band-path filter (Zeiss R510) or spectral detector set between 530 and 560 nm (Leica TCS-SL). YFP was visualized by excitation with an argon laser at 514 nm. Emission was detected with spectral detector set between 525 and 570 nm (Leica TCS-SL). GFP/YFP double imaging was visualized using Leica TCS-SL confocal microscope. Excitation was with an argon laser at 488 nm. GFP emission was detected by setting the spectral detector between 500 and 525 nm. YFP emission was detected by setting the spectral detector between 530 and 570 nm. A channel dye separation process (Leica TCS) was performed to separate between the GFP and YFP signals. Alternatively, GFP and YFP signals were separated by a 
scan between 500 and 570 nm, at 5-nm increments, followed by spectral dye separation procedure (Leica TCS). Both procedures yielded similar results.
Actin Imaging. To obtain maximal signal-to-noise ratios and minimize bleaching, the argon laser was set to 
30% and the acusto optical tunable filter in all scans did not exceed 16%. All scans were carried at 512 x 512 pixel resolution with repeated scanning of two lines and two frames. Typically, 70–100 (pending on cell size) 0.4-µm-thick sections were taken. Maximum intensity projections and nearest point three-dimensional images of multiple confocal scans were used to view actin.
Image Analysis. Image analysis was performed with Zeiss AxioVision, Zeiss confocal laser microscopy (CLSM)-5, Leica LCS, Photoshop 7.0 (Adobe Systems, Mountain View, CA) and ImageJ. Quantifications of signal intensities were performed with Zeiss CLSM-5 or Leica LCS in the following manner. In 8 bit/pixel scans, the potential (signal intensity) on the photomultiplier tubes is converted by the computer to a value on a dimensionless 0–256 scale (28), were 0 represents no signal and 256 maximal intensity.
Plasmolysis Experiments
To induce plasmolysis, leaves or whole seedlings were submerged in 400 mM mannitol solution for several minutes and then mounted on microscope slides in the same solution.
Cell Fractionation and Protein Identification
Proteins from leaves of GFP-Atrac10CA and GFP-AtAAT1 transgenic plants were extracted in 10 mM HEPES-KOH, pH 7.5, 10% glycerol, 5 mM 
-mercaptoethanol, and plant protease inhibitor mix (Sigma-Aldrich) and incubated for 20 min on ice. To precipitate insoluble material, extracts were centrifuged at 14,000 x g for 15 min at 4°C. The resulting supernatants were collected and centrifuged again at 125,000 x g for 1.5 h. The insoluble pellet was incubated on ice for 20 min in 50 mM HEPES-KOH, pH 7.5, 50 mM NaCl, 1% Triton X-100, 5 mM -mercaptoethanol, and protease inhibitor mix. The samples were centrifuged at 14,000 x g and for 20 min and then at 125,000 x g for 1.5 h to remove the Triton X-100–insoluble material. The Triton X-100–insoluble pellet from GFP-Atrac10CA plants was resuspended in 50 mM HEPES-KOH, pH 7.5, 50 mM NaCl, 1% Triton X-100, 1.5% SDS, 5 mM -mercaptoethanol, and protease inhibitor mix. The SDS soluble, Triton X-100–insoluble samples were centrifuged at 14,000 x g for 20 min, and the resulting supernatants were further centrifuged at 125,000 x g for 1.5 h. GFP-Atrac10CA and GFP-AtAAT1 fusion proteins were detected by immunoblots decorated with mouse 
-GFP monoclonal antibodies (StressGen Biotechnologies, San Diego, CA), and goat -mouse horseradish peroxidase-conjugated secondary antibodies (Bio-Rad Blotting grade; Bio-Rad, Hercules, CA). Immunoblots were developed with EZ-ECL detection kit (Biological Industries, Bet Haemek, Israel)
Membrane Flotation Centrifugation
