Plant Cell Polarity: The Ins-and-Outs of Sterol Transport

Current Biology, Vol. 13, R781–R783, September 30, 2003, ©2003 Elsevier Science Ltd. All rights reserved. DOI 10.1016/j.cub.2003.09.023

Plant Cell Polarity: The Ins-and-Outs of Sterol Transport Hazel Betts and Ian Moore

Plant cells and tissues develop polarities essential for completion of development. Recent studies highlight the important role of membrane sterols in plant cell polarity and suggest that sterols may influence the dynamic trafficking of proteins with polar distributions in the plasma membrane.

A colleague introduces his lectures on the ability of plants to adapt to their environment with a photograph of a polar bear on the frozen tundra. The image is taken from a popular natural history publication which eulogises the ability of mammals to colonise diverse hostile environments — but the reason for showing this photograph is of course to point out the underlying vegetation which supports the large furry creatures (literally and ecologically) yet barely registers on the consciousness of the viewer. So it is perhaps with plant membrane lipids; as we focus on the antics of the attention-grabbing proteins, we tend to overlook the contribution made by the lipids in the backdrop. Several new papers [1–4], the most recent in Current Biology [4], bring membrane sterols to stage-centre, highlighting their importance for the establishment of several aspects of cell and tissue polarity. Sterols are components of eukaryotic membranes which are known to be required for polar distribution of certain proteins in animal cells [5]. Cholesterol, the main sterol in animal cells, influences the polar trafficking of proteins through the formation of specific membrane microdomains or ‘lipid rafts’ [6]. Rafts are thought to induce clustering of membrane proteins through cooperative interactions between proteins, cholesterol and sphingolipids [1]. The cell biology of plant sterols has received comparatively little attention. Lipid rafts have not been identified in plants, but a recent proteomic analysis [7] confirmed that GPI-anchored proteins, which associate with sterol-rich rafts in mammals, are abundant in plants. One of these has an asymmetric distribution in polarised cells, where it is required for anisotropic cell expansion [8]. Plant membranes contain very little cholesterol — about 3–10% of all sterols — but other sterols, principally sitosterol, are present. Campesterol, the second most abundant plant sterol in Arabidopsis, is a precursor of brassinosteroid signalling molecules, which influence many aspects of plant development. Evidence that sterols also have developmental roles independent of brassinosteroid signalling has recently come from the analysis of the SMT1 and SMT2 genes of Arabidopsis [1,3]. SMT1 and SMT2 Department Plant Sciences, University of Oxford, South Parks Road, Oxford OX1 3RB, UK. E-mail: [email protected]

Dispatch

encode sterol methyltransferases 1 and 2, respectively. These enzymes act at key branch-points in the sterol biosynthetic pathways (Figure 1). Mutants defective in SMT1 or SMT2 activity exhibit altered ratios of sitosterol, campesterol and cholesterol, accompanied by defects in numerous aspects of development. The mutant phenotypes are not influenced by the application of brassinosteroid precursors or inhibitors of brassinosteroid metabolism, so it appears that the developmental consequences of each mutation are attributable to altered membrane sterol content, rather than altered brassinosteroid signalling [1,3]. Willemsen et al. [3] recently attributed many of the pleiotropic developmental defects in the SMT1deficient orc mutant (smt1orc) to loss of cell and tissue polarity. Several aspects of the mutant phenotype can potentially be explained by the observation that the so-called PIN proteins are mislocalised: some of these proteins normally adopt precise polar distributions in the membranes of specific auxin-transporting cells, and they are implicated in polar transport of auxin during the establishment of morphogenic gradients and tropic curvatures (reviewed in [9,10]). How may altered membrane sterol composition in smt1orc affect PIN protein distribution? Possible causes include defects in a sterol-based signalling system, failure to maintain membrane microdomains analogous to mammalian lipid rafts, or defects in sterol-dependent cargo sorting or vesicle targeting. Direct evidence for any of these possibilities is lacking, but Grebe et al. [4] report evidence for a close association between sterol-enriched membranes and certain membrane proteins, including PIN2, in the endocytic recycling pathways of Arabidopsis roots. This is of interest because the polar distribution of PIN proteins has been shown to be dependent on continuous recycling between the plasma membrane and an internal compartment [11,12]. Grebe et al. [4] used the fluorescent sterol-binding antibiotic filipin to visualise sterol location and trafficking in young epidermal cells of Arabidopsis roots, and found that sterols are abundant in the plasma membrane but can also be detected in structures that associate closely with Golgi stacks. This is consistent with previous biochemical studies of membrane sterol distribution [13]. The Golgi-associated structures may be trans-Golgi cisternae [4], which are known to be enriched for cholesterol in mammalian cells. While filipin may label Golgi cisternae, the filipinlabelled structures overlap with a trans-Golgi marker only at their margins, so an alternative view is that these sterol-enriched structures represent a distinct compartment maintained close to the Golgi stacks, from which it is imperfectly resolved by light microscopy. Given the motility of the Golgi, such an association would be intriguing, but this compartment is perhaps similar to the ‘partially coated reticulum’, which has

