Shade avoidance and the regulation of leaf inclination in Arabidopsis


Jack L. Mullen. Fax: +1 812 855 6082; e-mail:


As a rosette plant, Arabidopsis thaliana forms leaves near to the ground, which causes the plant to be vulnerable to shading by neighbours. One mechanism to avoid such shading is the regulation of leaf inclination, such that leaves can be raised to more vertical orientations to prevent neighbouring leaves from overtopping them. Throughout Arabidopsis rosette development, rosette leaves move to more vertical orientations when shaded by neighbouring leaves, exposed to low light levels or placed in the dark. After dark-induced reorientation of leaves, returning them to white light causes the leaves to reorient to more horizontal inclinations. These light-dependent leaf movements are more robust than, and distinct from, the diurnal movements of rosette leaves. However, the movements are gated by the circadian clock. The light-dependent leaf orientation response is mediated primarily through phytochromes A, B and E, with the orientation varying with the ratio of red light to far-red light, consistent with other shade-avoidance responses. However, even plants lacking these phytochromes were able to alter leaf inclination in response to white light, suggesting a role for other photoreceptors. In particular, we found significant changes in leaf inclination for plants exposed to green light. This green light response may be caused, in part, by light-dependent regulation of abscisic acid (ABA) biosynthesis.


Leaf orientation is regulated in many plant species, changing with development and in response to environmental cues. For example, a number of plants adjust their leaf orientations to track the sun (Wainwright 1977; Lang & Begg 1979). Many legumes and malvaceous plants demonstrate nyctinastic sleep movements in their leaves, which have been postulated to protect the plants from chilling injury (Darwin 1880; Smith 1974; Beck et al. 1982; Enright 1982) or aid in photoperiod measurement (Bünning & Moser 1969). Furthermore, in many tree species, leaves shaded because of a position low in the leaf canopy tend to be held in a more horizontal position than leaves experiencing full sun (McMillen & McClendon 1979), which is thought to facilitate light capture for photosynthesis in the canopy shade.

Plants respond to shading by neighbouring vegetation – which may compete with them for light – by sensing changes in light intensity and in the spectral quality of light in the red and far-red regions of the spectrum (Ballaré 1999). These changes in light quality are detected by the phytochrome family of photoreceptors and transduced to cause a series of changes in growth comprising the shade-avoidance syndrome, which includes increased internode elongation, increased petiole elongation, decreased leaf blade area, increased apical dominance and an accelerated flowering (for review, see Smith & Whitelam 1997; Ballaré 1999).

Although internode elongation is generally one of the more dramatic responses to shading, rosette plants, such as Arabidopsis, do not undergo substantial internode extension during the vegetative stage of growth. However, regulation of leaf positioning by light is one mechanism by which rosette plants can respond to neighbouring plants competing for light. Shade avoidance-related leaf positioning has been reported in Impatiens (Whitelam & Johnson 1982), tobacco (Hudson & Smith 1998; Pierik et al. 2004a) and Arabidopsis (Vandenbussche et al. 2003; Millenaar et al. 2005), with leaf inclination varying with the ratio of red light to far-red light. In Arabidopsis, mutants deficient in phytochrome are hyponastic (Ballaré & Scopel 1997; Vandenbussche et al. 2003). In addition to this regulation by light quality, leaf inclination is also dependent upon light intensity, with leaves in Arabidopsis becoming more horizontal with increasing light intensity (Hangarter 1997; Millenaar et al. 2005; Fig. 1). While this intensity-dependent response of rosette species is consistent with a role in shade avoidance, it differs from the response of many non-rosette species, in which leaves generally become more vertical with increasing light intensity as a protective mechanism against photodamage from excess light interception (King 1997; Valladares & Pugnaire 1999; Falster & Westoby 2003).

Figure 1.