The experiments were performed as described previously (Ono and Freed, 1999) with the following modifications. Tissues were frozen with liquid N2 and ground to powder by using pestle and mortar. Protein were extracted by adding 1 ml of extraction buffer (50 mM HEPES-KOH, pH 7.5, 10 mM KCl, 5 mM EDTA, 5 mM EGTA, 10% sucrose, 1 mg/ml phenylmethylsulfonyl fluoride, and protease inhibitor cocktail; Roche Diagnostics, Indianapolis, IN) per 100 mg of tissue and incubated for 20 min on ice. The extract was filtered through 80-µm nylon mesh and precipitated at 3000 x g for 10 min. The pellet was discarded and the supernatant was collected for further analysis. Next, 250 µl of the supernatant was mixed with 1.25 ml of 85% sucrose in TE buffer (10 mM Tris, pH 8.0, and 1 mM EDTA, pH 8.0). This mixture was placed at the bottom of a 13-ml 10, 65, and 75% sucrose density step gradient as described previously (Ono and Freed, 1999). Centrifugal separation was performed with SW41 swinging bucket rotor (Beckman Coulter, Fullerton, CA) at 100,000 x g for 18 h. At the end of centrifugation, 1.25-ml samples were collected from the top of the gradient. To precipitate proteins, 0.5-ml samples were mixed with 125 µl of 50% trichloroacetic acid and 1 µg of bovine serum albumin and then incubated at -20°C for 10 min. Proteins were precipitated by centrifugation at 10,000 x g for 20 min and in turn the precipitated protein pellets were dried in SpeedVac. Dried protein pellets were dissolved in SDS-PAGE sample buffer and resolved on gels SDS-gels as described above. Protein immunoblots were performed as described above. All centrifugation steps were carried out at 4°C.
Tissue Sectioning and Staining
Tissues were fixed with formaldehyde acetic acid alcohol, dehydrated in graded ethanol series, and embedded in paraffin blocks by using standard procedures (Ruzin, 1999). Embedded tissues were sectioned with a rotary microtome (Shandon Scientific, Cheshire, England). To view cells, tissues were stained with safranin/fast-green according to standard procedures (Ruzin, 1999).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
Next, we examined weather the changes in leaf structure reflected altered cell morphogenesis. Cells of GFP-Atrac10CA transgenic plants had abnormal structures (Figure 2). Epidermal cells lost their lobing (Figure 2F). The palisade mesophyll cells became isomorphic rather than elongated along the adaxial-abaxial axis, and air spaces were smaller in the spongy mesophyll (Figure 2B). These changes made it difficult to discriminate between the palisade and spongy mesophyll in cross sections (Figure 2, A and B). Epidermal cells from plants expressing a wild-type GFP-AtRAC10 (GFP-AtRAC10), a dominant negative GFP-AtRAC10 (GFPAtrac10DN), or nonexpressing cells from GFP-Atrac10CA partially silenced plants, in which only a subset of the cells expressed the recombinant protein, were similar to cells of wild-type plants (Figure 2, E and G–L). The analysis of the partially silenced plants also showed that the loss of lobing in the epidermal cells was directly correlated to the level of GFP-Atrac10CA fluorescence in the cells (Figure 2, K and L). The data presented in Figure 2 indicated that the changes in cell morphology were induced by the expression of the constitutive active form of AtRAC10. To further study the function of AtRAC10, its subcellular localization was investigated.
|
Plasma Membrane Localization and Function of AtRAC10
Previously, we have shown that AtRAC10 is attached to the plasma membrane by virtue of its palmitoylation on two C-terminal cysteines (Lavy et al., 2002). When palmitoylation was inhibited, AtRAC10 accumulated in the nucleus due to a polybasic domain located proximal to its C-terminal end. Unlike other type-II RACs, a fraction of AtRAC10 was often located in the nucleus (Lavy et al., 2002).