Dispatch R782

Cycloartanol SMT1

Cholesterol

smt1orc FK/HYD2 HYD1 24-methylenelophenol

Sterols

smt2

SMT2 SMT3

Sitosterol smt1orc, smt2, hyd1, fk/hyd2

Campesterol

smt2 hyd1, fk/hyd2

Stigmasterol

hyd1, fk/hyd2 smt1orc, smt2

Brassinosteroids Current Biology

Figure 1. The sterol biosynthetic pathways in plants. The sites of action of enzymes and mutants discussed in the text are shown as are the effects of selected mutants on steady-state levels of the principal membrane sterols and cholesterol (blue, increase; red, decrease) [1,3,10,18].

been observed close to the trans-Golgi cisternae in ultrastructural studies of other plant species and has previously been suggested to act as an early endocytic compartment (discussed in [14]). PIN recycling occurs through a pathway requiring the actin cytoskeleton and the GNOM gene product [11,12,15], a GDP/GTP exchange factor (GEF) for small GTPases of the ARF class. ARF GTPases are required for recruitment of vesicle coats during vesicle budding and coat selection at various stages of secretory and endocytic transport. GNOM is a target of the drug brefeldin A (BFA), which causes PIN proteins and other recycling membrane proteins to accumulate intracellularly in ‘BFA compartments’ [11,12,15]. Grebe et al. [4] report further insights into the mode of BFA action and the composition of BFA compartments in Arabidopsis roots (Figure 2). They used filipin to show that both BFA and the actin depolymerising drug cytochalasin D induce membrane sterols to accumulate in the same compartments as plasma membrane proteins, including PIN2. Pulse-labelling with filipin followed by BFA treatment suggested that the internalised membrane sterols are derived from the plasma membrane rather than new synthesis. It is assumed that, at steady state, PIN proteins and sterolcontaining membranes continually recycle to and from an internal endosomal compartment, and that BFA and cytochalasin D inhibit the return of internalised

proteins to the plasma membrane, trapping them in the intracellular compartment (Figure 2). As the internalisation of sterols and plasma membrane proteins exhibited similar kinetics and pharmacological sensitivities, the internalisation pathways seem likely to share similar mechanisms. The internal sterol-enriched compartments induced by BFA and cytochalasin D have been shown in this [4] and other studies to contain plasma membrane proteins besides PIN proteins, so the sterol-associated recycling pathways are apparently not specific to PIN proteins or to proteins with a polar distribution. But the ‘BFA compartment’ is in reality an agglomeration of discrete vesicles (or tubules) which may have distinct biochemical identities, so it may be that only some are enriched for sterols and certain plasma membrane proteins [4]. Furthermore Grebe et al. [4] found that AUX1, a putative auxin influx carrier which does not exhibit a polar distribution in epidermal cells, is not found in epidermal BFA compartments, suggesting AUX1 trafficking occurs by a different pathway (or at a different rate) to sterols, PIN2, and other plasma membrane proteins. Interestingly, in smt1orc mutants, AUX1 still exhibits its typically polar distribution in protophloem cells so it may be that polar trafficking of AUX1 is not strongly dependent on normal sterol composition [3]. Taken together, these results show that plant sterols are important for polar localisation of membrane proteins such as PIN proteins, and that the sterols are internalised and recycled together with certain plasma membrane proteins. But it is still not clear how altered membrane sterol composition affects cell polarity. It cannot result simply from the reduction in sitosterol content, because PIN1 and PIN2 are normally distributed in fackel/hyd2 and hyd1 mutants, which have reduced sitosterol and campesterol content [2]. So PIN mislocalisation in smt1orc is more likely to result either from its particular ratio of major membrane sterols or, interestingly, perhaps from the increased cholesterol content which is most pronounced in this mutant. The location of PIN proteins in smt2 and smt3 mutants could be revealing. Filipin labelling did not provide any evidence of polarised sterol distribution in the plasma membrane, but as filipin cannot discriminate between the major sterols, it remains possible that individual plasma membrane domains and endocytic organelles may have differing sterol composition. Alternatively, the altered sterol content of smt1orc mutants may affect cargo sorting or vesicle targeting on the endocytic recycling pathways. The mechanisms of membrane internalisation are still largely obscure, however, so it will be important to identify more molecular components of the recycling pathways and to establish their respective roles in trafficking of PINs, AUX1 and sterols. One candidate suggested by Grebe et al. [4] is the Arabidopsis Rab GTPase ARA6, because a fusion protein of this protein with green fluorescent protein (GFP) was seen to associate with some of the filipinlabelled structures. Rab GTPases are key regulators of many aspects of vesicle transport and can act as local organisers of membrane microdomains [16]. ARA6 is a member of the F1 subclass of plant Rabs, which are unique among Rabs in that they attach to membranes