Effect of light intensity on leaf inclination in Arabidopsis. Plants were placed for (a) 48 h in darkness, (b) 35 µmol m−2 s−1 white light or (c) 100 µmol m−2 s−1 white light. (d) Increased leaf inclination in Arabidopsis caused by overtopping by a leaf from a neighbouring plant. Arrow indicates the overtopped leaf. The leaf on top was pushed up by the upward growth of the lower leaf. The overtopping began 2 d before the image was taken.

Many plants that reorient their leaves do so by means of reversible changes in the volume of cells within a pulvinus. The molecular mechanisms controlling these turgor-driven volume changes are becoming increasingly well understood (Coté 1995; Koller 2000). However, many plants, including Arabidopsis, lack discrete pulvini, and in these plants, leaf movements are caused by differential growth in the leaves (Koller 2000; Millenaar et al. 2005). In order to gain a better understanding of the mechanisms of growth-dependent leaf movements, we have investigated the regulation of leaf positioning in Arabidopsis, focusing on the involvement of specific light-signalling pathways.


Plant material and growth conditions

Seeds of Arabidopsis thaliana (L.) Heynh. were sown on moist Scott’s Plug Mix (Scotts-Sierra, Marysville, OH, USA) in 75 mm pots and stratified at 4 °C for 2–3 d. Seeds of the phytochrome mutants phyA-201 and phyB-1 were provided by Professor R. Sharrock (Montana State University, Bozeman, MT, USA), and phyD-1, phyE-1 and the triple mutant were from Professor G. Whitelam (University of Leicester, UK). All phytochrome-deficient mutants were in the Landsberg erecta background. Seeds of hy1-100 were provided by Dr J. Chory (The Salk Institute, La Jolla, CA, USA). The phototropin mutants were in the Columbia background. The phototropins phot1-5 (Liscum & Briggs 1995; Huala et al. 1997) and phot2-1 (Kagawa et al. 2001) were provided by Dr E. Liscum and Dr T. Kagawa, respectively. Seed for the mutants cry1-1, cry2-1, npq1-1 and npq2-1 were obtained from the Arabidopsis Biological Resource Center at The Ohio State University, Columbus, OH, USA and are in the ecotype Columbia, except for cry1, which is in the Landsberg erecta background. Plants were grown under fluorescent lighting (≈100 µmol m−2 s−1) on a 12 h photoperiod. Plants were fertilized with K-grow All-Purpose Plant Food (K-mart, Troy, MI, USA) every 2 weeks and were used for experimentation when they were 4–5 weeks old. For experiments using abscisic acid (ABA), plants were watered daily with 100 mL of aqueous solution containing 0.1, 1.0 or 20 µm ABA; leaf inclination was measured after 4 d.

Measurement of leaf inclination

For most experiments, the leaf angle was determined trigonometrically from measurements of leaf length and height of the leaf tip above the surface of the growth medium. The value of leaf inclination is the angle of the leaf above horizontal, with the base of the petiole at the vertex. For the kinetic experiments, leaves were imaged using infrared illumination (940 nm LED, Radio Shack, Fort Worth, TX, USA) and a charge-coupled device camera (Marshall Electronics, Culver City, CA, USA) connected to a computer with a frame-grabber board (Imagenation Corp., Beaverton, OR, USA). Leaf inclination was measured from the images using Image Tool (a port of NIH Image, University of Texas Health Science Center, San Antonio, TX, USA). For the time-course experiments with mutants, only leaves that were very young at the outset (10–15 mm in length) were used, because leaf inclination also changes developmentally. For the other experiments, a larger range of leaves were measured (10–25 mm in length).

Light sources

White light was provided by cool white fluorescent lamps (F40CWRSWM, General Electric, Fairfield, CT, USA). Red light was supplied by a red light-emitting diode (LED) array (QBeam 2200 with lamp QB1310CS, Quantum Devices, Barneveld, WI, USA) and passed through a red Plexiglas filter (Rohm and Haas no. 2423; Dayton Plastics, Columbus, OH, USA). For blue and green illumination, light from halogen flood lamps (150 W Quartzline; General Electric) was filtered through 5 cm of a 1.5% (w/v) solution of CuSO4·5H2O and then through either a blue Plexiglas filter (Rohm and Haas no. 2045; Dayton Plastics) or a green interference filter (λ = 550 ± 25 nm, 03FIB008, Melles Griot, Rochester, NY, USA).


