Plant Cell Physiol. 45(5): 573–582 (2004)JSPP © 2004

Microtubules of Guard Cells are Light Sensitive

Maoz Lahav 1, Mohamad Abu-Abied 1, Eduard Belausov 1, Amnon Schwartz 2 and Einat Sadot 1, 3

1 The Department of Ornamental Horticulture, The Volcani Center, Bet-Dagan 50250 Israel 2 The Robert H. Smith Institute of Plant Science and Genetics, The Faculty of Agriculture, Food and Environmental Sciences, The Hebrew Univer-sity of Jerusalem, Rehovot, Israel

;Guard cells of stomata are characterized by ordered

bundles of microtubules radiating from the ventral sidetoward the dorsal side of the cylindrical cell. It was sug-gested that microtubules play a role in directing the radialarrangement of the cellulose micro-fibrils of guard cells.However, the role of microtubules in daily cycles of openingand closing of stomata is not clear. The organization ofmicrotubules in guard cells of Commelina communis leaveswas studied by analysis of three-dimensional immunofluo-rescent images. It was found that while guard cell microtu-bules in the epidermis of leaves incubated in the light wereorganized in parallel, straight and dense bundles, in thedark they were less straight and oriented randomly nearthe stomatal pore. The effect of blue and red light on theorganization of guard cell microtubules resembled theeffects of white light and dark respectively. When stomatawere induced to open in the dark with fusicoccin, microtu-bules remained in the dark configuration. Furthermore,when incubated in the light, guard cell microtubules weremore resistant to oryzalin. Similarly, microtubules of Arabi-dopsis guard cells, expressing green fluorescent protein–tubulin α 6, were disorganized in the dark, but were organ-ized in parallel arrays in the presence of white light. Thedynamics of microtubule rearrangement upon transfer ofintact leaves from dark to light was followed in single sto-mata, showing that an arrangement of microtubules typi-cal for light conditions was obtained after 1 h in the light.Our data suggest that microtubule organization in guardcells is responsive to light signals.

Keywords: Arabidopsis thaliana — Commelina communis —Guard cells — Light — Microtubules.

Abbreviations: ABA, abscisic acid; GFP, green fluorescent pro-tein; TUA6, tubulin α 6; MAP, microtubule-associated protein; MBD,microtubule-binding domain; EF-1α, elongation factor 1α; DIC, dif-ferential interference contrast.

Introduction

Guard cells are highly specialized multisensory cells thatflank stomatal pores in the leaf epidermis. Guard cells control

stomatal aperture via signal-induced changes in their volumeand shape. Such signal-mediated movement of guard cells reg-ulates the CO2 influx into leaves for photosynthesis and the lossof water vapor by transpiration.

Guard cell signaling integrates hormonal stimuli, light,CO2 levels and other environmental stimuli to regulate sto-matal aperture. A network of ion channels in the plasmamembrane and the vacuolar membrane of guard cells, whichcontrols ion fluxes and stomatal movement, has been character-ized. Overall, the current understanding of how guard cellsignalling networks operate is well advanced (Assmann andShimazaki 1999, Blatt 2000, Fedoroff 2002, Hetherington2001, Ng et al. 2001, Schroeder et al. 2001a, Schroeder et al.2001b, Zeiger 2000). However, an intriguing challenge forfuture studies is to decipher how the plant cytoskeleton, whichis inherently associated with signalling (Geitmann and Emons2000, Kost and Chua 2002, Kost et al. 1999, Lloyd and Hussey2001, Moore et al. 1998, Nick 1998, Volkmann and Baluska1999, Wasteneys 2002) integrates and responds to these signal-ling cascades.

Guard cells contain cortical microtubules that are radiatingfrom the ventral pore side toward the dorsal side (Marc et al.1989a, Marc et al. 1989b, Marc and Palevitz 1990, McDonaldet al. 1993, Palevitz and Hepler 1976). This specific organiza-tion of microtubules has been implicated in guard cell develop-ment, where microtubules are thought to guide the depositionof cellulose microfibrils in distinctive orientation and distribu-tion, which is crucial to the ability of stomata to open and closeupon changes in guard cell volume (Palevitz and Hepler 1976).

Interestingly, unlike in other cells, such as pea epidermalepicotyl or Arabidopsis root cells where microtubules changeorientation from transverse to longitudinal when elongationdeclines (Sugimoto et al. 2000, Granger and Cyr 2001, Yuan etal. 1994), and in leaf epidermal cells where microtubules pos-sess different orientation at different stages of leaf develop-ment (Fu et al. 2002), in guard cells the strict organization ofradially orientated microtubules remains throughout their life-time. This suggests that microtubules of guard cells are uniquein their function and, most likely, in their associated proteins oreffectors as well. Nevertheless, neither a mechanical role nor arole in signaling perception is known for microtubules inmature guard cells and their involvement in stomatal move-ments is controversial. Assmann and Baskin (1998) reportedthat treatment with colchicine (a microtubule-disrupting agent)

