Planta (2003) 217: 206–217 DOI 10.1007/s00425-003-0985-8

ORIGINAL ARTICLE

Elison B. Blancaflor Æ Guichuan Hou Æ Kent D. Chapman Elevated levels of N -lauroylethanolamine, an endogenous constituent of desiccated seeds, disrupt normal root development in Arabidopsis thaliana seedlings

Received: 21 October 2002 / Accepted: 27 December 2002 / Published online: 12 February 2003 Springer-Verlag 2003

Abstract N-Acylethanolamines (NAEs) are prevalent in ficking to and/or from the cell surface. The rapid me- desiccated seeds of various plant species, and their levels tabolism of NAEs during seed imbibition/germination decline substantially during seed imbibition and germi- may be a mechanism to remove this endogenous class of nation. Here, seeds of Arabidopsis thaliana (L.) Heynh. lipid mediators to allow for synchronized membrane re- were germinated in, and seedlings maintained on, organization associated with cell expansion. micromolar concentrations of N-lauroylethanolamine Electronic Supplementary Material Supplementary (NAE 12:0). NAE 12:0 inhibited root elongation, material is available for this article if you access the increased radial swelling of root tips, and reduced root article at http://dx.doi.org/10.1007/s00425-003-0985-8. hair numbers in a highly selective and concentration- A link in the frame on the left on that page takes you dependent manner. These effects were reversible when directly to the supplementary material. seedlings were transferred to NAE-free medium. Older seedlings (14 days old) acclimated to exogenous NAE by Keywords Arabidopsis Æ Cytoskeleton (microtubules) Æ increased formation of lateral roots, and generally, these Development (roots) Æ N-Acylethanolamine Æ lateral roots did not exhibit the severe symptoms Membrane dynamics Æ Vesicle trafficking observed in primary roots. Cells of NAE-treated primary roots were swollen and irregular in shape, and in many Abbreviations ER: Endoplasmic reticulum Æ DMSO: cases showed evidence, at the light- and electron-micro- dimethyl sulfoxide Æ GFP: green fluorescent protein scope levels, of improper cell wall formation. LA: lauric acid Æ NAE: N-acylethanolamine Æ NAPE: Microtubule arrangement was disrupted in severely N-acylphosphatidylethanolamine Æ PLD: phospholipase distorted cells close to the root tip, and endoplasmic D Æ numerical designation for acyl groups, number of reticulum (ER)-localized green fluorescent protein carbons in acyl chain: number of double bonds in acyl (mGFP5-ER) was more abundant, aggregated and dis- chain tributed differently in NAE-treated root cells, suggesting disruption of proper cell division, endomembrane orga- nization and vesicle trafficking. These results suggest that Introduction NAE 12:0 likely influences normal cell expansion in roots by interfering with intracellular membrane traf- N-Acylethanolamines (NAEs) are lipid mediators derived from the hydrolysis of the membrane phosphol- Electronic Supplementary Material Supplementary material is ipid, N-acylphosphatidylethanolamine (NAPE; Schmid available for this article if you access the article at http:// et al. 1996). In animal systems, this reaction constitutes dx.doi.org/10.1007/s00425-003-0985-8. A link in the frame on the part of the endocannabinoid signaling pathway which left on that page takes you directly to the supplementary material. regulates a variety of physiological processes, including K.D. Chapman (&) cell proliferation (De Petrocellis et al. 2000), immune cell Division of Biochemistry and Molecular Biology, signaling (Berdyshev 2000), neurotransmission (Wilson Department of Biological Sciences, and Nicoll 2002), and embryo development (Paria and University of North Texas, Denton, TX 76203, USA Dey 2000). In addition, this pathway appears to be E-mail: [email protected] Fax: +1-940-5654136 involved in cytoprotection in some tissues (Hansen et al. 2000). E.B. Blancaflor Æ G. Hou Plant Biology Division, NAEs have been identified and quantified in a variety The Samuel Roberts Noble Foundation, of plant tissues (Chapman et al.1998, 1999), and reports Ardmore, OK 73401, USA of biological activities of these lipids in plants are 207 beginning to appear (Chapman 2000). For example, of cells and tissue types in the elongated root is deter- increases in NAE 14:0 in elicited tobacco cell suspen- mined by the positional cues that are established during sions (Chapman et al. 1998) and elicited leaves of embryogenesis (Benfey and Scheres 2000). In the matu- tobacco plants (Tripathy et al. 1999) were sufficient to ration zone, the rapid directional expansion of cells activate phenylalanine-ammonia lyase (PAL) gene eventually ceases with specific cells in the epidermis dif- expression (Tripathy et al. 1999). More recently, NAEs ferentiating to form root hairs, which, in contrast to the were shown to selectively inhibit the activity of phosp- diffuse growth characteristic of cells in the elongation holipase D (PLD) a in vitro in a chain-length-dependent zone, expand primarily by tip growth. Underlying these manner, and application of NAEs to epidermal sections two different mechanisms of cell expansion are various of tobacco and Commelina communis abrogated abscisic dynamic processes that include ion fluxes (Hepler et al. acid (ABA)-induced stomatal closure (Austin-Brown 2001), cytoskeletal reorganization (Bibikova et al. 1999; and Chapman 2002), a process previously shown to Barlow and Baluska 2000), trafficking of membrane involve PLDa activity (Sang et al. 2001). vesicles (Hawes et al. 1999) and modifications in cell wall Desiccated seeds of a variety of plant species con- properties (Cosgrove 2000). tained NAEs with chain lengths of 12C to 18C (sum- The rapid depletion of NAEs in imbibing/germinat- marized in Chapman 2000). Total NAE content ranged ing seeds (Chapman et al. 1999; Shrestha et al. 2002) from about 400 ng g)1 FW in pea to 1,600 ng g)1 FW in points to a physiological role for this metabolic pathway cottonseed (Chapman et al. 1999). Moreover, in cotton, in seed germination and seedling growth. To test this pea and peanut seeds, total NAE content declined hypothesis, we germinated Arabidopsis thaliana seeds sharply after 4–8 h imbibition. Recently, two competing and maintained seedlings on sustained levels of NAE pathways were identified (and shown to be activated by above the physiological concentrations measured in imbibition) for the metabolism of these seed NAEs desiccated seeds. In this paper, we show that NAE 12:0, in vivo – a lipoxygenase (13-LOX)-mediated pathway but not NAE 16:0 or free fatty acid analogues of NAEs, for the peroxidation of polyunsaturated NAE species, induced a multitude of morphological and cellular and an amidohydrolase-mediated pathway for the defects in primary roots of