When plant tissues are incubated in high-concentration osmoticum solutions, such as 30% glycerol or mannitol, water exits cells due to differences in water potential. In this process, designated plasmolysis, the plasma membrane is detached from the cell wall and the cytoplasm and vacuole shrink, but cell walls stay intact, maintaining overall cell structure. After plasmolysis, fluorescently labeled plasma membrane is detected as thin lines that are detached from the cell wall and the membrane becomes visible using Nomarsky differential interference contrast (DIC) (Figure 3K). After plasmolysis of either leaf epidermal cells or swollen root hairs, GFP-Atrac10CA remained attached to the plasma membrane, and no perinuclear and transvacuolar cytoplasmic strands that would have indicated vaulor membrane localization were detected (Figure 3, C, D, G, and H). For comparison, after expression of free GFP, the protein was detected as a thick, patchy line circumventing the cell, in cytoplasmic strands and nuclei (Figure S1, A–C). After plasmolysis, fluorescence accumulated in larger patches (Figure S1, D–F). This contrasts the membrane-bound GFP-Atrac10CA, which showed as a sharp fluorescing line before and after plasmolysis. Occasionally, elongated root hairs expressing GFP-Atrac10CA had swellings that were associated with accumulation of GFP-Atrac10CA protein (Figure 3, I and J).
|
To substantiate GFP-Atrac10CA localization at the plasma membrane and obtain further insight into its possible functions, total protein extracts were analyzed by membrane flotation centrifugation assays (Figure 4A). In these assays, the protein extract is placed at the bottom of a sucrose density gradient. After centrifugal separation, membranes float and the soluble fraction remains at the bottom of the gradient (Ono and Freed, 1999). In membrane flotation centrifugation assays, GFP-Atrac10CA was detected in the third and fourth fractions from the top of the gradient (Figure 4A). This localization in the gradient, which corresponds to the border between 10 and 65% sucrose, was expected for membrane-attached proteins (Ono and Freed, 1999). The soluble protein fraction, which in plants is characterized by a major 50-kDa band of the large subunit of ribulose biphosphate carboxylase oxygenase enzyme complex, remained at the bottom of the gradient (our unpublished data). Further experiments were carried out to verify whether AtRAC10 could be extracted out of the membranes by Triton X-100.
|
Protein extracts from GFP-Atrac10CA plants were fractionated into soluble and insoluble fractions. The insoluble fraction was fractionated into Triton X-100–soluble and –insoluble fractions. In turn, the Triton X-100–insoluble fraction was solubilized with SDS. The results (Figure 4B) show that the majority of the GFP-Atrac10CA protein existed in the insoluble fraction with most protein in the Triton X-100–insoluble fraction. Control experiments with a known membrane protein were used to verify that the inability to extract GFP-Atrac10CA by Triton X-100 did not result from the experimental system. AtAAT1 is a highly hydrophobic plasma membrane-localized amino acid transporter containing 14 putative transmembrane domains (Frommer et al., 1995). Protein extracts prepared from AtAAT1 transgenic plants were fractionated into soluble and insoluble fractions. In turn, the insoluble fraction was solubilized with Triton X-100. Although no GFP-AtAAT1 protein was detected in the soluble fraction, the fusion protein was extracted from the membrane by Triton X-100 (Figure 4C). The same expression system and buffers were used to express and extract GFP-Atrac10CA and GFP-AtAAT1, indicating that the inability to solubilize GFP-Atrac10CA with TritonX-100 was not a result from the experimental conditions. The partitioning of GFP-AtRAC10CA between Triton X-100–soluble and –insoluble membranes suggested that this protein might be fractionated into specific membrane microdomains (Simons and Toomre, 2000). More experiments were carried out to study the function of AtRAC10 at the plasma membrane.
