Antiphase Light and Temperature Cycles Affect … · Ethylene application restored leaf growth in 2DIF conditions, and constitutive ethylene signaling mutants maintain robust leaf

Antiphase Light and Temperature Cycles AffectPHYTOCHROME B-Controlled Ethylene Sensitivityand Biosynthesis, Limiting Leaf Movement and Growthof Arabidopsis1[C][W]

Ralph Bours, Martijn van Zanten, Ronald Pierik, Harro Bouwmeester, and Alexander van der Krol*

Laboratory of Plant Physiology, Wageningen University, 6708 PB Wageningen, The Netherlands (R.B., H.B.,A.v.d.K.); and Plant Ecophysiology, Institute of Environmental Biology (M.v.Z., R.P.), and Molecular PlantPhysiology (M.v.Z.), Utrecht University, 3584 CH Utrecht, The Netherlands

ORCID ID: 0000-0001-6585-7572 (A.v.d.K.).

In the natural environment, days are generally warmer than the night, resulting in a positive day/night temperature difference(+DIF). Plants have adapted to these conditions, and when exposed to antiphase light and temperature cycles (cold photoperiod/warm night [2DIF]), most species exhibit reduced elongation growth. To study the physiological mechanism of how light andtemperature cycles affect plant growth, we used infrared imaging to dissect growth dynamics under +DIF and 2DIF in the modelplant Arabidopsis (Arabidopsis thaliana). We found that 2DIF altered leaf growth patterns, decreasing the amplitude and delayingthe phase of leaf movement. Ethylene application restored leaf growth in 2DIF conditions, and constitutive ethylene signalingmutants maintain robust leaf movement amplitudes under 2DIF, indicating that ethylene signaling becomes limiting under theseconditions. In response to 2DIF, the phase of ethylene emission advanced 2 h, but total ethylene emission was not reduced.However, expression analysis on members of the 1-aminocyclopropane-1-carboxylic acid (ACC) synthase ethylene biosynthesisgene family showed that ACS2 activity is specifically suppressed in the petiole region under2DIF conditions. Indeed, petioles ofplants under 2DIF had reduced ACC content, and application of ACC to the petiole restored leaf growth patterns. Moreover,acs2 mutants displayed reduced leaf movement under +DIF, similar to wild-type plants under 2DIF. In addition, wedemonstrate that the photoreceptor PHYTOCHROME B restricts ethylene biosynthesis and constrains the 2DIF-inducedphase shift in rhythmic growth. Our findings provide a mechanistic insight into how fluctuating temperature cycles regulateplant growth.

In nature, during the day (light), temperatures areusually higher than during the night (dark). Correspond-ingly, most plants show optimal growth under such syn-chronous light and temperature cycles. Increasing thedifference between day and night temperature (+DIF)results in increased elongation growth in various species,a phenomenon referred to as thermoperiodism (Went,1944). The opposite regime, when the temperature ofthe day (DT) is lower than the temperature of the night(NT), is called 2DIF (negative DT/NT difference).Under 2DIF conditions, the elongation growth of stemsand leaves of various plant species is reduced (Maas and

van Hattum, 1998; Carvalho et al., 2002; Thingnaes et al.,2003). Arabidopsis (Arabidopsis thaliana) plants grownunder2DIF (DT/NT 12°C/22°C) displayed a reductionin leaf elongation of approximately 20% compared withthe control (DT/NT 22°C/12°C; Thingnaes et al., 2003).2DIF is frequently applied in horticulture to producecrops with a desirable compact architecture without theneed for growth-retarding chemicals (Myster and Moe,1995). Despite the economic importance of the appli-cation of such temperature regimes in horticulture, themechanistic basis of the growth reduction under 2DIFis still poorly understood.