Arabidopsis. Collectively, our hydrolysis of both saturated and unsaturated NAEs results point to a role for NAEs in root cell expansion/ (Shrestha et al. 2002). During the same period of seed elongation, possibly as a lipid mediator regulating imbibition and seed germination, the biosynthesis of membrane trafficking to and from the cell surface. NAPE increased substantially (Sandoval et al. 1995; Chapman and Sprinkle 1996) and the enzymatic machinery for NAPE biosynthesis was localized to Materials and methods membranes derived from the ER, Golgi, and plasma membranes (Chapman and Sriparameswaran 1997). It Plant material and treatment may be that the rapid metabolism of NAEs during seed imbibition and germination is an important requisite for Seeds of Arabidopsis thaliana L. Heynh. (ecotype: Columbia) were normal seedling growth. surface-sterilized in 95% ethanol and 20% bleach followed by extensive washing with sterile, deionized water prior to spreading Seed germination involves the breaking of dormancy onto sterile filter paper. Individual seeds were planted on 0.8% and the resumption of growth processes in the embryo, phyta-agar (Gibco) layered onto 62 mm·48 mm sterile coverslips. which is triggered partly by imbibition. Early seedling The composition of the medium was essentially as described in growth is sustained by metabolizable substrates stored in Legue et al. (1997). For the different treatments, a stock solution of 10 mM NAE 12:0 in 67% dimethyl sulfoxide (DMSO) was added the seed but subsequent establishment of the plant as a to the medium prior to autoclaving and layering onto coverslips to self-sustaining organism requires a coordinated pattern make final NAE 12:0 concentrations of 10, 20, 30, 40, 50 or of cell division and expansion that is regulated by a 100 lM. Arabidopsis media with an equivalent amount of DMSO, variety of external stimuli and endogenous factors (Esau NAE 16:0, lauric acid (LA) and palmitic acid were prepared for solvent and specificity controls. After planting, cover glasses were 1977). Although there have been several endogenous placed in 90-mm sterile plastic Petri dishes and placed in a growth compounds (e.g. plant hormones), and a variety of genes chamber with a 16 h light (134 lmol photons m)2 s)1), 8 h dark identified (Benfey and Scheres 2000) that regulate plant cycle at 22 C. The Petri dishes were positioned at a 60 angle from cell division and expansion, the complexity of these the horizontal so that the roots could grow along the bottom of the processes makes it likely that other yet unidentified coverslip for subsequent microscopic observation and image cap- ture. endogenous constituents regulate these important N-Acylethanolamine types were synthesized from ethanolamine cellular processes. and the corresponding acylchloride and purified as described pre- Roots are critical to plant establishment since a con- viously (Chapman et al. 1999). Purity was greater than 99% as stant supply of nutrients and water is necessary to sustain determined by GC–MS (Chapman et al. 1999). plant growth. The cell division patterns that are initiated in the meristematic region of the root (Schiefelbein et al. 1997) are modified by rapid directional expansion that Root hair counts, root length and radial expansion measurements leads to cells that are significantly longer than their width. Images of the primary root were captured using an SZX12 stere- This consequently forms the typical elongated shape of omicroscope (Olympus) equipped with a SPOT RT digital camera the root (Dolan et al. 1993), and the regular arrangement (Diagnostic Instruments, Sterling Heights, Mich., USA). For length 208 measurements, images of the primary root (approx. 20 per treat- [goat anti-rat IgG conjugated to fluorescein isothiocyanate (FITC); ment/experiment) were collected 3 days after planting and images Sigma]. After three additional washes in PME buffer, roots were collected every day thereafter for up to 8 days after planting. mounted in 20% Mowiol 4-88 (Calbiochem) containing the anti- The number of root hairs was obtained from seedling roots fade reagent phenylenediamine (0.1%) in phosphate-buffered saline 4 days after planting. Root hair counts were made from images of (PBS), pH 8.5. the root captured at a point beginning 0.5 mm from the base of the hypocotyl and approximately covering a length of 1 mm distal to this point. For measurements of radial expansion, longitudinal Confocal microscopy images of 10–12 roots/treatment were captured 9 days after planting and the width of the root at 100, 300, 500, 700 and 900 lm Seeds of Arabidopsis plants stably expressing ER-localized green from the tip was measured. All measurements were made from fluorescent protein (mGFP5-ER; Haseloff and Siemering 1998) digital images using the Metamorph 4.5 software (Universal were obtained from the Arabidopsis stock center (CS9139; Ohio Imaging Corp, West Chester, Pa., USA). All experiments were State University, Columbus, Ohio, USA) and planted in NAE repeated at least three times to ensure reproducibility of measure- 12:0-containing agar plates as described above. Both living and ments. Means, error, and statistical comparisons between data sets fixed root samples were imaged with a Bio-Rad 1024 ES con- (t-tests) were calculated with SigmaPlot v6.1 software. focal laser scanning microscope equipped with a 63·N.A. 1.2, For high-resolution growth measurements, Arabidopsis seeds water-immersion objective. For living roots expressing mGFP5- were initially germinated in NAE-free medium. After 3 days, ER, time-lapse sequences were generated using the time series seedlings with straight primary roots were selected and trans- function of the Lasersharp 3.2 software. Movie sequences were planted onto Arabidopsis medium containing 50 lM NAE 12:0 or generated from single optical sections at the mid-plane of se- medium with the equivalent solvent control (DMSO). Roots were lected root cells that were captured at 3-s intervals for 1.5 min. mounted on a Nikon TMD inverted microscope, which was posi- Individual stacks were assembled using the Metamorph 4.5 tioned on its side so the stage was vertical. Roots were visualized program. GFP and FITC were detected by exciting samples with using a 4 objective and images of vertically growing roots were · the 488-nm line of the Krypton–Argon laser and emission de- captured every 30 min for up to 7 h using a C2400-75i camera tected at 522 nm while propidium iodide (Sigma) was excited (Hamamatsu, Tokyo, Japan) running on Metamorph 4.5 image using the 488-nm line and emission detected at 590 nm. All acquisition software (Universal Imaging Corp). images were processed using Adobe Photoshop 5.0L.E. and printed on an Epson 900 ink-jet printer (Epson America).

