Skip to main content
NCBI home page
The Plant Cell logoLink to The Plant Cell
. 2012 Jan 31;24(1):192–201. doi: 10.1105/tpc.111.093849

The Influence of Light on Microtubule Dynamics and Alignment in the Arabidopsis Hypocotyl[W]

Adrian Sambade a,1, Amitesh Pratap b,1, Henrik Buschmann a, Richard J Morris b, Clive Lloyd a,2
PMCID: PMC3289555  PMID: 22294618

This article used mutants in the light/gibberellin pathway to examine microtubules under dark-like conditions and found that light inhibits the polymerization and reorientation of microtubules at the onset of growth.

Abstract

Light and dark have antagonistic effects on shoot elongation, but little is known about how these effects are translated into changes of shape. Here we provide genetic evidence that the light/gibberellin–signaling pathway affects the properties of microtubules required to reorient growth. To follow microtubule dynamics for hours without triggering photomorphogenic inhibition of growth, we used Arabidopsis thaliana light mutants in the gibberellic acid/DELLA pathway. Particle velocimetry was used to map the mass movement of microtubule plus ends, providing new insight into the way that microtubules switch between orthogonal axes upon the onset of growth. Longitudinal microtubules are known to signal growth cessation, but we observed that cells also self-organize a strikingly bipolarized longitudinal array before bursts of growth. This gives way to a radial microtubule star that, far from being a random array, seems to be a key transitional step to the transverse array, forecasting the faster elongation that follows. Computational modeling provides mechanistic insight into these transitions. In the faster-growing mutants, the microtubules were found to have faster polymerization rates and to undergo faster reorientations. This suggests a mechanism in which the light-signaling pathway modifies the dynamics of microtubules and their ability to switch between orthogonal axes.

INTRODUCTION

The development of plants is plastic, their form modulated by environmental factors that influence the rate and direction of growth. Light and dark have major effects on morphogenesis and induce alternative developmental strategies (de Lucas et al., 2008). In the dark, the embryonic shoot uses energy stored in the seed to undergo rapid elongation, conveying the seed leaves up through the soil. But once in the light, and once the unfolded cotyledons become photosynthetic, the switch to autotrophic growth is accompanied by the photomorphogenic inhibition of hypocotyl elongation. The protein interaction framework that coordinates opposing effects of light and dark on the length of Arabidopsis thaliana hypocotyls is increasingly clear (Reed et al., 1993; de Lucas et al., 2008). This involves positive and negative regulation of transcription factors that mediate the effects of light perception and the growth-promoting effects of gibberellins (GAs). Despite the increasing clarity of these signaling pathways, very little is known about the way in which the activities of these molecules are transcribed into higher-order changes of shape. Ultimately, these pathways must influence the self-organizing properties of cytoplasmic microtubules that align the shape-controlling cellulose microfibrils in the cell wall.

Generally, transverse microtubules are believed to accompany rapid elongation, whereas a switch to longitudinal alignment accompanies growth inhibition (Lloyd, 2011). However, a major barrier to studying living microtubules in rapidly growing cells is that the light necessary for visualizing the plants invokes growth inhibition (Gendreau et al., 1997; Le et al., 2005). Indeed, the confocal laser beam can be experimentally used for stopping growth and inducing microtubules to reorient to the longitudinal (Paredez et al., 2006). Numerous previous studies on dark-grown shoots have therefore relied upon the use of fixatives to provide a snapshot of microtubule alignment in dark-grown material (reviewed in Lloyd, 2011). However, growth is not uniform along the Arabidopsis hypocotyl (Gendreau et al., 1997; Le et al., 2005), frustrating general correlations between microtubule alignment and the rate of expansion. In addition to being transverse, it has been reported that microtubules in shoots can be oblique, random, or even longitudinal in supposedly elongating cells, and this has tended to undermine a role for microtubules in guiding the growth of shoots (Lloyd, 2011).

To see microtubules undergoing rapid elongation in living cells, we used complementary strategies to mitigate the effects of light under the confocal microscope. This involved the use of long-hypocotyl mutants and overcoming light-inhibition with GA.

The red/far-red light photoreceptor, phyB, is the major phytochrome in Arabidopsis, occupying an important step in the light-signaling pathway. In this study, we used the phyB-1 (Koorneef et al., 1980; Reed et al., 1993) mutant, which has a long hypocotyl phenotype. Elongation of hypocotyls is promoted by a family of phytochrome-interacting transcription factors (PIFs), but their growth-promoting effects are antagonized by phyB and the GA-suppressive DELLA proteins (Khanna et al., 2007; de Lucas et al., 2008; Lorrain et al., 2008). To abrogate this repression, we used seedlings in which PIF5 was overexpressed (Lorrain et al., 2008). In addition, exogenous GA annuls growth inhibition by destabilizing DELLA (de Lucas et al., 2008), so this phytohormone was added to wild-type seedlings as an alternative means of increasing the length of hypocotyls under the light microscope (Duckett and Lloyd, 1994). Finally, we studied another mutant, hy1, based on disruption of the HY1 locus involved in phytochrome biosynthesis (Muramoto et al., 1999). This mutation also produces a long-hypocotyl phenotype in the light, ranging in length between that of phyB-1 and the wild type.