3 Corresponding author: E-mail, vhesadot@volcani.agri.gov.il; Fax, +972-3-9669583.

573

Guard cell microtubules574

failed to interfere with stomatal function, suggesting that micro-tubules are not required for stomatal movements. In contrast,Fukuda et al. (1998) reported that propyzamide (a microtubule-disrupting herbicide) and taxol (a microtubule stabilizingagent) interfered with stomatal function, suggesting that micro-tubules are required for stomatal movements. Marcus et al.(2001) used several microtubule-disrupting herbicides as wellas over-expression of microtubule-binding domain of micro-tubule associated protein 4 – green fluorescent protein (GFP–MBD) and showed that microtubules are necessary for stomatalopening in a specific manner prior to the ionic fluxes. Similarly,over-expression of the microtubule-associated protein elonga-tion-factor-1-α (EF1-α) inhibited stomatal opening upon whitelight illumination (Moore and Cyr 2000). In addition, whereasabscisic-acid (ABA) induction of stomatal closure was accom-panied by microtubule fragmentation in guard cells of Vicia faba(Jiang et al. 1996), guard cell microtubules of Commelina com-munis (Eun and Lee 1997) and Arabidopsis thaliana (Lemichezet al. 2001) remained intact after ABA treatment. On the otherhand, guard cells of V. faba showed dynamic changes in micro-tubule organization during stomatal opening and closing(Fukuda et al. 1998, Fukuda et al. 2000, Yu et al. 2001).

In the present research we have compared the organiza-tion and integrity of microtubules in guard cells when incu-bated in the light or in the dark. We report here that, in the

light, microtubules are radiating straight from the ventral sideof guard cells and are arranged in parallel arrays. In contrast, inthe dark, microtubules range from less parallel to randomly ori-ented at the region closer to the ventral side of the cylindricalguard cell. This “light-condition configuration” is cell volume-independent and was reproduced by blue light illumination, butnot by red light. In addition, we show that microtubules ofguard cells incubated in the light are more stable. A differentorganization of microtubules was also observed in live guardcells of A. thaliana expressing GFP–tubulin. The kinetics of thechanges of microtubules in stomata transferred from dark tolight is demonstrated.

Results

The organization of guard cell microtubules differs betweenleaves that are incubated in the light or in the dark

It has been previously shown that in guard cells of V. fabamicrotubules change during cycles of opening and closing ofstomata (Fukuda et al. 1998, Fukuda et al. 2000, Yu et al.2001).

In the present study we followed the organization ofmicrotubules in guard cells of C. communis after exposingdetached leaves to different light treatments. Following thetreatment, abaxial epidermes were peeled, fixed, and microtu-

Fig. 1 Differences in microtubule organization in guard cells incubated in the light or in the dark. Leaves of Commelina communis were incu-bated in the light (A–C) or in the dark (D–F). Epidermal peels were fixed and microtubules were labeled with anti-tubulin antibody. Images wereacquired by a confocal microscope and are three-dimensional reconstructions of a series of optical sections. (A–C) Rotation around the axis of thepore of stomata incubated in the light: (A) –35°; (B) 0°; (C) +35°. (D–F) Rotation around the axis of the pore of stomata incubated in the dark:(D) –35°; (E) 0°; (F) +35°. Scale bar = 10 µm.

Guard cell microtubules 575

bules were immuno-stained with anti-tubulin antibody. Imageswere acquired by a confocal microscope as follows; series ofoptic sections were acquired, projected and three-dimensionalreconstruction of the whole cell was performed and rotated rightor left. For the whole movie of rotating stomata see supplemen-tal data where movies 1 and 2 correspond to stomata in the lightand in the dark respectively. This approach enabled us to followindividual microtubules along and surrounding the whole cell.It is worth noting here that during fixation the turgor pressureof guard cells is lost but cytoskeletal structures are preserved(Assmann and Baskin 1998, and our own data). It is shown thatin guard cells incubated in the light most of the microtubulesare straight, and that they radiate from the ventral side of thecylindrical guard cell toward the dorsal side, encircle the dorsalside and return back toward the ventral side. Most of the micro-tubules are arranged in parallel arrays. This three-dimensionalview conforms to previous observations and models describingin detail microtubule arrangement in guard cells (Marc et al.1989a, Marc et al. 1989b, Marc and Palevitz 1990). In con-trast, in guard cells incubated in the dark there are microtu-bules with random orientation that form crossovers mainly nearthe ventral pole. Interestingly, the differences between the lightand dark configurations of microtubules were less pronouncedwhen experiments were carried out with epidermal peels.

Blue and red light effects on microtubules of guard cells resem-ble the white light and dark effects respectively

Blue light is one of the environmental stimulators affect-ing opening of stomata (Assmann and Shimazaki 1999,

Assmann et al. 1985, Gorton et al. 1993, Schroeder et al.2001a, Shimazaki et al. 1986, Zeiger 2000). Blue light percep-tion by guard cells is mediated by the receptors phototropins 1and 2 (Kinoshita et al. 2001), and leads to phosphorylation ofthe C-terminus of a plasma membrane ATPase and its activa-tion (Kinoshita and Shimazaki 1999). Since changes in micro-tubule organization were observed upon illumination by whitelight (Fig. 1), we analyzed the relative efficiency of the blueand the red spectral regimes in affecting microtubule organiza-tion. The abaxial side of C. communis leaves was illuminatedby blue or red light for 2.5–3 h, removed, fixed and immuno-stained for microtubules. Fig. 2aA and 2b show that guard cellsof leaves illuminated by blue light possessed microtubulesarranged in the white-light-condition configuration, that is,in straight and parallel bundles. Nevertheless, the blue-light-condition configuration of microtubules differs from that of thewhite light configuration with more crossing over of microtu-bules observed near the ventral zone (Fig. 2aA). Guard cellsilluminated with red light possessed microtubules that resem-bled the dark-condition configuration (Fig. 2aB, 2b). This sug-gests that blue light can initiate a signaling pathway affectingmicrotubule organization in guard cells.