Light and electron microscopy Supplementary material NAE-treated roots and corresponding controls were fixed in 2.5% glutaraldehyde in 25 mM potassium phosphate buffer (pH 7.0) Time-lapse movies of ER dynamics in living root cells in response for 2 h. The roots were washed three times with the same buffer to NAE 12:0 can be viewed online. followed by post-fixation in 1% OsO4 at 4 C overnight. After extensive washing in buffer, roots were dehydrated through a graded acetone series, and embedded individually in Spurr’s ep- oxy resin (Electron Microscopy Science, Ft. Washington, Pa., Results USA). For light microscopy, longitudinal sections were cut with an MT-X ultramicrotome (Research & Manufacturing Co., Tuc- son, Ariz., USA). Semi-thin (1 lm) median longitudinal sections The rapid depletion of NAEs in imbibing seeds suggests for light microscopy were placed onto glass slides and stained with an endogenous physiological role for this metabolic 0.5% (w/v) toluidine blue and counterstained with 0.05% (w/v) pathway in seed germination and seedling growth. One basic fuchsin. Sections were examined with a Nikon Microphot- FX microscope and images captured using a DC 200 digital straightforward approach to test this was to germinate camera (Leica Microsystems). For cell viability determination, seeds and maintain seedlings on levels of NAE above the roots of living seedlings were stained with 10 lg/ml propidium physiological concentrations measured in desiccated iodide for 10 min and imaged with a confocal microscope as seeds. Several morphological abnormalities were evident described below. For transmission electron microscopy (TEM), gold ultrathin in seedling roots when Arabidopsis thaliana (ecotype sections (90 nm) were cut using a diamond knife. Ultrathin sections Columbia) seedlings were germinated and grown in were mounted onto 200-mesh copper grids and stained with 1% the continual presence of N-lauroylethanolamine aqueous uranyl acetate and 0.1% lead citrate. Samples were then (NAE 12:0, Fig. 1). These effects were specific to viewed at 80 kV on a JEOL JEM-100CXII transmission electron microscope. NAE 12:0, since similar levels of lauric acid (12C free fatty acid) resulted in seedling phenotypes indistinguishable from untreated seedlings, or those treated with DMSO solvent alone. Generally, NAE 12:0 treatment resulted in Immunofluorescence labeling of microtubules a concentration-dependent reduction in primary root Imaging of microtubules by indirect immunofluorescence length, and at lower concentrations of NAE 12:0 (e.g. microscopy was as described in Harper et al. (1996). Briefly, roots 20 lM) there was an increased prevalence of lateral roots (NAE-treated and corresponding controls) were fixed in 3.7% compared to control seedlings or those treated with higher formaldehyde in PME (50 mM Pipes, 5 mM MgSO4,10mM EGTA) buffer (pH 6.9) and 5% DMSO (v/v). Roots were kept in levels of NAE 12:0 (e.g. compare Fig. 1a top, middle, and fixative for 2 h, washed extensively in PME buffer and secured onto lower panels). In the presence of 50 lM NAE 12:0, tips of glass coverslips using a thin film of agar. After a 30-min treatment primary roots (Fig. 1a, b) and lateral roots (Fig. 1c) ap- in a cocktail of 0.5% pectolyase Y23, 1% (w/v) bovine serum peared swollen after several days of growth. In addition, albumin (BSA; Sigma), 10% (v/v) glycerol and 0.2% (v/v) Tri- ton X-100, samples were incubated in a monoclonal rat anti-yeast roots grown in 50 lM NAE 12:0, were devoid of root a-tubulin (clone YOL1/34; Accurate Chemicals, Westbury, N.Y., hairs, either in young seedlings (3 days old, Fig. 1d) or USA) overnight followed by 2 h incubation in secondary antibody older seedlings (9 days old, Fig. 1c). 209

Fig. 1 a–d General morpho- logical effects of NAE 12:0 on roots of Arabidopsis thaliana seedlings. a Nine-day-old plants had severely reduced primary root growth at 50 lM NAE 12:0. Lower NAE concentrations (e.g. 20 lM) also inhibited primary root growth but the effects were less severe compared to higher NAE doses. Lower doses of exogenous NAE 12:0 appeared to promote the formation of lateral roots as evident from the longer lateral roots of plants grown in 20 lM NAE (arrows) compared to the free fatty acid controls. b, c Primary (b) and lateral roots (c) of 9-day-old plants displayed extensive swelling at the root tip region in response to 50 lM NAE 12:0 but not to 50 lM lauric acid. d In addition to the reduced root length and radial swelling, roots of 3-day-old Arabidopsis seedlings maintained on 50 lM NAE 12:0 were completely devoid of root hairs. LA Lauric acid. Bars = 1 mm (a), 100 lm (b–d)

Reduction in root elongation rate To determine how soon NAE exerted its inhibitory effect on root elongation, we transplanted 3-day-old A reduction in primary root length of seedlings could be Arabidopsis seedlings onto medium supplemented with due to a reduced growth rate of roots or to delayed seed 50 lM NAE 12:0 and captured images of the roots every germination and developmental delay of seedlings. In 30 min for up to 7 h using an inverted compound mi- the presence of NAE 12:0, there did not appear to be a croscope. Like roots of seedlings germinated and contin- substantial delay in seed germination, but rather the uously grown on NAE 12:0 (Fig. 2a), the growth of roots effect on root length of seedlings appeared to be due to a of transplanted 3-day-old seedlings was substantially in- reduced rate of growth (Fig. 2). Lengths of seedling hibited (Fig. 2c). The higher temporal and spatial reso- primary roots, growing in agar media supplemented lution afforded by our imaging system allowed us to detect with various concentrations of NAEs or free fatty acid, a statistically significant reduction in root elongation as were measured at daily intervals and plotted versus time early as 3 h after transplanting (Fig. 2c). Roots trans- (example shown in Fig. 2a). The rate of root growth in planted onto 50 lM NAE 12:0 became devoid of root mm/day was calculated by a linear regression of these hairs and, after longer periods of NAE 12:0 exposure, data and plotted versus the concentration of lipid added displayed increased radial swelling (Fig. 1; see below). (Fig. 2b). A pronounced dose-dependent reduction in It is important to note that the NAE treatment was root elongation rate was noted for seedlings grown in not generally toxic to seedlings, as growth rate, even at NAE 12:0, but not for seedlings grown in lauric acid or, 100 lM NAE 12:0, was not reduced to zero (Fig. 2a), N-palmitoylethanolamine (NAE 16:0), indicating that and as detailed below, roots reverted to normal growth the reduction in growth rate was quite specific for the rates when removed from NAE 12:0. Moreover, vital NAE 12:0. The half-maximal effective concentration staining with propidium iodide indicated that the cells (EC50) for NAE 12:0 was estimated to be 28 lM. were indeed viable (see Fig. 4c). 210