Current Biology R783

AUX1 PIN2

AUX1 PI

Lti6a-GFP

N2 Lti6a-

GFP PM

BFA (10µM)

cytoD

cytoD

E

E

E

(sterols, Lti6a-GFP, FM4-64)

E ?

BFA (>25µM) (or 10µM BFA in der1 mutant)

?

? Golgi

ER

Current Biology

X BFA compartment (PIN2, sterols, Lti6A, GNOM, PM ATPase, FM4-64 but not AUX1)

Figure 2. The effect of BFA in young epidermal cells of Arabidopsis roots. Membranes enriched for sterols are indicated by red lines. Putative endosomes (E) observed in the presence of BFA and cytochalasin D (cytoD) are green. The unidentified filipinlabelled compartment that overlaps with a trans-Golgi marker in untreated cells is marked ‘?’. CytoD or low concentrations of BFA (left side of figure) result in the accumulation of sterols and the plasma membrane (PM) marker Lti6a-GFP with FM4-64 in small dispersed endosomes. Higher concentrations of BFA (right side) produce larger ‘BFA compartments’ surrounded by Golgi stacks. BFA compartments may represent a distinct organelle upstream of endosomes (E) or an agglomeration of these endosomes, perhaps through perturbation of the actin cytoskeleton, as suggested by the BFA hypersensitivity of der1 mutants, and [19]. FRAP studies reveal unperturbed biosynthetic transport of Lti6a–GFP (and presumably sterols) to the plasma membrane in the presence of BFA or cytoD in these cells, accumulation in BFA compartments after internalisation from the plasma membrane, and delayed inhibition of internalisation by cytoD. Data principally from [4] but also [11,12,15].

via amino-terminal myristoylation and palmitoylation rather than carboxy-terminal isoprenylation [17]. It is tempting to speculate that this property of F1 Rabs is related to association with particular membrane lipids; we think it is fair to say, however, that the evidence of a role for ARA6 in endocytic sterol transport is suggestive rather than conclusive. Firstly, in epidermal cells, ARA6–GFP and filipin do not quantitatively colocalise so it is not clear whether they label different domains on the same compartment, a mixture of common and distinct compartments, or distinct compartments that are sometimes in close proximity. Secondly, in response to cytochalasin D and BFA treatments, whereas plasma membrane proteins and the dye FM4-64 exhibit extensive colocalisation with filipin, ARA6-labelled compartments seem to cluster around the filipin-labelled structures but, like the Golgi stacks, remain distinct. Thus they do not appear to be the principal compartments in which sterols accumulate when recycling is blocked by drug treatment. It will be important to establish whether ARA6 is required for recycling of membrane proteins and sterols. It will also be of interest to investigate the distribution and role of the related F2 subclass of Rabs, which are involved in vacuolar protein sorting and are