The microtubule organization is dependent on light but not onstomatal opening or on the activation of plasma membraneATPase

Next we addressed the question of whether microtubuleorganization was dependent on stomatal aperture. Leaves weretreated in the dark with fusicoccin, a drug that activates plasma

Fig. 2 Blue and red light confers the light and darkeffects on guard cell microtubules respectively. Abaxialleaf surfaces of C. communis were illuminated either byred or blue light. Epidermal peels were then fixed andstained with an anti-tubulin antibody (2a upper panel) orexamined under differential interference contrast micros-copy (2a lower panel). Guard cells were scored for thestraight and parallel “light-condition configuration” ofmicrotubules (light-like), non straight and non parallel“dark-condition configuration” organization (dark like),or “intermediate” (2b). Experiments were reproduced sev-eral times. Scale bar of fluorescent images = 10 µm, andof differential interference contrast (DIC) images =20 µm.

Guard cell microtubules576

membrane ATPases, and, eventually, leads to stomatal opening(Assmann and Schwartz 1992, Johansson et al. 1993). It isdemonstrated in Fig. 3aC that fusicoccin, which was applied tothe intact leaf via the incubation medium, promoted partialopening of the stomata in the dark. Nevertheless the microtu-bule configuration in these guard cells resembled the dark-condition configuration (Fig. 3aB, C, 3b). Taken together, thedata suggest that blue light affected microtubules independentof stomatal aperture and upstream or irrespective to the activa-tion of plasma membrane ATPase.

Microtubules of guard cells in the light are more resistant tooryzalin treatment

We next asked whether the different configuration ofmicrotubules in guard cells incubated in the light vs. dark isaccompanied by a difference in their stability. Leaves were pre-

incubated in white light or in the dark for 2.5 h and then trans-ferred to buffer containing 5 µM oryzalin, a microtubule dis-rupting herbicide, for different periods of time, from 5 min to60 min. Oryzalin binds dimers of α and β plant tubulin withhigh affinity to form an oryzalin/tubulin complex that eventu-ally leads to microtubule depolymerization. It was found that inthe light, mild disruption of microtubules was observed in~10% of guard cells after 5, 15 or 30 min of oryzalin treatment(Fig. 4aG, H, I, 4b). After 60 min of treatment with oryzalin inthe light, 70% of guard cells contained severely disruptedmicrotubules (Fig. 4aJ, 4b). In contrast, when leaves were incu-bated in the dark prior to and during the oryzalin treatment,severe disruption of microtubules was observed already 5 minfollowing the application of the drug in 40% of guard cells(Fig. 4aB). More than 50% of guard cells possessed severelydisrupted microtubules after 15 and 30 min of oryzalin treat-

Fig. 3 Microtubule organization is dependent on light butnot on stomatal opening. Leaves of C. communis were eitherincubated in the light (3aA), or dark (3aB), or in the dark inthe presence of fusicoccin (Fc) (3aC). Following these treat-ments epidermal peels were fixed, stained for microtubuleswith anti tubulin antibody (3a upper panel), or examinedunder differential interference contrast microscopy (3a lowerpanel). Guard cells were scored for the straight and parallel“light-like” microtubule organization, non-straight andnon-parallel “dark like” organization, or “intermediate”.Experiments were reproduced several times. Scale bar offluorescent images = 10 µm, and of DIC images = 20 µm.

Guard cell microtubules 577

ment (Fig. 4a C, D, 4b) while after 60 min, 80% of guard cellsin the dark contained no microtubules (Fig. 4aE, 4b). DMSOalone at 0.05% did not affect microtubule organization (notshown). Interestingly, similar results were obtained whenleaves were incubated with 1 mM colchicine (another microtu-bule disrupting agent). Microtubules of guard cells that wereincubated in the light were hardly affected by colchicine treat-ment of 1 h while after this time, microtubules of guard cellsthat were incubated in the dark were severely disrupted (notshown). The mild treatment of intact leaves with microtubuledisrupting agents revealed the difference in microtubule stability.