Fig. 2a–c Effects of exogenous NAEs on primary root length of Arabidopsis. a Kinetics of elongation of Arabidopsis roots grown in NAE 12:0, free fatty acid (FFA-lauric acid controls) or DMSO controls. Seeds of Arabidopsis were planted on phyta-agar medium supplemented with increasing concentrations of NAE 12:0 and root length measured daily for up to 8 days. Elongation of primary roots declined with increasing NAE 12:0 concentrations but root elongation of lauric acid-treated and DMSO controls was not Fig. 3a–c Radial swelling of Arabidopsis roots induced by exoge- affected. Data points are means ± SD from 22–25 or more roots/ nous NAE 12:0 8 days after planting. a Dose-dependent increase treatment and are representative of a typical experiment. Replicate in radial swelling of Arabidopsis primary roots in response to NAE experiments exhibited similar results with these NAE concentra- 12:0. At concentrations greater than 40 lM, increased radial tions, and additional experiments with various concentrations of expansion was evident at the region of the root spanning lauric acid (LA) or NAE 16:0 were not different from DMSO-only 100–900 lm from the tip. b, c At 100 (b) and 900 (c) lm from controls. b Dose–response curves of root growth inhibition by the tip, concentrations as low as 30 lM NAE 12:0 resulted in NAE 12:0, NAE 16:0 and lauric acid controls. The rate of root significant radial swelling compared to lauric acid (LA) and DMSO growth was calculated by a linear regression of the data presented controls. Data points are means and SD from 10 roots per in a (for each treatment), and plotted versus the concentration of treatment. For b, statistical comparison of treatment vs. control by lipid added. NAE 12:0 concentrations at or above 50 lM resulted t-test: * indicates P<0.02; ** indicates P<0.000001; *** indicates in maximum inhibition of elongation rate while NAE 16:0 and P<0.0000001. For c, statistical comparison of treatment vs. control lauric acid had minimal or no effect on root growth rate even at the by t-test: * indicates P<0.04; ** indicates P<0.002; *** indicates concentrations of NAE 12:0 that caused maximal growth P<0.000001 inhibition. c Kinetics of root elongation of 3-day Arabidopsis seedlings transplanted onto 50 lM NAE 12:0. Root length was Root tip swelling measured every 30 min for up to 7 h. A significant difference in root length was observed 3 h after transplanting. Statistical comparison of treatment vs. control by t-test: * indicates Root tip swelling in primary roots of 9-day-old seedlings P<0.04; ** indicates P<0.002 also was dependent upon NAE 12:0 concentration at 211 several distances between 100 and 900 lm from the root the organization of the endomembrane system, marked tip (Fig. 3a). The mean radial width of roots (intact roots by ER-localized GFP (mGFP5-ER; Haseloff and of growing seedlings) at both 900 lm (Fig. 3b) and Siemering 1998). Root cells in 4- to 5-day-old NAE- 100 lm (Fig. 3c) from the tip was significantly different treated seedlings appeared to accumulate and traffic from DMSO-only controls at concentrations above 30 lM NAE 12:0. The mean radial width of roots of seedlings treated with 50 lM lauric acid (12:0 free fatty acid) was not significantly different from DMSO controls and root tips did not swell in NAE 16:0, indicating that the effect of root tip swelling was specific for NAE 12:0.

Cellular organization and membrane trafficking defects

A closer examination of root tips at the cellular level indicated that the swollen root tips were comprised of disorganized cell files, and cells in the elongation zone were considerably swollen and irregular in shape (Fig. 4a). Some evidence of improperly formed cell walls was evident near the meristematic region in treated roots (Fig. 4b), and may have been in part responsible for the disorganization of cells in the swollen tip. However, other intracellular differences were noted in cells of the elongation zone. Projections of 30 optical sections of portions of propidium iodide-stained roots using con- focal microscopy showed clearly the defects in cell wall organization and cell shape in NAE-treated roots (Fig. 4c), and confirmed that these extremely distorted cells generally remained viable since cells excluded the dye. Also, there appeared to be a distinct difference in

c Fig. 4a–d Light-microscopic examination of fixed and living primary roots of Arabidopsis seedlings grown in 50 lM NAE 12:0. a Longitudinal sections of 8-day-old primary roots display extensive swelling and disorganized cell files in the tip region (arrow) in response to NAE 12:0 but not in DMSO or lauric acid controls. The swollen region is characterized by a population of large isodiametric cells and small distorted cells. b Higher magnification image in the region of the root close to the apex that displays extreme cell morphological distortions. Other defects induced by NAE 12:0 include the incomplete formation of cell plates (arrowhead) and the formation of oblique walls (arrow). c Viability of root cells was determined by staining with propidium iodide and imaging with a confocal microscope. Propidium iodide stains the nuclei of dead cells and labels the walls of living cells. Therefore, in addition to being a viability stain, it is a good indicator of tissue architecture in living roots. Projection of 30 optical sections through the root surface shows the typical elongated shape of the cells in control roots and the irregular cell shapes in NAE-treated roots. Despite the distorted appearance of NAE-treated roots, propidium iodide is excluded from the cells indicating that the cells are viable. d 4-day-old root of Arabidopsis stably expressing mGFP5, which is retained in the ER and thus allows visualization of ER dynamics in vivo. A single optical section of cells in the elongation zone of control roots exhibits fluorescence along the cell periphery (arrows). An optical section through the cell surface shows the presence of fusiform bodies (arrowhead; Hawes et al. 2001) in addition to the reticulate pattern typical of the ER network. In NAE 12:0-treated roots, cells are characterized by the aggregation of fusiform bodies (arrowhead) and smaller fluorescent bodies that extend beyond the cell periphery and appear to accumulate at the center of the cell (arrows). See supplementary material for time-lapse sequences of ER dynamics. Bar = 50 lm(a), 10 lm(b–d) 212 material through the central portion of the cell, unlike NAE-treated cells showed differences in cellular orga- the organization and movement of ER tubules observed nization, vesicle occurrence, and cell wall appearance in untreated root cells, which was confined to the cell (compare Fig. 5a, b with c–e). Cell wall stubs, which periphery (Fig. 4d, and supplementary material). are indicative of incomplete wall formation and wall Ultrastructural examination of root cortical cells invaginations, were prevalent at the electron-micro- using transmission electron microscopy supported scopic level (Fig. 5c). Walls of NAE-treated roots were inferences made from light microscopy, in that also wavy and uneven in thickness giving the cells their irregular and distorted shapes (Fig. 5d). These distorted Fig. 5a–e Transmission electron micrographs of ultrathin sections cells had large sections of membrane material pro- of cells from control and NAE-treated roots. Vacuolated cells from truding into the cytoplasm (Fig. 5d), which at times control roots are characterized by straight primary cell walls that were surrounded by multi-vesicular bodies (Fig. 5e). In are uniform in thickness (a) and are roughly perpendicular at the wall intersections (arrows in b). These patterns give rise to the addition, vesicular material appeared to accumulate in normal rectangular shape of cells in control roots. The distorted what appeared to be the middle lamella (Fig. 5e, inset); morphology of NAE-treated roots is manifested by cells with however, it is not entirely clear whether these incomplete walls (arrow in c), wall invaginations (double arrow- vesiculated regions are middle lamella or remnants of heads in c) and walls with uneven thickness (arrows in d). In some severely distorted cells, the plasma membrane appears to extend cytoplasm resulting from defects in wall formation. toward the vacuolar region (arrowhead in d). e Some of these Nevertheless, these results suggest that normal, efficient membrane protrusions (arrows) are surrounded by numerous membrane trafficking events, probably to and/or vesicles (arrowheads). Inset in e shows the accumulation of vesicles from the cell surface, were disrupted by NAE 12:0 (arrows) in unknown compartments that appear to be either middle lamella or remnants of cytoplasm resulting from irregular wall treatment, and that this was manifested in an overall formation. cw Cell wall, m mitochondrion, pd plasmodesmata, v disorganization of cell division and expansion, and vacuole. Bars = 0.5 lm ultimately a distortion of cell/root shape. 213