located on a compartment whose structure is altered in the gnom mutant [15]. To labour the theatrical metaphor, membrane sterols have emerged into the limelight of centre stage, but the opening acts in this play suggest that the plot may become rather more convoluted before all is revealed. References 1. Carland, F.M., Fujioka, S. Takatsuto, S., Yoshida, S. and Nelson, T. (2002). The identification of CVP1 reveals a role for sterols in vascular patterning. Plant Cell 14, 2045–2058. 2. Souter, M., Topping, J., Pullen, M., Friml, J., Palme, K., Hackett, R., Grierson, D. and Lindsey, K. (2002). hydra mutants of Arabidopsis are defective in sterol profiles and auxin and ethylene signaling. Plant Cell 14, 1017–1031. 3. Willemsen, V., Friml, J., Grebe, M., van der Toorn, A., Palme, K. and Scheres, B. (2003). Cell polarity and PIN protein positioning in Arabidopsis require STEROL METHYLTRANSFERASE1 function. Plant Cell 15, 612–625. 4. Grebe, M., Xu, J., Mobius, W., Ueda, T., Nakano, A., Geuze, H.J., Rook, M.B., and Scheres, B. (2003). Arabidopsis sterol endocytosis involves actin-mediated trafficking via ARA6-positive early endosomes. Curr. Biol. 19 August issue. 5. Keller, P. and Simons, K. (1998). Cholesterol is required for surface transport of influenza virus hemagglutinin. J. Cell Biol. 140, 1357–1367. 6. Simons, K. and Toomre, D. (2000). Lipid rafts and signal transduction. Nat. Rev. Mol. Cell Biol. 1, 31–39. 7. Borner, G.H.H., Lilley, K., Stevens, T.J. and Dupree, P. (2003). Identification of glycosylphosphatidylinositol-anchored proteins in Arabidopsis. A proteomic and genomic analysis. Plant Physiol. 132, 568–577. 8. Schindelman, G., Morikami, A., Jung, J., Baskin, T.I., Carpita, N.C., Derbyshire, P., McCann, M.C. and Benfey, P.N. (2001). COBRA encodes a putative GPI-anchored protein, which is polarly localized and necessary for oriented cell expansion in Arabidopsis. Genes Dev. 15, 1115–1127. 9. Friml, J. (2003). Auxin transport- shaping the plant. Curr. Opin. Plant Biol. 6, 7–12. 10. Reinhardt, D. (2003). Vascular patterning: more than just auxin? Curr. Biol. 13, R485–R487. 11. Steinmann, T., Geldner, N., Grebe, M., Mangold, S., Jackson, C.L., Paris, S., Galweiler, L., Palme, K. and Jurgens, G. (1999). Coordinated polar localization of auxin efflux carrier PIN1 by GNOM ARF GEF. Science 286, 316–318. 12. Geldner, N. and Friml, J. Stierhof YD., Juergens G. and Palme K. (2001). Auxin transport inhibitors block PIN1 cycling and vesicle trafficking. Nature 413, 425–428. 13. Moreau, P., Hartmann MA., Peret AM., Sturbois-Balcerzak and Cassagne C. (1998).Transport of sterols to the plasma membrane of leek seedlings. Plant Physiol. 117, 931–937. 14. Staehelin, L.A. and Moore, I. (1995). The plant Golgi apparatus: structure, functional organisation and trafficking mechanisms. Annu. Rev. Plant. Physiol. Plant Mol. Biol. 46, 261–288. 15. Geldner, N. Anders, N., Wolters, H., Keicher, J., Kornberger, W., Muller, P., Delbarre, A., Ueda, T., Nakano, A. and Jurgens, G. (2003). The Arabidopsis GNOM ARF-GEF mediates endosomal recycling, auxin transport and auxin-dependent plant growth. Cell 112, 219–230. 16. Zerial, M. and McBride, H. (2001). Rab proteins as membrane organisers. Nat. Rev. Mol. Cell Biol. 2, 107–117. 17. Ueda, T., Yamaguchi, M., Uchimiya, H. and Nakano, A. (2001). Ara6, a plant-unique novel type Rab GTPase, functions in the endocytic pathway of Arabidopsis thaliana. EMBO J. 20, 4730–4741. 18. Schrick, K., Mayer, U., Martin, G., Bellini, C., Kuhnt, C., Schmidt, J. and Juergens, G. (2002). Interactions between sterol biosynthesis genes in embryonic development of Arabidopsis. Plant J. 30, 1179–1190. 19. Petrasek, J., Cerna, A., Schwarzerova, K., Elckner, M., Morris, D.A. and Zazimalova, E. (2003). Do phytotropins inhibit auxin efflux by impairing vesicle traffic? Plant Physiol. 131, 254–263.