Using long-hypocotyl seedlings expressing the microtubule plus-end marker, EB1a-green fluorescent protein (GFP), combined with long-term time-lapse confocal microscopy (Chan et al., 2007; Buschmann et al., 2010), it has been possible to follow the development of microtubule alignment over time in actively elongating hypocotyl epidermal cells. A key finding is that the longitudinal array not only marks the end of growth but also occurs in young cells and precedes bursts of growth. We used computational modeling to demonstrate how the striking bipolarization of this array predisposes the formation of a microtubule star, which mediates the switch to the transverse array favoring rapid elongation. Mutants show that light inhibits the speed of this transition. Light also inhibits the speed of microtubule polymerization, and because treadmilling microtubules play such an important part in the movement and insertion of wall precursors (Crowell et al., 2009), this inhibition provides an explanation for the negative effects of light on cell expansion.

RESULTS

Different Regions of the Hypocotyl Respond Differently to Light

Initially, to provide a detailed survey of the timing and the position along the hypocotyl at which elongation occurred, Arabidopsis seedlings were placed on a cold stage of a Phillips/FEI XL30 field electron gun–scanning electron microscope. This allowed highly accurate measurements to be made of the lengths of epidermal cells along the entire hypocotyl at different time points over days 2 to 5 (Figure 1), defining the regions to be later studied in living cells under the fluorescence microscope. This showed that in the Arabidopsis ecotype Landsberg erecta (Ler) wild type, there was little growth at days 2 and 3, but by day 4, cells were considerably longer in the lower part of the hypocotyl (Figure 1B). In the mutants (phyB, hy1) and the PIF5 overexpressor (PIFox), by contrast, this burst of expansion was seen to occur 1 d earlier. In addition, expansion continued further into the upper portion of the hypocotyl (Figure 1C). It is known that photomorphogenic inhibition of growth is stronger in this upper one-half of the wild-type Arabidopsis hypocotyl (Gendreau et al., 1997; Le et al., 2005), which is where phyB, hy1, and PIFox mutant hypocotyls are found to continue to elongate. We therefore selected cells from two hypocotyl regions: cells six to nine, as measured from the base of the hypocotyl, and cells 12 to 14 from the more light-sensitive region nearer the shoot apex.

Figure 1.

Figure 1.

Open in a new tab

Hypocotyl Elongation in Constant Light.

(A) Seedling growth over days 2 to 4. The inset shows a scanning electron micrograph of a 2-d-old Ler EB1a-GFP seedling; the main figure shows a seedling at day 4.

(B) The x axis shows cell number from bottom of hypocotyl, and the y axis shows cell length in micrometers.

(C) The Ler hy1mutant (EB1-GFP). Note that the hy1 mutant in (C) grows longer than the wild-type control in (B) and that cells in the upper part of the mutant hypocotyl continue to elongate at day 4.

Bar in (A) = 500 μm.

A Microtubule Star Marks the Onset of Growth

Because elongation was discontinuous, it was possible to capture the initiation of growth in long-term movies made from projections of z-stacks (Chan et al., 2007) (Figure 2A). At the onset of growth, cells underwent a defined sequence of microtubule realignments in which a radial array of microtubules, or “star” (Figure 2), formed a convenient marker for the onset of growth. Stars were clearly discernible in phyB, hy1, and PIFox, and in wild-type ecotypes (Columbia [Col-0] and Ler) expressing EB1a-GFP, as well as in Col-0 wild-type cells expressing GFP-tubulin (Figure 2C). Sometimes, microtubule tracks underwent rotations as previously described for light-grown hypocotyls (Chan et al., 2007), but these could occur after or during growth, unlike the star that reproducibly marked the imminent onset of growth. The significance of the isotropic radial star, specifying no direction in particular, is that it is transitional between the two orthogonal longitudinal and transverse axes.

Figure 2.

Figure 2.

Open in a new tab

Microtubule Alignment during the Onset of Growth.

(A) Montage of time projections (3 min each) obtained at 20-min intervals in a time-lapse series made using LerEB1a-GFP. To the left, kymographs of the top and bottom of the cell over this time period show 1) that the time-lapse series is taken before growth starts and 2) that the beginning of growth (arrow) is marked by the first signs of microtubule star formation.

(B) The same process of star formation in the faster-growing Ler-hy1-EB1a:GFP line (note the rapid elongation in the kymograph). Microtubules become more transversely aligned after star formation. Frames are separated by 20 min.

(C) Snapshots of microtubule star-like arrays in the wild type (WT), hy1, phyB, and cry1 mutants and PIFox plants expressing EB1a:GFP. The Col-0 wild type expresses GFP-TuA.

Bars in (C) = 5 μm (except for GFP-tubulin, where bar = 10 μm).

To check that these patterns were not induced by blue light from the laser beam, we used the cryptochrome mutant cry1 in the Ler background; cry1 is compromised in blue light detection and is synergistic with phyB for cell growth (Sellaro et al., 2009). cry1 mutants expressing EB1a-GFP showed the same sequence of events, including the star (Figure 2C), as described for the phytochrome mutants.

Supplemental Movie 1 online shows that the appearance of microtubule stars was synchronized between neighboring cells. In darkness, but not in light, a tipward wave of elongation occurs along the Arabidopsis hypocotyl (Gendreau et al., 1997). The appearance of stars may therefore be part of the physiological mechanism to reset the microtubule network for growth.