Light sensitivity of microtubules in live GFP–tubulin expressingcells

In order to further confirm the differences in the configu-ration of guard cell microtubules observed in the light or dark,we followed GFP-labeled microtubules in A. thaliana express-ing GFP–tubulin α 6 (TUA6) (Ueda et al. 1999). Leaves wereincubated in the dark or in the light and guard cells of the abax-ial side of the leaf were inspected by the confocal microscope.It is demonstrated in Fig. 5 that when leaves were incubated inthe light, guard cell microtubules were straight and parallelwhile in dark treated leaves more than 50% of guard cells pos-

sess a diffuse pattern of GFP, suggestive of unpolymerizedtubulin. In order to follow the dynamic changes of microtubulesin guard cells, leaves were incubated in the dark, transferred tolight while on the microscope stage, and the organization ofmicrotubules followed in single stomata. The results of two dif-ferent experiments are demonstrated in Fig. 6. It is shown thatafter 2 h in the dark, the GFP signal was diffuse, suggestingthat the microtubules were mostly depolymerized. However,following 0.5 h of white light illumination, organized microtu-bules appeared, although some microtubules with different ori-entations remained (Fig. 6B, E, arrows). Following 1 h of lightillumination, microtubules became organized in parallel arrays(Fig. 6C, F, arrows). Interestingly, during this time in the light,microtubules of the subsidiary cells changed their highly varia-ble organization to become perpendicular to the guard cellmicrotubules (Fig. 6C, F, arrows). Similarly, microtubules ofguard cells and of subsidiary cells were perpendicular in sto-mata complexes of C. communis (Fig. 1B, 2aA). Moreover,Fig. 6 demonstrates that the changes in microtubule organiza-tion in guard cells preceded the increase in stomatal aperture.These results confirm and reinforce our data obtained fromfixed cells.

Fig. 4 Microtubules of guard cells incubated in the light are moreresistant to oryzalin treatment than microtubules of guard cells in thedark. Leaves of C. communis were incubated either in the light or inthe dark and then transferred to the same buffer containing oryzalin inthe light or in the dark for additional time periods as indicated. (a) Epi-dermal peels were fixed and stained with anti-tubulin antibody. (A, F)Control without oryzalin; (B, G) 5 min; (C, H) 15 min; (D, I) 30 min;(E, J) 60 min. (A–E) white light; (F–J) dark. (b) Guard cells werescored for disrupted or intact microtubules; the chart represents theaverage of three experiments. Scale bar = 10 µm.

Guard cell microtubules578

The dynamics of microtubules of guard cells transferred fromdark to light

In order to study the dynamics of microtubule organiza-tion in guard cells that were transferred from dark to light, wefollowed the changes in microtubule arrangement in single sto-mata every 5 min, upon transferring the leaves from dark tolight. As shown in Fig. 7a, microtubular organization changed

from a diffuse fluorescent signal to clear filaments within thefirst 5 min of white light illumination. Furthermore, whiletransverse microtubules retain their orientation through out theexperiment, “disoriented” oblique microtubules appeared anddisappeared from image to image (Fig. 7a arrows). This is fur-ther demonstrated in Fig. 7b, in which the images after 20, 25and 30 min (from Fig. 7a) were colored red, blue and greenrespectively and projected together. It is shown that while sev-eral transverse microtubules contain white pixels, indicatingco-localization of the three colors throughout the 10-minperiod, and therefore delineating static microtubules (Fig. 7barrow), dis-oriented microtubules are indicated in blue (Fig. 7barrow heads). A movie of this experiment (supplementarymovie 3) shows clearly that oblique microtubules are transient.In addition, reorientation of microtubules at the subsidiary cellscan be observed (supplementary movie 3). At this stage it is notknown whether the oblique microtubules in guard cells, shrinkas a result of catastrophe, or reorient and co-align with trans-verse microtubules during this time. Nevertheless, at the end ofthe process, when stomata are wide open, oblique microtu-bules were no longer observed Fig. 5A.

Discussion

A number of reports have described changes in microtu-bule organization in mature guard cells during cycles of open-

Fig. 5 Differences in Guard cell microtu-bules of GFP–TUA6 Arabidopsis plants incu-bated in the dark or light. Leaves ofArabidopsis plants expressing GFP-TUA6were incubated in the dark or in the light andthen inspected by the confocal microscope.(A) Open stomata; (B) closed stomata. Eachimage is a projection of six optical sections,1 µm apart, covering the whole stomata topto bottom. Scale bar 10 µm.

Fig. 6 Changes in microtubule organization of guard cells trans-ferred from dark to light. Leaves of Arabidopsis plants expressingGFP–TUA6 were incubated in the dark for 2 h prior to the experi-ments. Shown here are two different experiments (A–C and D–F) ofsingle stomata inspected by the confocal microscope. The first imageacquisition (A, D) showing microtubule organization in the dark. Theleaf was then illuminated by white light while positioned on the micro-scope stage. Two more image acquisitions were performed after 0.5 h(B, E) and 1 h (C, F) of white light illumination. It is shown that afterdark incubation the GFP signal in guard cells is diffuse (AD). After0.5 h of light illumination, organized microtubules appear but they arenot uniformly oriented (arrows in B and E). Microtubules of guardcells are parallel to each other and perpendicular to those of subsidiarycells after 1 h of light (C and F arrows). Scale bar = 10 µm.