When leaves of Arabidopsis orient to more vertical inclinations under reduced irradiances, the changes in orientation are not only caused by changes in the inclination at which the leaves emerge from the stem but also by differential growth in the expanding leaves, particularly in the petiole, causing curvature along the leaves (Fig. 1a–b). This light-dependent change in leaf positioning allows the plant to respond to shading by proximal neighbours. When an Arabidopsis rosette leaf was overtopped by leaves from neighbouring plants, the inclination of the shaded leaf increased (Fig. 1d). This movement of the shaded leaf is not a result of physical contact with the overtopping leaf (data not shown), although the upward growth of the shaded leaf is robust enough to physically push the overtopping leaf up. This response also appears to be confined to the leaf where sensing occurs and not systemically spread, as overtopped leaves raise preferentially (Fig. 1d).

To understand the nature of these growth changes better, we examined the kinetics of the leaf movements following the removal of light. When placed in darkness at midday, changes in leaf inclination could be observed within ≈ 2 h, with an initial large increase in inclination followed by a slower phase of leaf raising (Fig. 2a). The latent period of the response depended on the timing of light removal. When plants were placed in darkness near the end of their normal photoperiod, the leaves remained near horizontal until the subjective morning, at which point they began rising towards vertical (Fig. 2a). The kinetics of these leaf movements show that this response is not a circadian movement, as the leaves generally continued to increase in inclination during the entire 24 h period. However, we did observe smaller diurnal changes in leaf positioning when plants were maintained in a 12 h photoperiod (Fig. 2b). In a diurnal pattern, leaf inclination gradually increased during the light period such that it was noticeably greater at subjective dusk compared with early in the light period. However, there was little change in inclination during the dark period until near the beginning of the next light period, when the leaves returned to near horizontal (Fig. 2b). The magnitude of the diurnal movements (≈ 15° change in inclination) was small compared with the acute response to light (≈ 50°); yet, the response to light appears to be gated by the circadian clock, because when leaves are placed in darkness near the end of the normal light period, they do not change inclination until the next expected light period (Fig. 2a).

Figure 2.

Kinetics of rosette leaf movement in response to light. (a) Response of leaves to light removal. The inclination of leaves was measured following the placement of plants in darkness either at the middle of day (●) or at the end of day (○) every 1 h for 24 h (mean ± SE, n = 8). (b) Diurnal changes in leaf orientation. Leaf inclination was measured every 2 h starting at the beginning of the 12 h photoperiod (mean ± SE, n = 6). White bars indicate light periods,; and dark bars indicate dark periods. Measurement of inclination was the angle of a leaf relative to horizontal, with more upright leaves having larger inclinations.

Because shade-avoidance responses are generally mediated by the phytochrome family of photoreceptors, we examined the leaf inclination of mutants deficient in individual phytochromes to determine which particular phytochromes are involved in controlling leaf inclination. At low fluence rates of white light (35 µmol m−2 s−1), young leaves are initially oriented at inclinations ≈ 40° from horizontal (Fig. 3), and as they develop, the leaves slowly become more horizontal. Young leaves of phyB are significantly more vertical in inclination than wild type [P < 0.05, one-way analysis of variance (anova) with Tukey post hoc test], and they are delayed in becoming more horizontal with age (Fig. 3a). In contrast, phyA and phyE plants had young leaves that were more horizontal than wild type (P < 0.05; Fig. 3a), suggesting an antagonistic relationship between these phytochromes and phyB. Leaves of phyD did not differ significantly in orientation from wild type.

Figure 3.

Leaf inclination of photoreceptor mutants. (a) The inclinations of young leaves of phytochrome-deficient mutants were followed for 6 d as they grew in 35 µmol m−2 s−1 continuous white light (mean ± SE, n = 10–15). (b) Average leaf inclination (mean ± SE, n = 10–18) in phototropin mutants. (c) Leaf inclination in cryptochrome mutants (mean ± SE, n = 10–18). Plants were placed in continuous illumination 2 d prior to inclination measurements. WT, wild type; Col, Columbia; Ler, Landsberg erecta.