The Longitudinal Cortical Array Is Strongly Bipolarized

To analyze the flow pattern of numerous fluorescent microtubule plus ends as the array’s axis shifted, we adapted a particle image velocimetry (PIV) method used for following particle flow (Keane and Adrian, 1992; Raffel et al., 2007) (see Image Analysis in Supplemental Text 1 and Supplemental Figures 1 and 2 online). In roots, the end of growth is marked by the reorientation of microtubules from transverse to oblique and finally to longitudinal (Granger and Cyr, 2001), but our findings demonstrate that microtubules in the hypocotyl can also be longitudinal before the onset of growth and are therefore not dead-end states.

PIV and kymographic analysis revealed another novel feature in that the longitudinal array is strongly bipolarized (Figure 3A; see Supplemental Figure 3 online). Microtubules are known to form domains sharing the same polarity (Dixit et al., 2006) that can migrate around the cell (Chan et al., 2007), but here the microtubule plus ends reproducibly segregate out in a fixed manner to opposite poles (Figure 3B; see Supplemental Movie 2 online). That is, most microtubules in the upper one-half of the cell moved toward the shoot apex, whereas those in the lower one-half moved toward the root. This was seen in all mutants and ecotypes (Figures 3B to 3D). Analysis of 93 cells in nine hypocotyls, including 5BA, col0-EB1, hy1-EB1, phyB-EB1, and PIFox-EB1, showed that at 2 to 3 d, 31 cells had longitudinal microtubules, and all of them were bipolarized. As well as being present in the longitudinal arrays that preceded growth, bipolarization was also detected in the longitudinal arrays that form after growth has finished, although the extent was attenuated.

Figure 3.

Figure 3.

Open in a new tab

Bipolarization and Stars in Early Growing Stages.

(A) to (D) Bipolarized longitudinal microtubule array in LerEB1a-GFP. (A) Kymographic analysis. The red arrows show that microtubule comets move in opposite directions, toward opposing cell ends, and crossover in the mid-region. This is confirmed in (B), where particle velocimetry plots the angle of EB1 comet movement over a 6-min period (sampling rate of 10 s per frame), showing domains moving to opposite poles. The color scale represents the direction, in degrees, in which EB1 comets are moving. Zero represents movement to the right, −90 represents movement to the top, and 90 represents movement to the bottom. In (C), the bipolarized movement of comets shown in (B) are quantified; the x axis shows the scale in degrees, and the y axis shows the number of measurements normalized to 1.0. The inset shows the velocity distribution in micrometers per minute. (D) Divergence analysis based on PIV calculations showing EB1 flow per unit area (see Image Analysis in Supplemental Text 1 online) along the longitudinal axis. The lower values at the cell ends confirm bipolarization of the microtubule array.

(E) to (H) The same analysis as above but for the long hypocotyl mutant Ler-hy1-EB1a-GFP. The star configuration is shown to share the underlying bipolarization (E).

(I) to (M) Effect of growth interference on microtubule alignment. (I) Frame from Supplemental Movie 3 online of cells from Ler-wild-type-EB1a-GFP seedling treated with NPA. Cessation of growth arrests the majority of cells at the star stage. (J) and (K), which are the two asterisked cells in (I), show stills from the movie, demonstrating that under growth arrest, the star reverts to longitudinal. (L) Successive time points from movie of a phyB cell showing the transition from longitudinal to star to transverse as the cell noticeably grows. (M)Time-lapse sequence of a wild-type plant treated with LatB and NPA, showing that the bipolarized longitudinal array is maintained in the presence of an auxin transport and the actin polymerization inhibitor, LatB.

Bars = 5 μm.

Microtubule Stars Initiate the Reorganization for Growth

To investigate whether bipolarization was linked to growth, we used the polar auxin-transport inhibitor, 1-N-naphthylphthalamic acid (NPA), on wild-type hypocotyls. NPA is known to inhibit the elongation of light-grown but not etiolated Arabidopsis hypocotyls (Jensen et al., 1998). Our observations suggested that stars were intermediates in the transition from the longitudinal to the transverse array in elongating epidermal cells (Figure 3L). But when NPA stopped growth, the reorientation sequence progressed no further than the star stage (Figure 3I). This is consistent with the idea that stars occur in the transitional stage immediately prior to growth and therefore do not require growth for their formation. It also shows that it is not the early emergence of microtubules from the long sidewalls that initiates the transverse array (for some occur in stars) but an additional step. Supplemental Movie 3 online shows that, under NPA arrest, stars and longitudinal arrays interchange synchronously without completing the sequence to transverse as seen in the wild type. For example, a still from that movie (Figure 3I) illustrates 18 out of 25 cells with stars. Figures 3J and 3K show that under these conditions the star reverts to a longitudinal array. The NPA result also suggests bipolarization is independent of polar auxin flux, consistent with findings that microtubule reorganization does not depend on auxin transport (Heisler et al., 2010).

To test whether the axiality of these longitudinal microtubules was dependent on longitudinal actin cables, NPA-treated hypocotyls were additionally treated with latrunculin B (LatB), which depolymerizes actin filaments and stops cytoplasmic streaming in Arabidopsis hypocotyl epidermal cells grown under these conditions (Chan et al., 2007). LatB had no effect on alignment or bipolarity of longitudinal microtubule arrays (Figure 3M).