Guard cell microtubules 579

ing and closing of stomata. Fukuda and colleagues followed theorganization of microtubules in guard cells of V. faba every 2 hfor 24 h (Fukuda et al. 1998, Fukuda et al. 2000). They foundthat in the morning hours, microtubules were organized in par-

allel arrays perpendicular to the axis of the pore, the same aswas previously described (Marc et al. 1989a, Marc et al. 1989b,Marc and Palevitz 1990, McDonald et al. 1993, Palevitz andHepler 1976). However, in the night microtubules were dis-

Fig. 7 The dynamics of guard cell micro-tubules during the transition of leaves fromdark to light. Leaves of Arabidopsis plantsexpressing GFP–TUA6 were incubated inthe dark for 2 h. (a) Time-lapse imageacquisition of single stomata was per-formed every 5 min for 35 min as indi-cated. During this time the microscopestage was illuminated by white light inbetween image acquisitions. Arrows marktemporary oblique microtubules. (b) Theimages of 20, 25 and 30 min were coloredred, blue, and green respectively, and pro-jected together. White pixels show staticmicrotubules (arrows), blue/red/green pix-els show dynamic microtubules (arrowheads). Scale bar = 5 µm.

Guard cell microtubules580

rupted, shorter and more randomly oriented (Fukuda et al.1998, Fukuda et al. 2000). While Fukuda and colleagues fixedthe cells and immunostained them, Yu and colleagues followedmicrotubules in live guard cells of V. faba after microinjectionof fluorescent tubulin. Similarly, they observed fine parallelarrays of microtubules when stomata were open in the light butin the dark, microtubules were crooked in appearance (Yu et al.2001). In the present study, we have followed the microtubuleconfiguration in guard cells of C. communis by immunostain-ing fixed tissues and in A. thaliana live cells expressing GFP–tubulin. We characterized the changes of microtubule organiza-tion in the light and dark and separated the light effect from theputative stomatal aperture effect. Similar to Fukuda and Yu weshow that microtubules in guard cells of C. communis, althoughnot crooked in appearance in the dark, are organized dif-ferently from microtubules in the light. Three-dimensionalconfocal images presented here revealed that in the dark,microtubules were disordered and especially near the ventralzone possess random orientation. In contrast, in the light guardcells contain straight parallel arrays of microtubules that radi-ate from the tip of the ventral side to the dorsal side, and sur-round the cell perpendicular to its long axis. In addition, it isdemonstrated here that these well-organized microtubule arraysin the light are more resistant to mild treatment with microtu-bule-disrupting agents such as oryzalin and colchicine. Oryza-lin, a dinitroaniline herbicide, binds plant tubulin dimers veryefficiently (Hugdahl and Morejohn 1993), and leads to inhibi-tion of microtubule polymerization (Morejohn et al. 1987a).Interestingly, the affinity of oryzalin to polymerized microtu-bules is lower than to unpolymerized tubulin (Hugdahl andMorejohn 1993). Thus, the higher sensitivity of microtubules ofguard cells in the dark to oryzalin may suggest a shift in theequilibrium of polymerized/unpolymerized tubulin toward theunpolymerized state in the dark. Furthermore, oryzalin bindsplant tubulin dimers much more efficiently than colchicine(Morejohn et al. 1987b, Hugdahl and Morejohn 1993) and themechanism of oryzalin binding to plant tubulin differs fromthat of colchicine. While the affinity of oryzalin to plant tubu-lin is high, the oryzalin/tubulin complex can dissociate com-pletely. In contrast, the affinity of colchicine to plant tubulin ismuch lower but the complex colchicine/tubulin does not read-ily dissociate (Hugdahl and Morejohn 1993). Nevertheless, weobtained similar results with both drugs, showing higher stabil-ity of guard cell microtubules in the light. Resistance of micro-tubules to disrupting agents can reflect stabilization of thepolymer by microtubule-associated proteins (MAPs). Thesecan affect the dynamic instability of microtubules by reducingthe rate of tubulin dissociation. Therefore, light-induced reor-ganization and stabilization of transverse microtubules can bemediated by specific MAPs that might possess a role in sto-matal function. Indeed, it was shown that over-expression ofmicrotubule-associated proteins GFP–MBD (Marcus et al.2001) or GFP–EF-1α (Moore and Cyr 2000) inhibited light-induced stomatal opening. This suggests a competition on

microtubule-binding sites between the exogenous GFP chime-ras and endogenous microtubule-associated proteins playing arole in light-induced stomatal opening. In addition, it wasshown that light and fusicoccin effects on stomatal aperturewere synergistic (Assmann and Schwartz 1992), suggestingthat light may initiate both ATPase-dependent and -independentsignaling pathways affecting stomatal opening. Taken together,these data and our results demonstrating that fusicoccin in thedark did not affect microtubule organization, suggest that afusicoccin-independent pathway induced by light might involvemicrotubules. The isolation of guard cell-specific microtubule-associated proteins is a fascinating challenge still to beexplored.