AtRAC10 and Actin Organization
Changes in plant cell morphology induced by activated RACs were previously associated with actin reorganization (Fu et al., 2001, 2002; Molendijk et al., 2001; Jones et al., 2002; Yang, 2002; Chen et al., 2003; Cheung et al., 2003). To determine whether expression of GFP-Atrac10CA induced reorganization in actin cytoskeleton, transgenic Arabidopsis lines expressing the F-actin binding domain of mouse Tn (Kost et al., 1998) fused to YFP (YFP-Tn) were created. The YFP-Tn transgenic lines were carefully selected to ensure that the expression of the transgene did not induce new phenotypes.
Three independent YFP-Tn lines were crossed into GFP-Atrac10CA background to obtain double transgenic plants. In root hairs of Col-0 YFP-Tn plants, the actin bundles were organized parallel to long axis of the cell (Figure 5A). Fine mesh but no actin bundles were observed under the tip of growing root hairs. More actin bundles were observed in swollen root hairs of GFP-Atrac10CA YFP-Tn plants. These actin filaments were organized at different angles, often arranging as a band oriented at a right angle to the original elongation axis of the root hair (Figure 5, B and C). The arrangement of the actin bundles, circumventing the swollen root hair, became clear by a 120° angle tilt of a three-dimensional projection of the image (Figure 5C).
|
To view both YFP-Tn and GFP-Atrac10CA markers, samples were excited with argon laser at 488 nm, and the GFP and YFP signals were separated using a channel dye separation algorithm (Leica TCS) (Figure 5D). Although some of the YFP signal was lost during this procedure, the GFP signal (green label) was detected at the cell surface and the YFP signal (red label) was detected mainly in the cortical region. Similar results were obtained when the GFP and YFP signals were resolved using a spectral dye separation algorithm (Leica TCS), after a -scan with the spectral detector of Leica TCS-SL confocal microscope (our unpublished data).
In yeast and animal systems, Rho GTPases have been reported to affect endocytosis and exocytose by actin-dependent and -independent mechanisms. Plant RACs have recently been suggested to play critical roles in regulation of vesicle trafficking, but the mechanisms of their action are still unknown (Molendijk et al., 2004). Moreover, the effects of plant RACs on membrane dynamics have not been addressed.
Vesicle Cycling and GFP-Atrac10CA Function
As in axonal growth, root hair elongation requires the polar movement of vesicles that add the necessary substances and plasma membrane components by fusing at the cell's growing tip. RACs regulate membrane vesicle movement in two major ways: first, vesicle movement depends on F-actin, the dynamic of which is regulated by RACs. Second, RACs may regulate vesicle movement by using mechanisms not directly related with the cytoskeleton. To examine whether RACs influence vesicular transport, membranes were visualized with the lipophilic dye FM4-64, a common probe used for detection of membrane endocytosis (Parton et al., 2001; Santos et al., 2001; Heese-Peck et al., 2002; Jochum et al., 2002).
When WT seedlings were submerged in FM4-64 solution, the dye was internalized immediately (Figure 6, A and B). Movies prepared from a time-lapse series of the same cell, consisting of 60 scans taken every 10 s for 10 min, depicted the cytoplasmic streaming and the internalization of FM4-64 into the root hair (Movies 1-1 and 1-2).
|
The majority (80–90%) of root hairs of the WT RAC GFP-AtRAC10 transgenic plants were similar to root hairs of untransformed plants (Figure 6, C–E). In these root hairs, FM4-64 was internalized like in untransformed plants (Figure 6, D and E). Movies prepared from time-lapse series of similar root hairs showed the cytoplasmic streaming and endocytosis of FM4-64 (Movies 2-1 and 2-2). Approximately 10–20% of root hairs of GFP-AtRAC10 plants were partially swollen (Figure 6, F–H). In these root hairs, internalization of FM4-64 was barely detectable, and most of the dye stayed on the membrane (Figure 6, G and H). Movies prepared from a time-lapse series of 60 scans taken every 10 s for 10 min depicted the cytoplasmic streaming and steady but slow internalization of FM4-64 into the root hair (Movies 3-1 and 3-2).