Modeling Provides Mechanistic Insight into Bipolarization, Star Formation, and the Transition to Transverse

To see whether bipolarization, as exemplified by Supplemental Movie 2 online, could be explained as the outcome of microtubule–microtubule interactions, we used rules of interaction pioneered by Dixit and Cyr (2004), in which shallow contacts favor coalignment (“zippering”) and steeper collisions tend to result in depolymerization (see Simulations in Supplemental Text 1 online). The dynamicity of individual microtubules was modeled as a two-state Markov process (see Simulations in Supplemental Text 1, Supplemental Figure 4, and Supplemental Table 1 online). In the bipolarized longitudinal array, most microtubules are seen to treadmill toward, and very few grow from, the short end walls (Figures 3A and 3E). This regional reduction in growing microtubule ends is clearly visible as the microtubule-poor “end zones” in transverse arrays (Lucas et al., 2011) and is also seen here (see Supplemental Movie 1 online). Addition to the model of end zones in which no microtubules were nucleated was sufficient to simulate the formation of the bipolarized longitudinal array (Figures 4A to 4C). Once this array was established, the simulation reached a quasi–steady state (see Supplemental Figure 5 online). In living cells, the next transition to the radial star was seen to be initiated, not by increased speed of microtubule growth (Figure 5C), but by increasing numbers of microtubules entering the outer cell face from the long sidewalls (Figure 5B). In the bipolarized array, very few microtubules were observed entering the outer wall from the subtending long sidewalls (Figure 5B). To simulate this in the model, we inhibited the entry of microtubules from the long sidewalls. To generate the star from the bipolarized array, it was sufficient to relax the near total inhibition from 98 to 85% (Figures 4D to 4F) (see Methods). Reversing this back to the near total inhibition reestablished the bipolarized array, mirroring the oscillation between star and bipolarized array seen during NPA treatment.

Figure 4.

Figure 4.

Open in a new tab

Modeling Bipolarization and Star Formation.

(A) The conditions required for modeling the bipolarized array. As indicated by the arrows, microtubules initiated on the field are free to leave the field, but few enter, as shown in Supplemental Movie 2 online. The boxes at the end zones indicate a decreased probability of microtubule nucleation, based on a Gaussian distribution along the length of the cell.

(B) A frame from the computer simulation. Each point is an arrowhead (visible at magnification), showing the resulting segregation of polarity based on angle-dependent microtubule-microtubule interactions.

(C) The velocity history encoded by colors, in which red indicates polymerization to the right, and blue indicates polymerization to the left.

(D) Schematic showing the increased number of microtubules entering from the sidewall to model the star.

(E) A frame from the simulation; red denotes microtubule plus ends moving downward yellow denotes the microtubule plus ends moving upward.

(F) A z-stack of frames showing the final stages of star formation (the last 40 frames from 250).

Figure 5.

Figure 5.

Open in a new tab

Microtubule Transitions.

(A) Time-lapse images showing the transition from star to transverse (at 20-min intervals) (speeds in micrometers per minute). Arrows indicate the spreading of transverse microtubules toward the cell ends.

(B) Number of sidewall microtubules (MT) before (T0) and after (T60) wedge formation (bars denote se).

(C) PIV analysis of Ler-phyB-EB1a-GFP showing homogeneous comet speed in a star.

(D) Speed of triangular wedges crossing cells in Ler and Col-0 wild type (WT) and phytochrome mutants hy1, phyB, and PIFox.

(E) Kymographs of plus-end comets in the wild type, phyB, and PIFox.

(F) PIV-calculated microtubule polymerization speed in wild-type plants and mutants.

In movies of living cells, increased numbers of oblique and transverse microtubules could be seen emerging from the long sidewalls during the transition from the longitudinal to the transverse axis, over and above the relatively small number of such microtubules observed during star formation (Figure 5B; see Supplemental Movie 1 online). This step-increase in long sidewall microtubules formed triangular wedges that tended to focus toward the center of the bipolarized array to produce the star-like configuration. In the star, analysis revealed that microtubules radiating toward the end walls remained unipolar, whereas the microtubule bundles associated with the sidewalls were bidirectional (Figures 3E and 3F). Next, Figure 5A illustrates a transition from the star to the transverse array. The increased densities of transverse/oblique microtubules in the center of the cell face spread toward the end walls to complete formation of the transverse array.

This transition to transverseness could be reproduced in all 10 simulation runs by relaxing the inhibition of microtubules entering from the long sidewalls from the 85% applied to star formation down to 30%. Running the model, this release of additional microtubules from the long sidewalls overcame the longitudinal bias still present in the star, resulting in a stable transverse array. Therefore, three states identified in living cells can be modeled by altering the numbers of microtubules passing onto the outer cell face from the long sidewalls: 1) the few such microtubules observed in the longitudinal bipolarized array, 2) the modest increase that produces the star, and 3) the greater increase that converts the star to the transverse array.