Previously, it was demonstrated that 2DIF affectsphytohormone signaling in plants. In pea (Pisum sativum),for instance, the 2DIF growth reduction correlated withincreased catabolism of the phytohormone GA (Stavanget al., 2005). In contrast to pea, active GA levels did notdecrease in response to2DIF in Arabidopsis (Thingnaeset al., 2003). On the other hand, the 2DIF growth re-sponse in Arabidopsis was associated with reducedauxin levels (Thingnaes et al., 2003). The photoreceptorPHYTOCHROME B (PHYB) has been shown to beimportant for the response to 2DIF, as phyB mutantsof Arabidopsis (Thingnaes et al., 2008) and cucum-ber (Cucumis sativus; Patil et al., 2003) are insensitiveto 2DIF.

1 This work was supported by the Top Technological InstituteGreen Genetics (grant no. 2CFL009RP to A.v.d.K.) and by the Nether-lands Organization for Scientific Research (VENI grant no. 863.11.008to M.v.Z.).

* Address correspondence to [email protected] author responsible for distribution of materials integral to the

findings presented in this article in accordance with the policy de-scribed in the Instructions for Authors (www.plantphysiol.org) is:Alexander van der Krol ([email protected]).

[C] Some figures in this article are displayed in color online but inblack and white in the print edition.

[W] The online version of this article contains Web-only data.www.plantphysiol.org/cgi/doi/10.1104/pp.113.221648

882 Plant Physiology�, October 2013, Vol. 163, pp. 882–895, www.plantphysiol.org � 2013 American Society of Plant Biologists. All Rights Reserved.

https://plantphysiol.orgDownloaded on March 22, 2021. - Published by Copyright (c) 2020 American Society of Plant Biologists. All rights reserved.

http://orcid.org/0000-0001-6585-7572

mailto:[email protected]

http://www.plantphysiol.org

mailto:[email protected]

http://www.plantphysiol.org/cgi/doi/10.1104/pp.113.221648

https://plantphysiol.org

In this work, the growth-related movement of matureArabidopsis rosette leaves was analyzed under control(+DIF) and 2DIF conditions. Under 2DIF, the ampli-tude of leaf movement was decreased and the phase ofmovement was later, compared with control plants. Thealtered leaf growth patterns observed in 2DIF couldbe restored by the application of ethylene. 2DIFreduced the expression of 1-AMINOCYCLOPROPANE-1-CARBOXYLIC ACID SYNTHASE2 (ACS2) in the petiole,which correlated with reduced 1-aminocyclopropane-1-carboxylic acid (ACC) levels and decreasedamplitude and delayed phase of leaf movement. Ourresults indicate that local ACS activity plays an impor-tant biological role, despite the fact that ethylene is agaseous and fast-diffusing hormone. In addition,we demonstrate that in the phyB9 mutant, the phaseof leaf movement is almost fully temperatureentrained. Finally, ethylene levels and sensitivity areincreased in phyB9, suggesting a role for PHYB inconstraining temperature-induced shifts in the phase ofleaf movement and dampening of leaf movementamplitude by controlling ethylene production andsensitivity.

RESULTS

Leaf Movement and Elongation Growth Are Reducedduring the Cold Photoperiod

To characterize the growth of Arabidopsis plants inresponse to antiphase light and temperature cycles(2DIF), we used the OSCILLATOR growth monitor-ing system, which enables accurate analysis of phase,amplitude, and period of growth-related leaf move-ments (Bours et al., 2012). Plants were pregrown under+DIF for 4 weeks, after which temperature cycles wereeither kept identical (control) or reversed to the op-posite 2DIF regime. The plants that were exposed to2DIF showed a visually lower leaf angle comparedwith the control plants (Fig. 1A), suggesting a differ-ence in diurnal leaf movement rhythm. Top-view imagesof the plants were recorded during 1 week of growthunder control and2DIF conditions (Supplemental MovieFile S1) and used for the measurement of projected leaflengths (Supplemental Fig. S1A) from which the pro-jected oscillations were extracted (Fig. 1B). These pro-jected oscillations were analyzed for peak amplitudeand for the timing of the peak amplitude relative to the