FM4-64 was not internalized in swollen root hairs expressing GFP-Atrac10CA, remaining instead on the membrane (Figure 6, I–K). Most of the dye remained in the plasma membrane even after 90-min incubation at room temperature (Figure 6, L–N). Movies prepared from a time-lapse series of 60 scans taken every 10 s for 10 min showed that although cytoplasmic streaming continued, only very little FM4-64 was internalized, and most of the dye remained on the plasma membrane (Movies 4-1 and 4-2). This result suggested that the inhibition of FM4-64 internalization was not associated with cessation of cytoplasmic streaming.
To illustrate the distribution of GFP and FM4-64 markers in the WT Col-0 and transgenic plants, signal intensities were quantified (Figure 7). The plots represent distribution of signal intensity per micrometer in GFP and FM4-64 channels in the line defined in the inset of each graph. The insets in A, B, C, and D correspond to Figure 6, B, E, H, and N, respectively. The plots provide quantitative illustration to internalization of FM4-64 and colocalization of the two signals. Collectively, these data suggested a correlation between the localization of activated RAC at the plasma membrane, root hair swelling, and inhibition of vesicle internalization.
|
Recent results indicate that actin turnover is required for root hair and plant cell growth (Ketelaar et al., 2003, 2004). Atrac10CA induced actin reorganization (Figure 5), and RACs have been shown to inhibit ADF/cofillin that is required for actin turnover (Chen et al., 2002; Chen et al., 2003). To examine whether actin stabilization inhibited endocytosis of FM4-64, seedling were treated with the actin-stabilizing drug jasplakinolide (Figure 8). Jasplakinolide stabilizes actin by preventing filament disassembly both in vitro (Bubb et al., 1994) and in vivo (McGrath et al., 1998; Cramer, 1999) and induces actin polymerization in a concentration-dependent manner (Bubb et al., 1994; Lee et al., 1998; Shurety et al., 1998) by increasing actin nucleation (Bubb et al., 2000). Additionally, jasplakinolide has been shown to be highly cell permeate in both animal (Cramer, 1999) and plant cells (Samaj et al., 2002). Aggregation and increase in thickness of actin bundles were detected in jasplakinolide-treated YFT-Tn root hairs (Figure 8A), whereas no effect was detected in control plants incubated in 0.1% methanol. FM4-64 was internalized into jasplkinolide-treated root hairs and accumulated in intracellular bodies (Figure 8C), reflecting the perturbation to the actin cytoskeleton. Thus, unlike the effects of Atrac10CA, which inhibited internalization of FM4-64, jasplakinolide interfered with intracellular vesicle cycling but not with FM4-64 internalization from the plasma membrane.
|
Next we sought to determine whether Atrac10CA activity altered vesicle cycling directly rather then indirectly, by processes such as growth inhibition. To this end, cells of wild-type and Atrac10CA plants were visualized after disruption of the balance between endocytosis and exocytosis.
Inhibition of Exocytosis and AtRAC10-induced Vesicle Fusion
Unless exocytosis is inhibited, vesicle movement between the plasma membrane and endosomes is often invisible due to its speed. To inhibit exocytosis and render movement visible, cells were treated BFA. BFA, a specific inhibitor of ADP ribosylation factor (ARF) GEF, arrests vesicle trafficking at various points along the secretory pathway (Peyroche et al., 1999; Nebenfuhr et al., 2002; Ritzenthaler et al., 2002; Geldner et al., 2003). BFA treatments induce intracellular accumulation of different plasma membrane proteins, sterols, and of cell wall components (Geldner et al., 2001, 2003; Baluska et al., 2002; Friml et al., 2002; Grebe et al., 2002, 2003).