Light Inhibits Microtubule Dynamics

Microtubule transitions were quantified by measuring the speed at which asymmetric wedges spread across the cell to form the transverse array. Image analysis showed that light mutants underwent faster transitions (Figure 5D). In addition to investigating transitions in knockout mutants (phyB, hy1), we used the PIF-overexpressing line (Ler-PIFox), in which the light-regulatory steps of the DELLA pathway are superseded. Kymography of the EB1 comets (Figure 5E) indicated that the speed at which wedges moved during the transition (as seen in Figure 5D) correlated with the speed at which individual microtubules grew. To quantify the behavior of individual microtubules in greater depth, we used PIV analysis (Figure 5F), in which ~500,000 to 800,000 grid points were used to sample the speeds of the individual comets (see Image Analysis in Methods). This produced the same trend as observed by kymography and confirmed that microtubules in hy1, phyB, and PIFox had a higher polymerization rate than controls. These results, produced with knockout and overexpression lines, signify a connection between the light-signaling pathway and microtubule polymerization (Figures 5E and 5F).

GA Restarts the Microtubule Sequence Required for Growth

To see whether the bipolarized-longitudinal/star/transverse sequence was related to the induction of growth irrespective of its trigger, we selected 4-d-old cells in regions 12 to 14. The growth of these cells is known to be reinitiated by dark treatment (Jensen et al., 1998) (Figures 6A to 6C), but to induce growth in the light we added gibberellic acid (GA3) (Duckett and Lloyd, 1994; Lucas et al., 2011). GA3 relieves the blockade of hypocotyl elongation caused by the GA-DELLA–signaling pathway responsible for coordinating the light response (de Lucas et al., 2008). Figures 6D and 6E illustrate the resulting elongation response. These cells, stimulated by exogenous hormone, retraced the longitudinal/star/transverse sequence observed in the light mutants (cf. Figure 6D with Figure 2B).

Figure 6.

Figure 6.

Open in a new tab

Restarting Cell Growth in the Dark and under GA3 Treatment.

(A) The inset shows a scanning electron microscopy of a Col-0 EB1a-GFP seedling grown for 5 d in the light. The main figure shows a seedling that had been grown in the light for 4 d and then transferred to the dark for a further 24 h.

(B) Average length of hypocotyls shown in micrometers with se. Left bar denotes seedlings grown 5 d in the light, right bar denotes seedlings grown 4 d in the light then 1 d in the dark.

(C) Hypocotyl cell length at day 5. The x axis shows cell number from bottom of hypocotyl, and the y axis shows cell length in micrometers.

(D) Microtubule alignment in the same cells before (T0) and after (T180 min) adding exogenous GA3 (10 μM).

(E) Individual hypocotyls grown in the light were measured at day 4 then remeasured at day 5, with the last day’s growth shown as a percentage. The right bar shows the enhancement of growth produced by adding 10 μM GA3 for the last day.

Bar in (A) = 300 μm.

DISCUSSION

We present new findings that clarify the process of microtubule reorientation in hypocotyls and offer mechanistic insight into the growth-suppressing effects of light.

PIV analysis of bulk plus-end flow led to the surprising finding that longitudinal arrays, long known to be markers of growth deceleration in the root (Granger and Cyr, 2001), can precede bouts of growth acceleration in the shoot. Modeling shows that interaction between treadmilling microtubules can sort out ordered parallel arrays of microtubules (Dixit and Cyr, 2004). Catastrophic depolymerization of microtubules meeting at steep angles, and coalignment (zippering) of microtubules meeting at shallower angles, explains how microtubules gain their parallel order, but further explanations are required to account for the bipolarization of the longitudinal array. Our data indicate that the longitudinal axis can be simulated by the addition to opposite cell ends of zones in which few microtubules are initiated. These microtubule-poor end zones have recently been noted by Lucas and colleagues (Lucas et al., 2011), and we also find them in living cells here. Adding such end zones to simulations provides the missing factor required to convert randomly oriented microtubules into a bipolarized array aligned parallel to the cell’s long axis. A similar phenomenon has just been reported in Drosophila oocytes, where microtubule initiation along the anterolateral cortex and its suppression at the posterior end accounts for the biased accumulation of microtubule plus ends toward the posterior pole (Parton et al., 2011). The main difference between that system and Arabidopsis is that we detect a strong reduction in microtubule initiation at both ends of the hypocotyl epidermal cells, consistent with the bipolarization of the longitudinal array.

This setting of the long axis by the bipolarized array establishes one of the two key vectors that characterize the binary nature of plant cell expansion: axial elongation versus lateral swelling. The modeling also indicates how the bipolarized longitudinal array, once established, is predisposed to form the radial star, because increased numbers of transverse microtubules are funneled toward the center of the bipolarized array. Although stars were not detected during rapid elongation, close analysis of movies revealed that it was the star that accurately marked the transition to rapid elongation.

The next stage in the reorientation process involved dense wedges of transverse microtubules that formed across the equator then spread toward the ends of the cells. Stopping growth with NPA stopped progression beyond the star stage, suggesting that growth itself may provide important vectorial information for the development of transverseness out of the star stage. Consistent with this, the concentration of NPA effective in stopping the growth of wild-type seedlings was unable to fully block the expansion of faster-growing phyB hypocotyls, which are reported to be less sensitive to NPA (Jensen et al., 1998). This reduced amount of growth was nevertheless sufficient for transverse arrays to develop out of the star stage in NPA (Figure 3L). The computer simulations show that a modest increase in the number of microtubules entering the outer cell face from the subtending anticlinal walls is sufficient to convert the bipolarized longitudinal array to a star. To complete the transition to the transverse array, it was necessary to increase the number of microtubules entering the outer face from the long sidewalls. Therefore, both modeling and live cell observations suggest that by modulating the number of microtubules entering from the long sidewalls, the cell is able to effect discrete transitions (from longitudinal to star, star to longitudinal, and star to transverse), consistent with different growth states.