Figure 1. Rhythmic growth and movement of leaves under control and 2DIF conditions. A, Two 36-d-old Arabidopsis (Col-0)rosette plants exposed to 2DIF (top) or control (+DIF; bottom) conditions and photographed at the end of the fourth photo-period. B, Projected oscillations for leaves developing 7 d under control (solid line) or 2DIF (dashed line) conditions. SE valuesare depicted per 20-min time point as shading; n = 8 leaves. Gray areas indicate the dark period. C, Closeup of B for day 4,depicting an example of amplitude, phase, and phase shift for control leaves compared with2DIF. Error bars represent SE; n = 8.D, Average amplitudes of days 2 to 6 calculated from the projected oscillations in B for leaves developing under control (blackbar) or 2DIF (gray bar) conditions. Error bars represent SE; n = 8. E, The phase of oscillations for leaves developing under 2DIFconditions (dashed line, white circles) shifts compared with the control (solid line, black triangles). Error bars represent SE; n = 8.[See online article for color version of this figure.]

Plant Physiol. Vol. 163, 2013 883

Ethylene Regulates Thermoperiodic Leaf Movement


http://www.plantphysiol.org/cgi/content/full/pp.113.221648/DC1




onset of the photoperiod, here referred to as the “phase”of the leaf movement rhythms (Fig. 1C). This analysisconfirmed that the visually lower leaf angle observedunder 2DIF (Fig. 1A) corresponds to reduced leafmovement amplitude relative to the control (Fig. 1D).Moreover, also the phase of movement shifted under2DIF compared with the control. Under control con-ditions, the phase of movement for small emergingleaves during day 1 is approximately 19 h (where t = 0h is dawn and t = 12 h is dusk). After 2 d, when theleaves matured, the phase of leaf movement stabilizedat around t = 16 h (Fig. 1E). In contrast, in response to2DIF, the phase of leaf movement shifted from t = 19 hduring the first day to t = 21 h, which persistedthroughout the next 6 d. Under2DIF, the timing of thepeak amplitude thus shifts to late night, whereas thelowest leaf position is measured at mid day. Conse-quently, the phase of movement of leaves developingunder 2DIF is delayed by 5 h compared with thecontrol (Fig. 1E).

Taken together, under 2DIF, upward movement islargely restricted to the warm night, whereas undercontrol conditions, the largest part of the upward leafmovement occurs during the warm day (Fig. 1B). Thissuggests that under alternating diurnal light andtemperature cycles, temperature, rather than light,contributes most to the leaf movement of Arabidopsis.

Analysis of the projected leaf length curves showedthat the reduced leaf movement during the day under2DIF coincides with decreased (projected) leaf elon-gation (Supplemental Fig. S1A), suggesting a relationship

between leaf movement and elongation. Measurementof absolute lengths of leaves developed under 2DIFconfirmed reduced elongation (Supplemental Fig. S1B).

Ethylene Affects the Amplitude and Phase of DiurnalLeaf Movement

It was previously demonstrated that the gaseousplant hormone ethylene is involved in upward leafmovement (hyponastic growth) in response to differ-ent stresses in Arabidopsis (Millenaar et al., 2005, 2009;van Zanten et al., 2010). Application of the ethyleneperception inhibitor silver thiosulfate (STS) to plantsgrowing under control +DIF conditions reduced leafmovement amplitude by approximately 50% com-pared with mock treatment (Fig. 2, A and B). Noadditional inhibitory effect on amplitude was notedwhen STS was applied to leaves developing under2DIF (Supplemental Fig. S2, A and B). The ethylene-insensitive mutants ethylene response1-1 (etr1-1) andethylene insensitive2-1 (ein2-1) showed a strongly re-duced leaf movement amplitude under control +DIFconditions (Fig. 2, A and C) as well as under 2DIFconditions (Supplemental Fig. S2, A and C). The am-plitude of leaf movement in these mutants undercontrol conditions is similar to that of wild-type plantsunder 2DIF (Fig. 1D) or wild-type plants treated withSTS under control conditions (Fig. 2A). STS applicationto leaves developing under control conditions initiallyshifted the phase of leaf movement from t = 16 h (mock