Microtubule organization in NAE-treated roots cortical microtubules was affected by NAE treatment, Arabidospis roots at 9 days of continuous exposure to Cellular elongation in roots has been shown to be NAE 12:0 were processed using immunofluorescence dependent on the cortical microtubule cytoskeleton (for microscopy, and microtubules in the regions that review, see Barlow and Baluska 2000). In fact, drugs that exhibited extreme cell distortion and swelling were either depolymerize or stabilize microtubules in roots examined using confocal microscopy. cause a significant reduction in root growth rate that In primary roots, cortical microtubules in cells of the eventually leads to radial expansion (Baskin et al. 1994). elongation zone are typically arranged transverse to the In order to determine whether the organization of longitudinal axis of the root. This orientation shifts to oblique or longitudinal arrays as the cells make their Fig. 6a–f Organization of cortical microtubules in root cells of transition into the maturation zone or when growth is 9-day-old Arabidopsis seedlings. a Microtubules in the epidermal prematurely inhibited (Blancaflor and Hasenstein 1995). cells throughout the elongation zone of control roots were After 9 days of continuous exposure to 50 lMNAE transverse to the longitudinal axis of the root. b A single optical section of cells in the distal elongation zone of control roots show 12:0, roots exhibited radial swelling. Instead of the cytoplasmic microtubules radiating from the surface of the nucleus typical cylindrical cell shape observed in the elongation (n) and connecting to the cell periphery. c Microtubules in the zone of control roots, cells of NAE-treated roots were meristematic region of control roots are associated with prepro- shorter and wider (Fig. 6; see also Figs. 4, 5). However, phase bands (arrowheads). Transversely oriented cortical microtu- despite the swollen cell phenotype, cortical microtubules bules are also characteristic of cells in interphase. d Bright-field and corresponding fluorescence image of root cells exposed to NAE in a majority of the cells remained transverse in both the 12:0. Roots grown in NAE 12:0 for 9 days were characterized by epidermis and cortex (compare Fig. 6a and d). In cells swollen cells in the elongation zone. However, microtubules in the from the distal elongation zone of control roots, radial epidermis retained their typical transverse orientation. e Bright- cytoplasmic microtubules were observed to connect the field and corresponding fluorescence image of roots grown in NAE 12:0 show abnormal cell division patterns within the meristem nucleus to the cell periphery (Fig. 6b). In addition, region (arrows) and microtubules in these cells were randomly microtubules in the meristem were associated with cell oriented. f A single optical section of an irregularly shaped cell division (e.g. preprophase bands) and transversely ori- close to the meristem showing an incomplete cell plate, which is ented cortical microtubules were observed in interphase characterized by intense tubulin labeling (arrowhead). Fragmented and disorganized microtubules were also observed in these cells cells (Fig. 6c). As described earlier, some roots grown (arrows). Bars = 100 lm (for bright-field images of whole roots), continuously in the presence of NAE 12:0 for 9 days 10 lm (for fluorescence images) displayed abnormal cell division patterns characterized 214 by oblique wall formation and the appearance of numerous irregularly shaped cells that varied in size (see Figs. 4, 5). Microtubules in these extremely disorganized cell files were oriented in random directions (Fig. 6e). Intense microtubule labeling also was associated with incompletely formed cell plates, and cytoplasmic microtubules in these misshapen cells appeared to be fragmented and disorganized (Fig. 6f).

Root hair formation

A reduction in the number of root hairs on Arabidopsis seedlings was noted with increasing concentration of NAE 12:0. Root hair number was unchanged from untreated control seedlings in the presence of DMSO alone (0.67% by volume, final), free lauric acid, or by growth in NAE 16:0, or free palmitic acid, indicating that like growth rate and tip swelling, the reduction in root hair numbers was quite specific to NAE 12:0 (Fig. 7a–c). The half-maximal effective concentration (EC50) for NAE 12:0 on root hair numbers was about 15 lM, somewhat less than that for either growth rate and tip swelling, perhaps reflecting the greater accessi- bility of NAE 12:0 to the root surface. Although not quantified, we noted that root hair length also was re- duced in the presence of 10 lM NAE compared with untreated or DMSO-only controls (Fig. 7a). Hence, at an NAE concentration below that at which cortical root cell expansion was perturbed, root hair cell elongation efficiency was reduced.