Hypothetically, this could be caused by physiological changes resulting in decreased microtubule catastrophe or increased nucleation and/or stability. One possibility, recently reported by Ambrose et al. (2011), is that the microtubule-associated protein CLASP accumulates at specific cell edges, in a tunable manner, to overcome depolymerization of microtubules at sharp boundaries.

The NPA results suggest that growth is required for the full transition to the transverse array. This raises the question of how expansion, or conversely its inhibition, could provide the coordinates for microtubule alignment and its 90° realignment during the growth cycle. Cell expansion has long been discussed to involve physical forces whose anisotropy is a function of the cell’s shape (Green, 1962; Hejnowicz et al., 2000); in an expanding cylinder, transverse stress is double that of the longitudinal stress. Hejnowicz and colleagues (Hejnowicz et al., 2000) hypothesize that microtubule reorientation is based on the interaction between the transverse stress of the individual turgid cell and the longitudinal tensile stress of the constricting epidermis, which must be cyclically relaxed to allow organ growth. That physical forces can play a role in plant morphogenesis was recently supported by studies showing how microtubules in the Arabidopsis shoot apex align along lines of principal stress (Hamant et al., 2008).

Numerous studies on the orientation of microtubules in shoots (reviewed in Lloyd, 2011) have failed to show a clear-cut correlation between microtubule alignment and cell elongation, raising doubts about whether microtubules can play a universal role in plant growth. Microtubules have been reported to be oblique, random, and longitudinal in addition to the transverse orientation predicted to accompany cell elongation (Green, 1962). However, it is now realized that not all shoot epidermal cells are elongating all the time (Gendreau et al., 1997; Le et al., 2005; this study), perhaps explaining the previous lack of correlation between microtubule alignment and the phase of rapid elongation. However, by accurately measuring the onset of elongation in movies of rapidly growing light mutants, it has been possible to demonstrate a clear sequence of reorientation in preparation for expansion. The sequence is, however, unexpected, because longitudinal arrays are found at the beginning of reorientation, and the random-looking radial star can now be seen to play a key part in the transition between the two main axes.

Having established the realignment sequence, these observations reveal a mechanism by which light regulates growth via its effect on microtubule dynamics. Cortical microtubules are far from static, undergoing treadmilling, branching, and minor realignments even when the overall array is maintaining its net alignment; for example, transverse (Shaw et al., 2003; Dixit and Cyr, 2004; Chan et al., 2009). This dynamicity is essential for the formation and movement of membrane-associated cellulose synthase complexes whose trafficking is blocked when microtubule turnover is blocked with taxol (Crowell et al., 2009). This suggests close coupling between microtubule dynamics and delivery of materials essential for new growth. Our observations show that the rate of microtubule polymerization is reduced in light-grown epidermal cells, providing an explanation for their reduced expansion during a finite growing phase. Recent studies indicate that cellulose-synthesizing complexes (CSCs) may be physically associated with microtubules via cellulose synthase-interactive protein1 (CSC1), a protein linker (Gu et al., 2010) whose interaction is sensitive to phosphorylation (Bischoff et al., 2011). The latter article also reported that the velocity of CSCs in the dark is inhibited by an interaction with microtubules that is overturned by the photoreceptor PHYB. All of this points toward an emerging mechanism whereby light can modulate the cellulose/microtubule complex that is the basis of directional plant growth. Our results show how the orientation of microtubules may be an inherent part of that mechanism.

METHODS

Plant Material

Arabidopsis thaliana seeds were sterilized using commercial bleach, chilled 2 d at 4°C, and grown in Murashige and Skoog medium as described (Chan et al., 2007), except that seedlings were transferred to observation chambers before microscopy. Arabidopsis Col-0 plants expressing GFP-TUA6 and EB1a-GFP, both driven by 35S promoter, are described in Chan et al. (2005). EB1a-GFP in Ler background is described in Buschmann et al. (2009). Mutant plants affected in light perception GA-DELLA pathway, displaying the long hypocotyl phenotype, phyB-1 (Reed et al., 1993), hy1 (Muramoto et al., 1999), and cry1 (European Arabidopsis Stock Centre; ID number N70) were crossed with EB1a-GFP. F3 double homozygous plants were used. F1 seeds were obtained by crossing the 35S:PIF5 overexpressor in the Ler background (courtesy of Dr. Philip Wigge, John Innes Centre) with EB1a-GFP in Ler.

Microscopy

Hypocotyls were imaged using a VisiTech spinning disc confocal microscope using either a 40/1.3 or a 60/1.4 numeric aperture oil-objective lens. GFP was excited using the 488-nm diode laser and emitted light filtered through a 500- to 550-nm band-pass filter. The microscope is equipped with a Hamamatsu ORCA-ER cooled charge-coupled device camera with 1 binning. Four-dimensional time-lapse series was collected and projected (max) using the MetaMorph Imaging System (www.moleculardevices.com). Time-lapse images were acquired either at low (four to six images every 30 s) or high (every 10 s) sampling rate with a time delay of 15 to 20 min (Buschmann et al., 2010).