Figure 2. Ethylene signaling controls amplitudeand phase of diurnal leaf movement. A, Averageamplitudes (days 2–6) calculated from the pro-jected oscillations (in B and C) for leaves treatedwith mock (gray bar) or 50 mM STS (black bar) andfor Col-0 (light gray bar), etr1-1 (black bar), andein2-1 (dark gray bar) developing under controlconditions. Error bars represent SE; n = 8. B, Pro-jected oscillations of Col-0 leaves under controlconditions treated with 50 mM STS (solid line)compared with mock (dashed line). Error barsrepresent SE; n = 8. Gray areas indicate the darkperiod. C, Projected oscillations for leaves ofetr1-1 (solid black line) and ein2-1 (solid grayline) compared with the Col-0 wild type (blackdashed line) under control conditions. Error barsrepresent SE; n = 8. Gray areas indicate the darkperiod. D, Phase shifts of leaf movements be-tween control and 2DIF conditions for leavestreated with mock or 50 mM STS and for Col-0,ein2-1, and etr1-1. Significant phase shifts (P ,0.05) are indicated with arrows, and nonsignifi-cant shifts are indicated with bars. Each arrowdepicts the direction and strength of the shift inphase: the start of the arrow indicates averagephase during days 2 to 6 under control condi-tions, and the arrowhead indicates the averagephase for2DIF (days 2–6). Error bars represent SE;n = 8.

884 Plant Physiol. Vol. 163, 2013

Bours et al.







treatment) to t = 20 h (STS; Fig. 2D; Supplemental Fig.S2D). This mimics the phase shift induced by 2DIF foruntreated plants (Fig. 2D). Three days after the start ofthe treatment, the phase of STS-treated leaves undercontrol conditions changed back to 18 h, indicatingthat the effect of a single STS application is transient(Supplemental Fig. S2D), which is expected, as novelsynthesis of ethylene receptors renders the plant sen-sitive again. Under 2DIF, STS application only causeda small phase shift from t = 21 h (mock) to t = 22 h(STS; Fig. 2D). In the ethylene-insensitive mutantsetr1-1 and ein2-1, the estimated phase did not shiftbetween the +DIF control and 2DIF conditions andwas around t; 20 to 22 h (Fig. 2D). Thus, the estimatedphase of leaf movement of these ethylene-insensitivemutants resembled the late phase of wild-type leafmovement under 2DIF (Fig. 2D). The differences be-tween the STS treatment and the ethylene-insensitivemutants (Fig. 2D) may arise in part due to ethylene-independent effects of STS on auxin (Strader et al., 2009).

Ethylene and Ethylene Signaling Restore Leaf Growth andMovement under 2DIF

Because our results indicate that ethylene signalingis involved in controlling the amplitude, phase, andelongation response of leaves to diurnal temperaturecycles, we reasoned that ethylene signaling or ethylenelevels may be limiting for leaf movement under 2DIF.To test this hypothesis, plants were sprayed with dif-ferent concentrations of the ethylene-releasing com-pound ethephon (2-chloroethylphosphonic acid) at thestart of the first 2DIF photoperiod (t = 0 h), and leafmovement was subsequently monitored over a periodof 7 d. Consistent with our hypothesis, a single ethe-phon application was sufficient to restore leaf movementamplitude under 2DIF, in a dose-responsive manner,to a level comparable to untreated control plants (Fig. 3,A and B). Ethephon application only marginally en-hanced the amplitude of leaf movement under controlconditions, and this effect became visible only 3 d aftertreatment (Supplemental Fig. S3, A and B). Subse-quently, we tested whether the loss-of-function mutantetr1-7 and the triple mutant etr1-6 etr2-3 ein4-4, whichboth have constitutive ethylene signaling phenotypes(Hua and Meyerowitz, 1998; Cancel and Larsen, 2002),are insensitive to the inhibitory effect of 2DIF on leafmovement and growth. Indeed, both mutants showeda robust and persistent amplitude of leaf movementunder 2DIF conditions, comparable to that of 0.5 mM