We show here that blue and white light had a similar effecton guard cell microtubules. Our observation that blue light canpromote the “light-specific configuration” of microtubules sug-gests the existence of a blue light signaling pathway targeting aspecific microtubule-associated protein or effector. The bluelight effect on stomatal aperture is mediated by phot1 andphot2 receptors (Kinoshita et al. 2001). Phot1 is a serine/threo-nine protein kinase, the substrates of which are still unknown(reviewed in Lin 2002). It is a fascinating assumption that theblue light receptors might initiate a signaling pathway activat-ing microtubule-associated proteins that may lead to microtu-bule reorganization. It is worth noting, that a blue light effecton microtubule reorganization and reorientation was previ-ously reported in maize coleoptiles, where a shift from trans-verse to longitudinal orientation of microtubules was observedon the side facing the blue light irradiation (Nick 1998, Nick etal. 1990).

Further, we demonstrate dynamic changes in microtu-bules of live guard cells of Arabidopsis plants expressing GFP–TUA6. Guard cells of these plants exerted microtubule reorgan-ization shortly after the application of light to a leaf previouslyincubated in the dark. Close examination revealed that obliquemicrotubules had a short lifetime while transverse microtu-bules formed and eventually conferred an array of parallelmicrotubules surrounding the cell. Dynamic instability of dis-cordant microtubule and/or their co-alignment with neighbor-ing microtubules characterize the transition of microtubuleorientation from transverse to longitudinal, or vice versa, whenelongation declines or in response to various plant hormones(Himmelspach et al. 1999, Shibaoka 1994, Yuan et al. 1994).However, the machinery underling microtubule reorganizationin guard cells transferred from dark to light is still to be deter-mined.

We believe that the results presented here support theassumption that microtubules of guard cells do have a role indaily cycles of opening and closing, a role that is still to bedetermined.

Guard cell microtubules 581

Material and Methods

PlantsC. communis plants were grown from seeds in pots containing

peat in a growth chamber at a constant temperature of 25°C and 16 hlight/8 h dark cycles. The youngest fully expanded leaves of 5- to 7-week-old plants were used for the experiments.

A. thaliana seeds of plants expressing GFP–TUA6, kindly pro-vided by Dr. Takashi Hashimoto (Nara Institute of Science and Tech-nology, Nara, Japan), were seeded in MS agar plates (Murashige andSkoog 1962) and after 2.5 weeks transferred to pots with peat. GFPtubulin was detected in 2- to 3-week-old seedlings.

Light and drug treatmentsThe experiments were performed in a temperature-controlled

room at 25±1°C. Light was provided from above by a bank of fluores-cent tubes at photosynthetic photon flux density of 200 µmol m–2 s–1 atthe leaf surface. Blue or red light was obtained by filtering white lightthrough blue or red Plexiglas (Rohm and Haas, Hayward, CA, U.S.A.).The photosynthetic photon flux density of red and blue light wasadjusted to 20 µmol m–2 s–1 at the leaf level. All the treatments wereperformed using detached leaves, while at the end of the treatments theabaxial epidermis was removed by forceps and was fixed immediately.Uniform stomatal opening to a range of 10–14 µm was achieved in C.communis by incubating the leaves under water surface in Petri dishesin the light for about 2–3 h. Stomata in leaves that were treated simi-larly but kept in the dark attained an aperture of no more than 4 µmwithin 2–3 h. Stomatal opening in the dark was achieved by holdingthe leaves under the surface of incubation medium that contained10 mM MES buffer adjusted to pH 6.1 with KOH and 2 µM fusicoc-cin (Sigma F0537). The fusicoccin treatment caused stomatal opening(mainly near the central vein) in the dark to 9±2 µm within 3 h. Oryza-lin (PS-410 Sigma) was prepared as 10 mM stock solution in DMSOand used at a final concentration of 5 µM and colchicine (SigmaC3915) was prepared at 100 mM in H2O and used at a final concentra-tion of 1 mM.

Immunofluorescence staining, confocal microscopy and image analysisThe following antibodies were used: a monoclonal anti-tubulin

antibody (Sigma, DM 1A T9026) and goat anti-mouse conjugated toCy3 (Jackson 115-165-072).

Epidermal peels were fixed for 1 h in PME buffer (50 mM PIPESpH 6.9, 5 mM EGTA, 2 mM MgSO4) containing 3% paraformalde-hyde. The peels were then washed in PBS and membrane permeabilitywas obtained by the freeze shattering method (Wasteneys et al. 1997).Blocking was performed in PBS with 1% BSA and 0.05% Triton X-100 at 4°C overnight. Incubations with the antibodies were for 1 h atroom temperature, followed by 3×30 min washing with PBS. Thepeels were then mounted in Elvanol under a cover slip. Live cell imag-ing was carried out by mounting intact Arabidopsis leaves under acover glass, and performing time lapse image acquisition of XYZscanning. Images were acquired by an Olympus IX81/FV500 confocalmicroscope using an argon laser (488 nm), a green helium/neon laser(543 nm) and the following objectives; PLAPO60Xoil /NA 1.4 WD0.15 mm, or PLAPO60XW/LSM /NA1.00 WD 0.15 mm. Image analy-sis was carried out using the MICA software (Cytoview, Israel).

Supplementary materialSupplementary material mentioned in the article is available to

online subscribers at the journal website: www.pcp.oupjournals.org.

Acknowledgments

We thank Prof. Benjamin Geiger from the Department of Molec-ular Cell Biology, the Weizmann Institute of Science, Rehovot, Israel,for fruitful discussions and help.