Reversibility

The profound physiological effects of NAE 12:0 on root growth and development were entirely reversible (Fig. 8). When 9-day-old Arabidopsis seedlings with short and swollen root tips (germinated and grown in 50 lM NAE 12:0) were transferred to NAE-free me- dium, primary root growth rate resumed to normal after a 1-day lag period (Fig. 8a). Root growth rate after Fig. 7a–c Effect of NAE 12:0 on root hair numbers from 4-day- recovery on NAE-free medium was estimated to be old Arabidopsis seedlings. Roots were grown on phyta-agar 5.6 mm/day, which was very similar to the elongation medium supplemented with increasing concentrations of NAE rate of untreated primary roots (compare to Fig. 2). 12:0 or 16:0. a Root hair numbers were counted by capturing Moreover, normal patterning and development of root images of the root covering a length of about 1 mm beginning at 0.5 mm from the base of the hypocotyls. Numerous root hairs hairs recovered on NAE-free medium, and root hairs within this region were observed in primary roots grown in NAE even emerged in the region that was previously barren. 16:0, lauric acid (LA) or DMSO controls but a reduction in the The ‘‘newly-formed’’ tip of the root was no longer length and overall numbers of root hairs were characteristic of swollen, and was indistinguishable from a root tip that roots grown at increasing concentrations of NAE 12:0. Bar = 100 lm. b The effect of NAE 12:0 on root hair numbers was had never been exposed to NAE 12:0 (Fig. 8b). dose-dependent. NAE 12:0 at 20 lM or higher significantly reduced root hair number while NAE 12:0 concentrations greater or equal to 40 lM resulted in the almost complete Discussion absence of root hairs. c Increasing concentrations of NAE 16:0 had minimal effects on root hair numbers even at concentrations of NAE 12:0 that abolished root hair formation. Root hairs N-Acylethanolamine (NAE) metabolism is part of the were counted on 10–20 roots per treatment and reported as the endocannabinoid signaling pathway in animal tissues, average number ± SE 215

a difference in the sensitivity of the molecular target(s) toward NAE in the root tissue, as is the case for PLDa, which is differentially inhibited by these two compounds (Austin-Brown and Chapman 2002); or, alternatively, there may be a different efficiency in the metabolism of these two NAEs by root cells. Perhaps several factors contribute to the selective effects of NAE 12:0. Never- theless, due to the similar structural nature of NAE 12:0 and NAE 16:0, it is unlikely that the difference in physiological effectiveness could be explained by solu- bility or tissue penetration. Because NAE 12:0 was found in all angiosperm seeds examined (Chapman et al. 1999), we conclude that its efficient catabolism during seed imbibition and germination is important for normal seedling growth and development. Alternatively, since NAEs are not completely depleted during imbibition (Shresta et al. 2002), it is possible that NAE 12:0 or other NAE types could exert their major effect during the early stages of seed germination. However, addi- tional work will be needed to clarify the role of NAEs in seed germination and post-germinative seedling growth. The pleiotropic effects of NAE 12:0 on seedling roots are reminiscent of roots treated with chemical inhibitors that disrupt the cytoskeleton (Baskin et al. 1994) or alter Fig. 8a, b The effect of NAE 12:0 on Arabidopsis primary root vesicle trafficking (Baskin and Bivens 1995). Most sim- growth is reversible. a Kinetics of primary root elongation of ilar was the effect of NAE 12:0 at concentrations at or Arabidopsis seedlings grown in 50 lM NAE 12:0 for 9 days and transferred to NAE-free medium. After a short lag of about 1 day, above 30 lM, which caused a significant increase in the growth rate of primary roots reverted back to the control rates. radial expansion of the root (Fig. 3). Although immu- b Primary roots of Arabidopsis growing for 9 days in 50 lM NAE nolabeling did not reveal any appreciable changes in 12:0 and 1 day after transfer to NAE-free medium. Roots before overall microtubule organization in NAE-treated roots transfer to NAE-free medium have swollen apices and no root hairs. After transfer to NAE-free medium, roots rapidly resume during stages of strong growth inhibition (data not normal growth as manifested by the reduction in tip swelling and shown) or in the swollen cells of the elongation zone the emergence of root hairs (arrows). Bar = 100 lm (Fig. 6), results from electron-microscopic analysis and plants expressing mGFP5-ER revealed abnormal ER and the tight regulation of NAE levels in vertebrates is dynamics, distorted and incomplete cell walls, and the important to proper functioning of a variety of physio- accumulation/aggregation of vesicles (Figs. 4, 5). Fur- logical systems (Di Marzo 1998). The conservation of thermore, immunofluorescence microscopy revealed the NAE occurrence and its metabolism in plant tissues occurrence of disorganized and fragmented microtu- (Chapman 2000) suggests that important physiological bules in severely disrupted cell files close to the root tip processes in plants also may be regulated at the cellular with prolonged exposure to NAE 12:0, suggesting that level by NAEs. Seed germination and seedling growth, defects in cell division occurred (Fig. 6e, f). marked by active metabolism of endogenous NAEs The apparent abnormalities in cell division (i.e. in- (Chapman et al. 1999; Shrestha et al. 2002), offer an complete/irregular walls) and the inhibition of root hair excellent system in which to examine this hypothesis. initiation, point to the endomembrane system and vesi- A multitude of profound physiological differences cle trafficking as potential targets of NAEs. In plants, were observed in roots of Arabidopsis seedlings that were cell plate formation begins with the assembly of germinated and grown in micromolar concentrations of the phragmoplast, a structure consisting of ER, Golgi- NAE 12:0 (Fig. 1). Although the mechanism(s) of ac- derived vesicles, and cytoskeletal elements. The complete tion is (are) not entirely clear at this point, all of the development of the cell plate requires the continuous effects were specific for NAE 12:0, since lauric acid, a delivery of vesicles carrying various wall precursors lipophilic 12C saturated free fatty acid, had no effect on (Otegui and Staehelin 2000). Recent studies on Arabid- seedling root morphology (Figs. 1, 2, 3, 7). Interestingly, opsis mutants defective in embryo development have led NAE 16:0, sometimes more abundant in seeds than to the identification of several molecular components of NAE 12:0, also did not have any effects on