To determine if there are other signalling pathways in the response – apart from that of the phytochromes – we examined phyB plants, as well as phyAphyBphyD and phyAphyBphyE triple mutant plants, for light-dependent changes in leaf inclination. Both phyB and the triple mutant plants had leaves which became more vertical when placed in darkness (P < 0.001; Fig. 4). This suggests that there is another signalling pathway involved in the leaf movements. In addition, the inclination of leaves of phyAphyBphyE in darkness was greater than that of wild-type leaves in darkness (P < 0.05), suggesting a possible role of phytochrome in regulating leaf inclination even in darkness. Conceivably, phyC could be the photoreceptor mediating leaf movement in the phytochrome triple mutant; however, in the phytochrome chromophore-deficient mutant hy1, we found a leaf inclination of 45 ± 2° (n = 12) for plants placed in 35 µmol m−2 s−1 white light and a leaf inclination of 63 ± 3° (n = 12) for plants in darkness, a significant difference (P < 0.01). Thus, it does not appear that phyC could explain the light-dependent leaf movements found in the phytochrome triple mutant.

Figure 4.

Light dependence of leaf inclination in phytochrome mutants. Plants were placed in darkness for 24 h prior to inclination measurements (mean ± SE, n = 11–19). The plants were then moved to 35 µmol m−2 s−1 white light for 24 h, and leaf inclination was measured again.

To investigate the role of blue light photoreceptors in the leaf movement response, we examined the leaf inclinations of mutants defective in individual members of both the phototropin and cryptochrome families of photoreceptors. Leaves of both phot1 and phot2 plants were more horizontal than wild type (P < 0.001; Fig. 3b). However, while the phototropin mutants had altered leaf positioning, the more horizontal inclination of the mutants is opposite of that expected if the phototropins were mediating the response. It seems more likely that the altered leaf inclination in the phototropin mutants was caused by a lack of leaf phototropism, as the illumination was from above the plants. The cryptochromes also appear not to play a major role in controlling leaf inclination, as the cry1 and cry2 mutants had leaf inclinations similar to wild-type plants (Fig. 3c).

To elucidate the role of the different photoreceptor families in the regulation of leaf inclination, we examined the inclination of leaves of plants illuminated with differing light qualities for 24 h. Consistent with a major role for phytochromes, leaves placed in red light (100 µmol m−2 s−1) had much lower inclinations than those in darkness (Fig. 5). However, leaves in red light had a higher inclination than those in an equivalent fluence rate of white light (P < 0.05), consistent with a proposed role for a photoreceptor apart from the phytochromes. In contrast, blue light had little effect on leaf orientation (Fig. 5). This is also consistent with the results from mutants deficient in blue light photoreceptors (Fig. 3b–c). Surprisingly, however, green light caused a significantly lowered inclination of leaves (P < 0.01; Fig. 5), raising the possibility that a photoreceptor that absorbs in the green range of wavelengths is involved in regulation of leaf inclination.

Figure 5.

Effect of light quality on leaf inclination. Plants were placed in darkness or 100 µmol m−2 s−1 light of the indicated quality for 24 h prior to inclination measurements (mean ± SE, n = 15–20).