Scanning electron microscopy was used to examine cell lengths in 2- to 5-d-old seedling hypocotyls. Seedlings were placed on a cold stage of a Phillips/FEI XL30 field electron gun–scanning electron microscope fitted with a cryostage.

Image Analysis

Cell measurements and kymographs were made using the multikymograph macro from ImageJ (http://reb.info.nih.gov/ij/) written by J. Rietdorf and A. Seitz at EMBL. Long-term movies of growing hypocotyls were aligned using Stackreg (http://bigwww.epfl.ch/thevenaz/stackreg/) plug-ins of ImageJ. Movies were edited using the brightness and contrast tool of ImageJ and the Time Stamper plug-in.

Supplemental Data

The following materials are available in the online version of this article.

Acknowledgments

We thank Philip Wigge, John Innes Centre, for advice and for reading the manuscript and Ceinwen Tilley, University of Leicester, for supplying seeds.

AUTHOR CONTRIBUTIONS

C.L. initiated the project, and R.J.M. proposed the modeling approach. A.S. performed all biological data gathering and, together with H.B., designed the mutant strategy. A.S., C.L., and H.B. contributed to conception, design, and interpretation of the cellular aspects. A.P. adapted the PIV method for this project and performed the computer simulations, which R.J.M. supervised. The article was drafted by C.L. and A.S. and was modified and approved by A.P., R.J.M., and H.B.