On treatment of wild-type roots with BFA, FM4-64–stained membranes accumulated in intracellular bodies, known as BFA bodies/compartments (Nebenfuhr et al., 2002; Ritzenthaler et al., 2002; Geldner et al., 2003; Grebe et al., 2003). Similarly, under the experimental condition used in this study, FM4-64–labeled membranes accumulated in BFA compartments after treatments of roots from wild-type seedlings with BFA (Figure 9A). Similar numbers of BFA bodies formed in root cells of BFA-treated GFP-AtRAC10 plants (Figure 9, C, E, and F). Much fewer BFA bodies were visible in BFA-treated GFP-Atrac10CA roots (Figure 9, C, H, and I). Moreover, the BFA bodies were only stained by FM4-64 and not with GFP (Figure 9, G and L). Quantitative analysis of BFA body numbers in 10 WT and GFP-AtRAC10 and eight GFP-Atrac10CA seedlings confirmed that their number was significantly lower (p 
0.001) in the GFP-Atrac10CA roots (Figure 9C). Quantification of signal intensities/pixel of the BFA bodies indicated that almost all the fluorescence came from the FM4-64 but not the GFP (Figure 9L). The maximum intensity projection stack image (Figure 9J) revealed that the labeling of BFA bodies with FM4-64 and not with GFP did not result from imaging of certain focal planes.
|
There were several interpretations to the results. GFP-Atrac10CA might have resided in membrane compartments that did not accumulate in BFA bodies. Alternatively, the ectopic expression of the activated RAC-disrupted vesicle accumulation in BFA bodies. The latter explanation was supported by data showing that BFA causes intracellular accumulation of different proteins and membrane components (Geldner et al., 2001, 2003; Friml et al., 2002; Grebe et al., 2002, 2003). More experiments were carried out to clarify this point.
When root hairs were examined a slightly different picture emerged (Figure 10). In wild-type root hairs, FM4-64 staining disappeared from the plasma membrane, and many intracellular bodies were visible after BFA application (Figure 10A). A similar picture emerged after BFA treatments of plants expressing the WT GFP-AtRAC10 where intracellular bodies stained with both FM4-64 and GFP occurred (Figure 10, C–E). In the GFP-Atrac10CA root hairs, FM4-64–stained bodies were visible, but in addition large vesicles measuring up to 20 µm in diameter, emanating from the plasma membrane, formed (Figure 10, G–I). In contrast to the BFA bodies in which only FM4-64 fluorescence was visible, the large vesicles were positive for FM4-64 and GFP signals. These, unique large vesicles, suggested that Atrac10CA disrupted vesicle cycling and inhibited the formation of BFA bodies.
|
As shown in Figure 2, like root hairs, leaf epidermal cells of GFP-Atrac10CA plants had distinct morphology. It was interesting to examine whether the expression of activated the RAC affected membrane dynamics in a similar manner in both root hairs and leaf epidermal cells (Figure 11).
|
On treatment of wild-type leaves with BFA, FM4-64 accumulated inside the cells in two ways: either cone-shaped structures formed in cellular growth lobes (Figure 11, A and B) or distinct BFA bodies were detected (Figure 11C). Inward movement of FM4-64–labeled membranes in the cellular growth lobes was revealed by time-lapse microscopy (Movie 5). Similar results were obtained after BFA treatments of leaves taken from transgenic plants expressing the WT GFP-AtRAC10 (our unpublished data). In some GFP-Atrac10CA cells, BFA induced formation of BFA bodies. Similar to roots, these BFA bodies mostly contained FM4-64 and no GFP signal, indicating that GFP-Atrac10CA–containing membranes did not cycle into the BFA bodies (Figure 11, E–G). In other GFP-Atrac10CA epidermal cells, large vesicles emanated from the plasma membrane, measuring up to 40 µm in diameter (Figure 11, I–K, and the lower cell E–G). Similar to the root hair cells, these large vesicles were stained by both FM4-64 and GFP, implying colocalization of the two dyes (Figure 11K, yellow label). Time-lapse microscopy revealed small vesicles budding from the plasma membrane and fusing into larger vesicles and fission of large vesicles from the membrane (Movie 6). The results from BFA treatments of either root hairs or leaf epidermal cells suggested that the activated RAC disrupted vesicle cycling, inhibiting their accumulation in BFA bodies, and instead promoting formation of large membrane vesicles.