Green light has been found to regulate stomatal aperture in Arabidopsis (Talbott et al. 2002). It has been suggested that this response is mediated by the carotenoid zeaxanthin (Frechilla et al. 2000; Talbott et al. 2003), as green light inhibition of stomatal opening is not observed in the zeaxanthin-deficient mutant npq1. Therefore, we examined whether the green light effect on leaf positioning also might be mediated by zeaxanthin. To this end, we measured leaf inclination in the mutants npq1 and npq2, which mediate the interconversion between violaxanthin and zeaxanthin (Niyogi, Grossman & Bjorkman 1998). Leaves of npq1, although deficient in zeaxanthin, were positioned similar to wild type, even in green light (Fig. 6). However, npq2 mutants had leaves which were more vertical in both white and green light (Fig. 6). Because npq2 encodes an enzyme in the ABA biosynthetic pathway, this suggested that green light may act by affecting ABA production. Consistent with this role for ABA in controlling leaf angle, npq2 plants treated with ABA responded by reducing the inclination of their leaves (Fig. 7). The effect was more dramatic in npq2 than in wild type, and at high ABA concentrations, some leaves were appressed against the soil, similar to the positioning of leaves under high fluence rates of light.

Figure 6.

Orientation of leaves of npq mutants in response to green light. Plants were placed in darkness or 100 µmol m−2 s−1 light of the indicated quality for 24 h prior to inclination measurements (mean ± SE, n = 8–20). WT, wild type.

Figure 7.

Effect of exogenous ABA application on leaf inclination. Leaf inclinations of WT and npq2 plants were measured following 4 d of treatment with the indicated concentration of ABA (mean ± SE, n = 8–14). Plants were grown under 35 µmol m−2 s−1 continuous white light during the treatment. ABA, abscisic acid; WT, wild type.


Arabidopsis plants respond to overtopping by increasing the inclination of their leaves, and this response can be initiated prior to the actual overtopping through the sensing of neighbouring plants (Fig. 1d). It is well established that light quality can provide a reliable cue of neighbour proximity and future competition for light. Light reflected from neighbouring plants has diminished levels of red light, relative to far-red light, because of chlorophyll absorption in the red wavelengths (Kasperbauer 1971; Holmes & Smith 1977). Because shifts in light quality occur in advance of direct shading, morphological responses to light quality, via phytochrome signalling, are hypothesized to provide a competitive advantage, allowing plants to establish a dominant position (Smith 1982; Ballaré, Scopel & Sanchez 1990; Schmitt 1997). For stem elongation, one of the more prominent shade-avoidance responses, individuals with greater internode elongation can achieve higher biomass and fitness than shorter plants (Dudley & Schmitt 1996; Weinig 2000), although elongation can be maladaptive when individuals are incapable of overtopping their neighbours (Weinig 2000). Arabidopsis does not undergo stem elongation during its vegetative stage of growth, so positioning of leaves may play an important role in competition for light. Increased variation in leaf inclination can increase light interception, although the effect is dependent on canopy arrangement (Niklas 1988). Thus, the increased leaf inclination may help maximize light interception, especially under the diffuse irradiance conditions of a shaded environment. In this regard, it is worth noting that in addition to the light quality effect, leaf inclination is also strongly dependent on fluence rate, although the response saturates with leaves appressed horizontally on the soil surface at fluence rates far below that of full sunlight (Fig. 1).

We found that the phytochrome most important in regulating leaf inclination in Arabidopsis rosettes was phyB (Fig. 3a), consistent with previous reports that phyB seedlings had increased leaf inclination (Ballaré & Scopel 1997; Vandenbussche et al. 2003). However, in cucumber, mutant plants deficient in phyB had normal leaf orientations (Ballaréet al. 1995). It may be that in cucumber, differences in interactions among different phytochromes allowed the plants to maintain normal leaf angles in the absence of phyB, as we found that phyA and phyE also affect leaf positioning in Arabidopsis. Such interactions among phytochromes might also explain differences in the light regulation of inclination in maize leaves where blue and far-red light, in addition to red light, also caused decreases in leaf angle (Fellner et al. 2003). In contrast, we found blue light to have no effect in Arabidopsis (Fig. 5), while far-red light caused leaves to orient at high inclinations similar to darkness in both Arabidopsis (Vandenbussche et al. 2003) and tobacco (Hudson & Smith 1998).