seedling the plant cytokinetic apparatus including the KNOLLE, morphology through the concentration ranges tested (up KEULE and AtSNAP33 proteins (Assaad et al. 2001). to 100 lM, Figs. 2, 7). This indicates a marked speci- The KNOLLE protein is a member of the syntaxin ficity for NAE 12:0, since NAE 16:0 is an acylethan- family of integral membrane proteins involved in vesicle olamide-like NAE 12:0 that differs only in the length of docking and fusion, and mutants in the KNOLLE gene the acyl chain. This difference in potency may be due to result in cytokinetic defects such as the accumulation 216 of unfused vesicles and incomplete wall formation membranes is well known, but has most often been (Waizenegger et al. 2000). KUELE and AtSNAP33 interpreted to indicate a degradative or signaling interact with the KNOLLE protein and mutations in function (Chapman 1998). In retrospect, many func- both genes also lead to extreme defects in cytokinesis tional associations of PLD with membranes could (Assaad et al. 2001). Interestingly, while the KNOLLE indicate a role in membrane trafficking (Munnik and mutants have long root hairs, KUELE mutants are de- Musgrave 2001). NAEs recently were shown to be void of, or display stunted root hairs. This indicates that potent regulators of PLDa activity in vitro and in vivo KNOLLE is involved in vesicle trafficking events specific (Austin-Brown and Chapman 2002), although NAEs to cytokinesis while KUELE may also be required for did not appreciably affect activities of PLDb and vesicle trafficking associated with tip growth (Assaad et PLDc. Future studies will be aimed at assessing a role al. 2001). The similarities in some of the cellular defects for PLD and NAE in membrane trafficking events between Arabidopsis seedling roots exposed to NAE and associated with cell elongation and expansion. cytokinesis-defective mutants suggests that the vesicle Another possible mode of action for NAE is through trafficking machinery of root cells could be a molecular binding to a cell surface receptor, such as the situation target of NAEs. that occurs in several vertebrate cell types (Reggio and Although disorganized microtubules in severely dis- Traore 2000). At least two types of cannabinoid (CB) torted cells were observed close to the root tip, cortical receptor that are activated by NAEs have been identified microtubules in most of the swollen cells in the elonga- in animal systems (reviewed in Pertwee 1997). Both types tion zone of NAE-treated roots remained transverse of NAE receptor are G-protein-coupled (belonging to (Fig. 6d). Despite the lack of microtubule reorganiza- the rhodopsin superfamily of transmembrane receptors) tion in these cells, the possibility that NAE-treatment and activate different intracellular signal-transduction causes the selective stabilization of microtubules in cells cascades depending upon the cell type (Howlett and that have already started longitudinal expansion cannot Mukhopadhyay 2000). Although no obvious orthologs be ruled out. For example, taxol, a drug that stabilizes in the Arabidopsis genome are readily identifiable by microtubules, also induces radial swelling in roots sequence similarity with animal CB receptors, in sepa- without altering the general orientation of cortical rate work in tobacco cell suspensions and microsomes of microtubules (Baskin et al. 1994). Aluminum treatment tobacco leaves, we have identified through radioligand- has also been shown to cause radial swelling in roots but binding assays, a membrane protein(s) which binds microtubules in the swollen cells of the epidermis and specifically and with high affinity to NAEs (data not outer cortex are stabilized while keeping their general shown). It is possible that an NAE 12:0 binding protein transverse orientation intact (Blancaflor et al. 1998). is present in roots and through activation (or inhibition) Therefore, similar to previous studies, a reorganization of cellular signaling machinery, mediates the effects of of cortical microtubules may not be correlated with NAE 12:0 on root cells. NAE-induced cell swelling/growth inhibition, and this Collectively, the results presented here continue to agrees with recent results showing that the growth status point to a role for NAEs in normal seed germination of the cell and cortical microtubule reorientation are not and seedling growth, and emphasize the importance of tightly correlated (Granger and Cyr 2001). Although our understanding the metabolic regulation of the levels of study did not include an examination of the actin cyto- this endogenous class of lipid mediators. Moreover, skeleton in NAE-treated roots, the dependence of these data are consistent with the concept of a generally organelle and vesicle dynamics on both the microtubule conserved NAE signaling pathway in both plants and and actin cytoskeleton (Steinborn et al. 2002) makes it animals, which when perturbed results in profound and likely that defects in cytoskeletal-mediated vesicle traf- widespread physiological consequences. ficking in both dividing and actively elongating cells could account for some of the abnormalities we Acknowledgments Financial support for this work was provided in observed in Arabidopsis seedling roots exposed to NAE. part by USDA-NRICGP agreement number 99-35304-8002 to Additional studies that will involve examining the actin K.D.C., NASA grant number NAG 2-1518 to E.B.B. and the Samuel Roberts Noble Foundation. We thank the Arabidopsis component of the cytoskeleton, as well as studies with stock center (Ohio State University) for the root expressing GFP-cytoskeletal/endomembrane reporter constructs mGFP5-ER seeds and Dr. Richard A. Dixon, Samuel R. Noble should yield important insights into how the cytoskel- Foundation, for critical reading of the manuscript. We also thank eton is modified or regulated by NAEs. Mr. David Garret, University of North Texas, for assistance with One possible target of NAE in root cells is phosp- transmission electron microscopy. holipase D (PLD). Recent evidence suggests that PLD is involved in the interaction of microtubules and membranes (Gardiner et al. 2001), and Munnik and References Musgrave (2001) suggested that plant PLD proteins may play a more generalized role in microtubule-me- Assaad FF, Huet Y, Mayer U, Jurgens G (2001) The cytokinesis gene KEULE encodes a Sec1 protein that binds the syntaxin diated membrane trafficking, similar to functions KNOLLE. J Cell Biol 152:531–543 ascribed to PLD proteins in yeast and animal systems Austin-Brown S, Chapman KD (2002) Inhibition of phospholi- (Cockcroft 2001). An association of PLD with plant pase Da by N-acylethanolamines. Plant Physiol 129:1892–1898 217