References

Assmann, S.M. and Baskin, T.I. (1998) The function of guard cells does notrequire an intact array of cortical microtubules. J. Exp. Bot. 49: 163–170.

Assmann, S.M. and Schwartz, A. (1992) Synergistic effect of light and fusicoc-cin on stomatal opening. Planta 98: 1340–1355.

Assmann, S.M. and Shimazaki, K. (1999) The multisensory guard cell. Sto-matal responses to blue light and abscisic acid. Plant Physiol. 119: 809–815.

Assmann, S.M., Simoncini, L. and Schroeder, J.I. (1985) Blue light activateselectrogenic ion pumping in guard cell protoplasts of Vicia faba L. Nature318: 285–287.

Blatt, M.R. (2000) Cellular signaling and volume control in stomatal movementsin plants. Annu. Rev. Cell. Dev. Biol. 16: 221–241.

Eun, S.O. and Lee, Y. (1997) Actin filaments of guard cells are reorganized inresponse to light and abscisic acid. Plant Physiol. 115: 1491–1498.

Fedoroff, N.V. (2002) Cross-talk in abscisic acid signaling. Sci. STKE 140: re10.Fu, Y., Li, H. and Yang, Z. (2002) The ROP2 GTPase controls the formation of

cortical fine F-actin and the early phase of directional cell expansion duringArabidopsis organogenesis. Plant Cell 14: 777–794.

Fukuda, M., Hasezawa, S., Asai, N., Nakajima, N. and Kondo, N. (1998)Dynamic organization of microtubules in guard cells of Vicia faba L. withdiurnal cycle. Plant Cell Physiol. 39: 80–86.

Fukuda, M., Hasezawa, S., Nakajima, N. and Kondo, N. (2000) Changes intubulin protein expression in guard cells of Vicia faba L. accompanied withdynamic organization of microtubules during the diurnal cycle. Plant CellPhysiol. 41: 600–607.

Geitmann, A. and Emons, A.M. (2000) The cytoskeleton in plant and fungal celltip growth. J. Microsc. 198: 218–245.

Gorton, H.L., Williams, W.E. and Assmann, S.M. (1993) Circadian rhythms instomatal responsiveness to red and blue light. Plant Physiol. 103: 399–406.

Granger, C.L. and Cyr, R.J. (2001) Spatiotemporal relationships between growthand microtubule orientation as revealed in living root cells of Arabidopsisthaliana transformed with green-fluorescent-protein gene construct GFP-MBD. Protoplasma 216: 201–214.

Hetherington, A.M. (2001) Guard cell signaling. Cell 107: 711–714.Himmelspach, R., Wymer, C.L., Lloyd, C.W. and Nick, P. (1999) Gravity-

induced reorientation of cortical microtubules observed in vivo. Plant J. 18:449–453.

Hugdahl, J.D. and Morejohn, L.C. (1993) Rapid and reversible high-affinitybinding of the dinitroaniline herbicide oryzalin to tubulin from Zea mays L.Plant Physiol. 102: 725–740.

Jiang, C.-J., Nakajima, N. and Kondo, N. (1996) Disruption of microtubules byabscisic acid in guard cells of Vicia faba L. Plant Cell Physiol. 37: 697–701.

Johansson, F., Sommarin, M. and Larsson, C. (1993) Fusicoccin activates theplasma membrane H+-ATPase by a mechanism involving the C-terminalinhibitory domain. Plant Cell 5: 321–327.

Kinoshita, T. and Shimazaki, K. (1999) Blue light activates the plasma mem-brane H(+)-ATPase by phosphorylation of the C-terminus in stomatal guardcells. EMBO J. 18: 5548–5558.

Kinoshita, T., Doi, M., Suetsugu, N., Kagawa, T., Wada, M. and Shimazaki, K.(2001) Phot1 and phot2 mediate blue light regulation of stomatal opening.Nature 414: 656–660.

Kost, B. and Chua, N.H. (2002) The plant cytoskeleton: vacuoles and cell wallsmake the difference. Cell 108: 9–12.

Kost, B., Mathur, J. and Chua, N.H. (1999) Cytoskeleton in plant development.Curr. Opin. Plant Biol. 2: 462–470.

Lemichez, E., Wu, Y., Sanchez, J.P., Mettouchi, A., Mathur, J. and Chua, N.H.(2001) Inactivation of AtRac1 by abscisic acid is essential for stomatal clo-sure. Genes Dev. 15: 1808–1816.

Lin, C. (2002) Blue light receptors and signal transduction. Plant Cell 14Suppl.: S207–225.

Lloyd, C. and Hussey, P. (2001) Microtubule-associated proteins in plants-whywe need a MAP. Nat. Rev. Mol. Cell. Biol. 2: 40–47.

Guard cell microtubules582

Marc, J., Mineyuki, Y. and Palevitz, B.A. (1989a) The generation and consolida-tion of a radial array of cortical microtubules in developing guard cells ofAllium cepa L. Planta 179: 516–529.