In tobacco, phyB was suggested to have a role in photoperiod-dependent circadian leaf movement rather than in acute shade avoidance-related leaf movement (Hudson & Smith 1998). Cotyledons of Arabidopsis also show strong circadian movements (Millar et al. 1995), and these movements are affected by phyB (Salome et al. 2002). However, while phyB may also have a similar role in Arabidopsis, circadian leaf movements in rosette leaves have much smaller amplitudes, so that the leaves maintain low inclinations throughout the night (Fig. 2). Yet, we have found that leaf orientation in phyB is greatly altered both under short days (Fig. 4 and data not shown) and when plants were grown under continuous illumination (Fig. 3), suggesting a more direct role in the shade-avoidance response of leaves. Moreover, the increase in leaf angle in response to light removal is gated by a circadian response. This gating was observed when plants moved to the dark in the middle of the subjective day showed a significantly faster reorientation response than when plants entered darkness at the end of the day (Fig. 2a). The rapid shade avoidance-induced stem-elongation response also shows regulation by circadian clock gates (Salter, Franklin & Whitelam 2003). Thus, while the robust light-dependent leaf movements in Arabidopsis rosettes differs from the circadian leaf movements observed for cotyledons and first true leaves (Dowson-Day & Millar 1999; Salome & McClung 2004), the movements are under circadian control.

Perhaps the most surprising finding of these studies is that in addition to the role of phytochrome in regulating leaf positioning, there appears to be another photoreceptor involved, as the phytochome mutants were still able to respond to changes in fluence rate (Fig. 4). Mutants deficient in phototropins had altered leaf inclination (Fig. 3b). However, the more horizontal inclination of the leaves of the phot mutants was likely caused by an impaired leaf phototropic response (data not shown), as the illumination in these experiments was from above. Moreover, blue light alone did not cause a decrease in leaf inclination compared with darkness (Fig. 5). This is in contrast with tobacco, where blue light reduced leaf inclination (Pierik et al. 2004b). However, under green light, Arabidopsis rosette leaves oriented to a significantly lower inclination than those placed in darkness. This response might be mediated by weak absorption of green light by phytochrome, as has been suggested for other green light responses (Mandoli & Briggs 1981). However, this would not explain the response in phytochrome mutants or the difference in leaf inclination between red and white illumination. Because carotenoids of the xanthophyll cycle have been implicated in other green light responses (Frechilla et al. 2000; Talbott et al. 2003), we examined mutants deficient in these carotenoids for changes in leaf inclination. We found altered leaf positioning in the npq2 mutant, but not in npq1 (Fig. 6), which suggested that regulation of ABA biosynthesis may be important in this response. Consistent with this idea, application of ABA decreased leaf inclination (Fig. 7). ABA has also been found to be involved in hyponastic growth of submerged leaves of Rumex (Cox et al. 2004). In addition to ABA, several other hormones have been implicated in leaf positioning (Vandenbussche et al. 2003; Pierik et al. 2004a). Therefore, the fine-tuning of leaf position in response to environmental and developmental signals is likely caused by complex interactions among these hormones.

Although no green light-specific photoreceptor has yet been identified in plants, a number of green light responses have been described (Klein 1964, 1965; Frechilla et al. 2000; Talbott et al. 2003; Folta 2004; Kim et al. 2004a and b). Because the phytochrome and cryptochrome photoreceptors can absorb some green light, it has been difficult to rule out a role for them in green light responses. However, the demonstration that green light can stimulate hypocotyl elongation in various photoreceptor mutants of Arabidopsis provides strong evidence for the presence of a green light photoreceptor that differs in action from known photoreceptors (Folta 2004). Our analysis of the green light regulation of leaf orientation represents another light response that may be dependent on an unidentified green light photoreceptor. Moreover, our results with the xanthophyll cycle mutant npq2 suggest that instead of xanthophylls serving as chromophores for absorbing green light (Eisinger et al. 2003; Talbott et al. 2002), it may be their function as precursors for ABA production that is important for this leaf orientation response. The data presented here demonstrate that regulation of leaf orientation is a complex response that is dependent on successful integration of multiple photoreceptors, hormonal signals and gating by the circadian clock.


This work was funded by the US National Aeronautics and Space Administration (NNA04-CC55G).