Depolymerization of Actin and Changes in Membrane Dynamics
It has been shown that vesicle recycling between plasma membrane and endosomes could be inhibited by the acin-depolymerizing agents CD and Lat-B (Geldner et al., 2001; Grebe et al., 2003). It was therefore interesting to examine whether the effect of actin-depolymerizing agents would be similar to that of BFA and would provide an additional control for the BFA experiments.
To verify the effectiveness of Lat-B and CD in our system, transgenic plants expressing the actin marker GFP-Tn were treated with either drug. Treatment with either Lat-B or CD induced complete depolymerization of both actin bundles and fine meshwork (Figure S2).
Leaves taken from wild-type and GFP-Atrac10CA transgenic plants were treated with Lat-B or CD (Figure 12). In wild-type leaves, either Lat-B or CD treatment induced intracellular accumulation of FM4-64, which was confined to growth lobes (Figure 12, A and E), as in the case of BFA treatments (Figure 11, A and B). In GFP-Atrac10CA transgenic leaves, Lat-B treatment induced the formation of vesicles, emanating from the plasma membrane, reaching up to 40 µm in diameter (Figure 12, B–D). Similar effects were obtained when GFP-Atrac10CA leaves were treated with CD (Figure 12, F–H).
|
The results in Figures 11 and 12 provide evidence that disruption of membrane cycling by Atrac10CA could be revealed by both BFA treatment, which inhibits the exocytic pathway, and by the disassembly of actin, which may block vesicular movement.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
AtRAC10 and Membrane Cycling
BFA induces intracellular accumulation of different membrane proteins, such as PIN1 and PIN3 putative auxin transporters, plasma membrane H+-ATPase, the auxin influx carrier AUX1, and the cytokinesis-specific syntaxin KNOLLE, suggesting that protein accumulation induced by the drug is nonspecific (Geldner et al., 2001, 2003; Friml et al., 2002; Grebe et al., 2002). Polar auxin transport has been shown to depend on continuous cycling of PIN1 between the PM and endosomes requiring the activity of the ARF-GEF GNOM (Geldner et al., 2001, 2003). Either BFA or the actin-destabilizing agents CD or Lat-B arrested cycling of PIN1, inhibiting polar auxin transport. The absence of GFP-Atrac10CA from BFA bodies (Figures 9 and 11), the reduction in numbers of BFA bodies (Figure 9), and the colocalization of GFP-Atrac10CA with FM4-64–labeled membrane in plasma membrane emanating, large vesicles after treatments with either BFA, Lat-B, or CD (Figures 10, 11, 12) indicated that the activated RAC disrupted cycling of membrane vesicles.
Comparison of root hair structure and vesicle internalization between GFP-AtRAC10 and the activated GFP-Atrac10CA plants suggested a correlation between RAC activation, root hair swelling, and inhibition of endocytosis. Some GFP-At-RAC10 root hairs were more swollen than in WT Col-0 plants but not as swollen as in GFP-Atrac10CA plants. Concomitantly, vesicle internalization in partially swollen GFP-AtRAC10 root hairs was reduced compared with WT cell but more pronounced compared with GFP-Arac10CA cells (Figures 6 and 7 and Movies 1-1–4-2).
Ectopic expression of different RACs has been shown to alter auxin-induced gene expression. It has therefore been proposed that RACs regulate auxin response (Tao et al., 2002). Careful evaluation should be carried out to distinguish between the effects of activated RACs on auxin-dependent gene expression and transport and internalization of auxin.