Marc, J., Mineyuki, Y. and Palevitz, B.A. (1989b) A planar microtubule-organizing center in guard cells of Allium: experimental depolymerizationand reassembly of microtubules. Planta 179: 530–540.

Marc, J. and Palevitz, B.A. (1990) Regulation of the spatial order of corticalmicrotubules in developing guard cells of Allium. Planta 182: 626–634.

Marcus, A.I., Moore, R.C. and Cyr, R.J. (2001) The role of microtubules inguard cell function. Plant Physiol. 125: 387–395.

McDonald, A.R., Liu, B., Joshi, H.C. and Palevitz, B.A. (1993) Gamma-tubulinis associated with a cortical-microtubule-organizing zone in the developingguard cells of Allium cepa L. Planta 191: 357–361.

Moore, R.C. and Cyr, R.J. (2000) Association between elongation factor-1alphaand microtubules in vivo is domain dependent and conditional. Cell Motil.Cytoskeleton 45: 279–292.

Moore, R.C., Durso, N.A. and Cyr, R.J. (1998) Elongation factor-1alpha stabi-lizes microtubules in a calcium/calmodulin-dependent manner. Cell Motil.Cytoskeleton 41: 168–180.

Morejohn, L.C., Bureau, T.E., Mole-Bajer, J. and Fosket, D.E. (1987a) Oryza-lin, a dinitroaniline herbicide, binds to plant tubulin and inhibits microtubulepolymerization in vitro. Planta 172: 252–264.

Morejohn, L.C., Bureau, T.E., Tocchi, L.P. and Fosket, D.E. (1987b) Resistanceof Roza microtubules polymerization to colchicine results from a low affinityinteraction of colchicine and tubulin. Planta 170: 230–241.

Murashige, T. and Skoog, F. (1962) A revised medium for rapid growth and bio-assays with tobacco tissue culture. Physiol. Plant. 15: 473–497.

Ng, C.K.-Y., Mcainsh, M.R., Gray, J.E., Hunt, L., Leckie, C.P., Mills, L. andHetherington, A.M. (2001) Calcium-based signaling systems in guard cells.New Phytol. 151: 109–120.

Nick, P. (1998) Signaling to the microtubular cytoskeleton in plants. Int. Rev.Cytol. 184: 33–80.

Nick, P., Bergfeld, R., Schafer, E. and Schopfer, P. (1990) Unilateral reorienta-tion of microtubules at the outer epidermal wall during photo- and gravit-ropic curvature of maize coleoptiles and sunflower hypocotyls. Planta 181:162–168.

Palevitz, B.A. and Hepler, P.K. (1976) Cellulose microfibril orientation and cellshaping in developing guard cells of Allium: the role of microtubules and ionaccumulation. Planta 132: 71–93.

Schroeder, J.I., Allen, G.J., Hugouvieux, V., Kwak, J.M. and Waner, D. (2001a)Guard cell signal transduction. Annu. Rev. Plant Physiol. Plant Mol. Biol. 52:627–658.

Schroeder, J.I., Kwak, J.M. and Allen, G.J. (2001b) Guard cell abscisic acid sig-nalling and engineering drought hardiness in plants. Nature 410: 327–330.

Shibaoka, H. (1994) Plant hormone-induced changes in the orientation of corti-cal microtubules: Alterations in the cross linking between microtubules andthe plasma membrane. Annu. Rev. Plant Physiol. Plant Mol. Biol. 45: 527–544.

Shimazaki, K., Iino, M. and Zeiger, E. (1986) Blue light-dependent protonextrusion by guard-cell protoplasts of Vicia faba. Nature 319: 324–326.

Sugimoto, K., Williamson, R.E. and Wasteneys, G.O. (2000) New techniquesenable comparative analysis of microtubule orientation, wall texture, andgrowth rate in intact roots of Arabidopsis. Plant Physiol. 124: 1493–1506.

Ueda, K., Matsuyama, T. and Hashimoto, T. (1999) Visualization of microtu-bules in living cells of transgenic Arabidopsis thaliana. Protoplasma 206:201–206.

Volkmann, D. and Baluska, F. (1999) Actin cytoskeleton in plants: from trans-port networks to signaling networks. Microsc. Res. Tech. 47: 135–154.

Wasteneys, G.O. (2002) Microtubule organization in the green kingdom: chaosor self-order? J. Cell Sci. 115: 1345–1354.

Wasteneys, G.O., Willingale-Theune, J. and Menzel, D. (1997) Freeze shatter-ing: a simple and effective method for permeabilizing higher plant cell walls.J. Microsc. 188: 51–61.

Yu, R., Huang, R.F., Wang, X.C. and Yuan, M. (2001) Microtubule dynamics areinvolved in stomatal movement of Vicia faba L. Protoplasma 216: 113–118.

Yuan, M., Shaw, P.J., Warn, R.M. and Lloyd, C.W. (1994) Dynamic reorienta-tion of cortical microtubules, from transverse to longitudinal, in living plantcells. Proc. Natl Acad. Sci. USA 91: 6050–6053.

Zeiger, E. (2000) Sensory transduction of blue light in guard cells. Trends Plant.Sci. 5: 183–185.

(Received September 5, 2003; Accepted February 23, 2004)