References

  1. Ambrose C., Wasteneys G.O. (August 16, 2011). A CLASP-modulated cell edge barrier mechanism drives cell-wide cortical microtubule organization in Arabidopsis. Nat. Commun. 2 (online), http://dx.doi.org/10.1038/ncomms1444 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bischoff V., Desprez T., Mouille G., Vernhettes S., Gonneau M., Hofte H. (2011). Phytochrome regulation of cellulose synthesis in Arabidopsis. Curr Biol. 21: 1822–1827 [DOI] [PubMed] [Google Scholar]
  3. Buschmann H., Hauptmann M., Niessing D., Lloyd C.W., Schäffner A.R. (2009). Helical growth of the Arabidopsis mutant tortifolia2 does not depend on cell division patterns but involves handed twisting of isolated cells. Plant Cell 21: 2090–2106 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Buschmann H., Sambade A., Pesquet E., Calder G., Lloyd C.W. (2010). Microtubule dynamics in plant cells. Methods Cell Biol. 97: 373–400 [DOI] [PubMed] [Google Scholar]
  5. Chan J., Calder G., Fox S., Lloyd C. (2005). Localization of the microtubule end binding protein EB1 reveals alternative pathways of spindle development in Arabidopsis suspension cells. Plant Cell 17: 1737–1748 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chan J., Calder G., Fox S., Lloyd C. (2007). Cortical microtubule arrays undergo rotary movements in Arabidopsis hypocotyl epidermal cells. Nat. Cell Biol. 9: 171–175 [DOI] [PubMed] [Google Scholar]
  7. Chan J., Sambade A., Calder G., Lloyd C. (2009). Arabidopsis cortical microtubules are initiated along, as well as branching from, existing microtubules. Plant Cell 21: 2298–2306 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Crowell E.F., Bischoff V., Desprez T., Rolland A., Stierhof Y.D., Schumacher K., Gonneau M., Höfte H., Vernhettes S. (2009). Pausing of Golgi bodies on microtubules regulates secretion of cellulose synthase complexes in Arabidopsis. Plant Cell 21: 1141–1154 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. de Lucas M., Davière J.M., Rodríguez-Falcón M., Pontin M., Iglesias-Pedraz J.M., Lorrain S., Fankhauser C., Blázquez M.A., Titarenko E., Prat S. (2008). A molecular framework for light and gibberellin control of cell elongation. Nature 451: 480–484 [DOI] [PubMed] [Google Scholar]
  10. Dixit R., Chang E., Cyr R. (2006). Establishment of polarity during organization of the acentrosomal plant cortical microtubule array. Mol. Biol. Cell 17: 1298–1305 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dixit R., Cyr R. (2004). Encounters between dynamic cortical microtubules promote ordering of the cortical array through angle-dependent modifications of microtubule behavior. Plant Cell 16: 3274–3284 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Duckett C.M., Lloyd C.W. (1994). Gibberellic acid-induced microtubule reorientation in dwarf peas is accompanied by rapid modification of an α-tubulin isotype. Plant J. 5: 363–372 [Google Scholar]
  13. Gendreau E., Traas J., Desnos T., Grandjean O., Caboche M., Höfte H. (1997). Cellular basis of hypocotyl growth in Arabidopsis thaliana. Plant Physiol. 114: 295–305 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Granger C.L., Cyr R.J. (2001). Spatiotemporal relationships between growth and microtubule orientation as revealed in living root cells of Arabidopsis thaliana transformed with green-fluorescent-protein gene construct GFP-MBD. Protoplasma 216: 201–214 [DOI] [PubMed] [Google Scholar]
  15. Green P.B. (1962). Mechanism for plant cellular morphogenesis. Science 138: 1404–1405 [DOI] [PubMed] [Google Scholar]
  16. Gu Y., Kaplinsky N., Bringmann M., Cobb A., Carroll A., Sampathkumar A., Baskin T.I., Persson S., Somerville C.R. (2010). Identification of a cellulose synthase-associated protein required for cellulose biosynthesis. Proc. Natl. Acad. Sci. USA 107: 12866–12871 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hamant O., Heisler M.G., Jönsson H., Krupinski P., Uyttewaal M., Bokov P., Corson F., Sahlin P., Boudaoud A., Meyerowitz E.M., Couder Y., Traas J. (2008). Developmental patterning by mechanical signals in Arabidopsis. Science 322: 1650–1655 [DOI] [PubMed] [Google Scholar]
  18. Heisler M.G., Hamant O., Krupinski P., Uyttewaal M., Ohno C., Jönsson H., Traas J., Meyerowitz E.M. (2010). Alignment between PIN1 polarity and microtubule orientation in the shoot apical meristem reveals a tight coupling between morphogenesis and auxin transport. PLoS Biol. 8: e1000516. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hejnowicz Z., Rusin A., Rusin T. (2000). Tensile tissue stress affects the orientation of cortical microtubules in the epidermis of sunflower hypocotyl. J. Plant Growth Regul. 19: 31–44 [DOI] [PubMed] [Google Scholar]
  20. Jensen P.J., Hangarter R.P., Estelle M. (1998). Auxin transport is required for hypocotyl elongation in light-grown but not dark-grown Arabidopsis. Plant Physiol. 116: 455–462 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Keane R.D., Adrian R.J. (1992). Theory of cross-correlation analysis of PIV images. Appl. Sci. Res. 49: 191–215 [Google Scholar]
  22. Khanna R., Shen Y., Marion C.M., Tsuchisaka A., Theologis A., Schäfer E., Quail P.H. (2007). The basic helix-loop-helix transcription factor PIF5 acts on ethylene biosynthesis and phytochrome signaling by distinct mechanisms. Plant Cell 19: 3915–3929 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Koorneef M., Rolff E., Spruit C.J.P. (1980). Genetic control of light-inhibited hypocotyl elongation in Arabidopsis thaliana (L.) Heynh. Zeitschrift fur Pflanzenphysiologie 100: 147–160 [Google Scholar]
  24. Le J., Vandenbussche F., De Cnodder T., Van der Straten D., Verbelen J.-P. (2005). Cell elongation and microtubule behavior in the Arabidopsis hypocotyl: Responses to ethylene and auxin. J. Plant Growth Regul. 24: 166–178 [Google Scholar]
  25. Lloyd C. (2011). Dynamic microtubules and the texture of plant cell walls. Int Rev Cell Mol Biol 287: 287–329 [DOI] [PubMed] [Google Scholar]
  26. Lorrain S., Allen T., Duek P.D., Whitelam G.C., Fankhauser C. (2008). Phytochrome-mediated inhibition of shade avoidance involves degradation of growth-promoting bHLH transcription factors. Plant J. 53: 312–323 [DOI] [PubMed] [Google Scholar]
  27. Lucas J.R., Courtney S., Hassfurder M., Dhingra S., Bryant A., Shaw S.L. (2011). Microtubule-associated proteins MAP65–1 and MAP65–2 positively regulate axial cell growth in etiolated Arabidopsis hypocotyls. Plant Cell 23: 1889–1903 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Muramoto T., Kohchi T., Yokota A., Hwang I., Goodman H.M. (1999). The Arabidopsis photomorphogenic mutant hy1 is deficient in phytochrome chromophore biosynthesis as a result of a mutation in a plastid heme oxygenase. Plant Cell 11: 335–348 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Paredez A.R., Somerville C.R., Ehrhardt D.W. (2006). Visualization of cellulose synthase demonstrates functional association with microtubules. Science 312: 1491–1495 [DOI] [PubMed] [Google Scholar]
  30. Parton R.M., Hamilton R.S., Ball G., Yang L., Cullen C.F., Lu W., Ohkura H., Davis I. (2011). A PAR-1-dependent orientation gradient of dynamic microtubules directs posterior cargo transport in the Drosophila oocyte. J. Cell Biol. 194: 121–135 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Raffel M., Willert C., Wereley S., Kompenhans J. (2007). Particle Image Velocimetry: A Practical Guide (Berlin: Springer-Verlag; ). [Google Scholar]
  32. Reed J.W., Nagpal P., Poole D.S., Furuya M., Chory J. (1993). Mutations in the gene for the red/far-red light receptor phytochrome B alter cell elongation and physiological responses throughout Arabidopsis development. Plant Cell 5: 147–157 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Sellaro R., Hoecker U., Yanovsky M., Chory J., Casal J.J. (2009). Synergism of red and blue light in the control of Arabidopsis gene expression and development. Curr. Biol. 19: 1216–1220 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Shaw S.L., Kamyar R., Ehrhardt D.W. (2003). Sustained microtubule treadmilling in Arabidopsis cortical arrays. Science 300: 1715–1718 [DOI] [PubMed] [Google Scholar]

Articles from The Plant Cell are provided here courtesy of Oxford University Press

RESOURCES