BFA treatments of Atrac10CA plants had different effects in roots, root hairs, and leaf epidermal cells. In young roots, only BFA bodies were detected, whereas in both root hairs and leaf epidermal cells large vesicles formed (Figures 9, 10, 11, 12). It is possible that the penetration of BFA into root hairs and leaf epidermal cells was greater than into other root tissues. Root hairs are adapted to absorbed materials from the environment. Leaves used in this study were taken from plants that were grown under 100% humidity, either eliminating or substantially reducing the amount of the waxy cuticle on their surface and thereby facilitating uptake of hydrophobic compounds such as BFA. Furthermore, penetration of BFA through the stomata pores could have enhanced its uptake into leaves. Thus, the actual concentrations of BFA to which different cells were exposed might have been variable, causing phenotypic differences. Alternatively, there could be inherent differences between different cell types and tissues, causing specific phenotypes.
Membrane Localization of AtRAC10
The partitioning of AtRAC10 into Triton X-100–insoluble membrane may indicate its existence in sterol-rich membrane domains. Sterol trafficking in plants was shown to be actin dependent and BFA sensitive, and the PIN2 putative auxin transporter was colocalized with sterols after BFA treatments (Grebe et al., 2003). Consistently, PIN1 localization was altered in sterol methyltransferase1 mutants (Willemsen et al., 2003), and several sterol biosynthetic Arabidopsis mutants were shown to be disturbed in cell and tissue polarity (Betts and Moore, 2003). The ectopic expression of mutant forms of AtRAC10 in Arabidopsis affected plant morphology, leaf mesophyll cell shape (Figures 1 and 2), vascular differentiation, and auxin response (our unpublished data). Future analysis of the function of RACs in endocytic and exocytic events, their localization to specific membrane compartments, and their effect on membrane docking and fusion would be crucial for understanding the diversity of the function of the RACs. It is tempting to speculate that the localized and temporal activation of AtRAC10 and possibly other RACs leads to changes in membrane dynamics, promoting cell growth.
RACs, Membrane Cycling, and the Actin Cytoskeleton
In yeast, Cdc42 and Rho promote homotypic vesicle docking during vacuole formation in an actin-dependent manner. Although Cdc42 and Rho1 promoted vesicle docking under in vitro conditions, it was suspected that some actin bundles were not dissociated from membranes during their separation and could have mediated vesicle fusion (Eitzen et al., 2001, 2002; Muller et al., 2001; Eitzen, 2003). Furthermore, the function of actin cytoskeleton in endocytosis is well established (Engqvist-Goldstein and Drubin, 2003).
In a more parallel situation to the one shown here, it has been shown that Cdc42 and Rho3 promote vesicle docking late in exocytosis during polar growth in yeast (Adamo et al., 1999, 2001). Yeast Rho1 and Cdc42 interact with the exocyst complex subunits Sec3 (Novick and Guo, 2002). It was proposed that these interactions are required for placement of the exocyst complex and the establishment of polar growth (Adamo et al., 1999; Guo et al., 2001). AtRAC10 was localized at the plasma membrane (Figures 3 and 4) and likely disrupted vesicle cycling at this location. Genes encoding putative components of the exocyst were identified in the Arabidopsis and other plant genomes (Jurgens and Geldner, 2002; Elias et al., 2003), and preliminary results suggest that GFP-AtEXO70 fusion protein was localized close to the plasma membrane after its transient expression in protoplasts (Elias et al., 2003). Future examination of the interactions between plant RACs and exocyst complex or other components of the secretory pathway would be required to elucidate the molecular mechanisms by which RACs regulate vesicle trafficking.
Similar to other plant RACs (Fu et al., 2001, 2002; Molendijk et al., 2001; Jones et al., 2002; Yang, 2002; Chen et al., 2003; Cheung et al., 2003), activated AtRAC10 induced reorganization of actin (Figure 5). In tobacco pollen tubes, expression of activated NtRAC1 induced swelling and increased phosphorylation of actin-depoly