Barlow PW, Baluska F (2000) Cytoskeletal perspectives on root Harper JDI, Holdaway NJ, Brecknock SL, Busby CH, Overall RL growth and morphogenesis. Annu Rev Plant Physiol Plant Mol (1996) A simple and rapid technique for the immunofluores- Biol 51:289–322 cence confocal microscopy of intact Arabidopsis root tips. Baskin TI, Bivens NJ (1995) Stimulation of radial expansion in Cytobios 87:71–78 Arabidopsis roots by inhibitors of actomyosin and vesicle Haseloff J, Siemering KR (1998) The uses of GFP in plants. In: secretion but not by various inhibitors of metabolism. Planta Chalfie M, Kain S (eds) Green fluorescent protein: strategies, 197:514–521 applications and protocols. Wiley, New York, pp 191–220 Baskin TI, Wilson JE, Cork A, Williamson RE (1994) Morphology Hawes CR, Brandizzi F, Andreeva AV (1999) Endomembranes and microtubule organization in Arabidopsis roots exposed to and vesicle trafficking. Trends Plant Sci 2:454–461 oryzalin or taxol. Plant Cell Physiol 35:935–942 Hawes CR, Saint-Jore C, Martin B, Zheng H-Q (2001) ER con- Benfey PN, Scheres B (2000) Root development. Curr Biol firmed as the location of mystery organelles in Arabidopsis 16:R813–R815 plants expressing GFP. Trends Plant Sci 6:245–246 Berdyshev EV (2000) Cannabinoid receptors and the regulation of Hepler PK, Vidali L, Cheung AY (2001) Polarized cell growth in immune response. Chem Phys Lipids 108:169–190 higher plants. Annu Rev Cell Dev Biol 17:159–187 Bibikova TN, Blancaflor EB, Gilroy S (1999) Microtubules regu- Howlett AC, Mukhopadhyay S (2000) Cellular signal transduction late tip growth and orientation in root hairs of Arabidospis by anandamide and 2-arachidonylglycerol. Chem Phys Lipids thaliana. Plant J 17:657–665 108:53–70 Blancaflor EB, Hasenstein KH (1995) Growth and microtubule Legue V, Blancaflor E, Wymer C, Perbal G, Fantin D, Gilroy S orientation of maize roots subjected to osmotic stress. Int (1997) Cytosolic free calcium in Arabidospis roots changes J Plant Sci 156:794–802 in response to touch but not gravity. Plant Physiol 114: Blancaflor EB, Jones DL, Gilroy S (1998) Alterations in the cyto- 789–800 skeleton accompany aluminum induced growth inhibition and Munnik T, Musgrave A (2001) Phospholipid signaling in plants: morphological changes in primary roots of Zea mays L. Plant holding on to phospholipase D. Science’s STKE http:// Physiol 118:159–172 stke.sciencemag.org/cgi/content/full/sigtrans;2001/111/pe42 Chapman KD (1998) Phospholipase activity during plant growth Otegui M, Staehelin LA (2000) Cytokinesis in flowering plants: and development and in response to environmental stress. more than one way to divide. Curr Opin Plant Biol 3:493–502 Trends Plant Sci 3: 419–426 Paria BC, Dey SK (2000) Ligand-receptor signaling with Chapman KD (2000) Emerging physiological roles for N-acylphos- endocannabinoids in preimplantation embryo development and phatidylethanolamine metabolism in plants: signal transduction implantation. Chem Phys Lipids 108:211–220 and membrane protection. Chem Phys Lipids 108:221–230 Pertwee RG (1997) Pharmacology of cannabinoid CB1 and CB2 Chapman KD, Sprinkle WB (1996) Development, tissue-specific receptors. Pharmacol Ther 74:129–180 and environmental factors regulate the biosynthesis of N-acyl- Reggio PH, Traore H (2000) Conformational requirements for phosphatidylethanolamine in cotton (Gossypium hirsutum L.). endocannabinoid interaction with the cannabinoid receptors, J Plant Physiol 149:277–284 the anandamide transporter, and fatty acid amidohydrolase. Chapman KD, Sriparameswaran A (1997) Intracellular localiza- Chem Phys Lipids 108:15–36 tion of N-acylphosphatidylethanolamine synthesis in cotyle- Sandoval JA, Huang Z-H, Garret DC, Gage DA, Chapman KD dons of cotton seedlings. Plant Cell Physiol 38:1359–1367 (1995) N-Acylphosphatidylethanolamine in dry and imbibing Chapman KD, Tripathy S, Venables B, Desouza A (1998) N-acy- cotton (Gossypium hirsutum L.) seeds: amounts, molecular lethanolamines: formation and molecular composition of a new species and enzymatic synthesis. Plant Physiol 109:269–275 class of plant lipids. Plant Physiol 116:1163–1168 Sang Y, Zheng S, Li W, Huang B, Wang X (2001) Regulation of Chapman KD, Venables B, Blair R Jr, Bettinger C (1999) plant water loss by manipulating the expression of phospholi- N-acylphosphatidylethanolamines in seeds: quantification of pase Da. Plant J 28:1–11 molecular species and their degradation upon imbibition. Plant Schiefelbein JW, Masucci JD, Wang H (1997) Building a root: the Physiol 120:1157–1164 control of patterning and morphogenesis during root develop- Cockcroft S (2001) Signaling roles of mammalian phospholipase ment. Plant Cell 9:1089–1098 D1 and D2. Cell Mol Life Sci 58:1674–1687 Schmid HH, Schmid PC, Natarajan V (1996) The N-acylation- Cosgrove DJ (2000) Expansive growth of plant cells. Plant Physiol phosphodiesterase pathway and cell signaling. Chem Phys Biochem 38:109–124 Lipids 80:133–142 De Petrocellis L, Melck D, Bisogno T, Di Marzo V (2000) En- Shrestha R, Noordimeer M, van der Stelt M, Veldink G, Chapman docannabinoids and fatty acid amides in cancer inflammation KD (2002) N-Acylethanolamines are metabolized by lipoxy- and related disorders. Chem Phys Lipids 108:191–209 genase and amidohydrolase in two competing pathways during Di Marzo V (1998) Endocannabinoids and other fatty acid deriva- cotton (Gossypium hirsutum L) seed imbibition. Plant Physiol tives with cannabimimetic properties: biochemistry and possible 130:391–401 physiological relevance. Biochim Biophys Acta 1392:153–175 Steinborn K, Maulbetsch C, Priester B, Trautmann S, Pacher T, Dolan L, Janmaat K, Willemsen V, Linstead P, Poethig S, Roberts Geiges B, Kuttner F, Lepiniec L, Stierhof Y-D, Scharwz H, K, Scheres B (1993) Cellular organization of the Arabidopsis Jurgens G, Mayer U (2002) The Arabidopsis PILZ group genes thaliana root. Development 119:71–84 encode tubulin-folding cofactor orthologs required for cell Esau K (1977) Anatomy of seed plants. Wiley, New York division but not cell growth. Genes Dev 16:959–971 Gardiner JC, Harper JDI, Weerakoon ND, Collings DA, Ritchie S, Tripathy S, Venables BJ, Chapman KD (1999) N-acylethanolam- Gilroy S, Cyr RJ, Marc J (2001) A 90-kD phospholipase D ines in signal transduction of elicitor perception: attenuation of from tobacco binds to microtubules and the plasma membrane. alkalinization response and activation of defense gene expres- Plant Cell 13:2143–2158 sion. Plant Physiol 121:1299–1308 Granger CL, Cyr RJ (2001) Spatiotemporal relationships between Waizenegger I, Lukowitz W, Assaad F, Schwarz H, Ju€rgens G, growth and microtubule orientation as revealed in living root Mayer U (2000) The Arabidopsis KNOLLE and KEULE genes cells of Arabidopsis thaliana transformed with the green- interact to promote vesicle fusion during cytokinesis. Curr Biol fluorescent protein gene construct GFP-MBD. Protoplasma 10:1371–1374 216:201–214 Wilson RI, Nicoll RA (2002) Endocannabinoid signaling in the Hansen HS, Moesgaard B, Hansen HH, Petersen G (2000) brain. Science 296:678–682 N-Acylethanolamines and precursor phospholipids – relation to cell injury. Chem Phys Lipids 108:135–150