ethephon-treated wild-type plants under 2DIF (Fig. 3,A and C; Supplemental Fig. S3C) or wild-type plantsunder control conditions (Fig. 2A). The single appli-cation of 0.5 mM ethephon to wild-type plants resultedin an almost biphasic growth-related movement of thetreated leaves. After processing of the data in OSCIL-LATOR, we used the highest peak value as a phaseindicator. The results indicate a modest phase shiftof leaf oscillation in response to ethephon treatment

under both 2DIF conditions (from mock t = 21 h toethephon t = 22 h; Fig. 3, A and D) and control con-ditions (from mock t = 16 h to ethephon t = 15 h;Fig. 3D; Supplemental Fig. S3B). The phase of leafmovement of the constitutive ethylene signaling mu-tants etr1-7 and etr1-6 etr2-3 ein4-4 was similar to thatof the wild type (t = 16 h) under control conditionsand, in contrast to the wild type, did not shift betweencontrol and 2DIF (Fig. 3, B and D). Taken together,these data indicate that ethylene stimulates both theamplitude and phase of leaf movement under 2DIF.

Quantification of the leaf length of ethephon-treatedplants showed that ethylene can also partly restore theinhibited leaf elongation under 2DIF (SupplementalFig. S3, E and F). In contrast, ethephon treatment ofplants developing under control conditions did notaffect leaf elongation (Supplemental Fig. S3, E and F),suggesting that ethylene is only limiting for leaf elon-gation under 2DIF.

These results may indicate reduced ethylene actionunder 2DIF. Therefore, we measured real-time ethyl-ene emissions of wild-type Columbia-0 (Col-0) plantsdeveloping under control and2DIF conditions (Fig. 4A).Ethylene accumulated in the headspace showed diur-nal oscillations under both control and2DIF conditions(Fig. 4A). The absolute amount of ethylene producedper gram of shoot fresh weight, however, was not re-duced for plants grown under2DIF compared with thecontrol (Fig. 4A). However, since 2DIF-grown plantshave an approximately 40% reduction of fresh weight,total emissions per plant are reduced under 2DIF. Thephase of peak ethylene emission for control plants wasat t = 6 h, while the phase of plants developing under2DIF was at 2 h and thus shifted 4 h backward com-pared with the control (Fig. 4A). Subsequently, ACSreporter plants were analyzed to determine the effectof 2DIF on local ethylene production in plants.

ACS2 Activity Is Altered under 2DIF

The rate of ethylene synthesis is dependent on theproduction of the biological precursor ACC by ACSenzymes (Kende, 1993). The Arabidopsis ACS genefamily contains eight functional members (Yamagamiet al., 2003), which can form a multitude of functionalhomodimers and heterodimers (Tsuchisaka et al., 2009).Furthermore, different ACS members display uniquespatial expression patterns (Tsuchisaka and Theologis,2004) and may respond differentially to 2DIF. To testif and which ACS gene(s) are differentially expressedin 2DIF versus control conditions, we compared GUSpresence in seven different promoter ACS::GUS re-porter lines (Tsuchisaka and Theologis, 2004). In five ofthese lines (ACS4, ACS5, ACS6, ACS9, and ACS11),GUS expression did not change in response to 2DIFconditions (Supplemental Fig. S4A). However, theACS2:GUS and ACS8:GUS reporter plants showedmarkedly different expression patterns under 2DIFconditions (Fig. 4, B and C).















In control conditions, ACS2 promoter activity wasvisible in the proximal side of the leaf, the petiole, andmidrib veins, while in leaves developed under 2DIF,the activity in veins and petioles was strongly reducedor absent (Fig. 4B). The ACS8 promoter showed in-creased GUS expression in the transverse edges of theblade (Fig. 4C). ACS8 was shown to be negativelyregulated by ethylene (Thain et al., 2004), and the in-creased activity under2DIF might indicate that ethylene-mediated repression of ACS8 is relieved. Because ACS2showed differential activity in the petiole, we testedwhether the activity of ACS2 in the petiole relates to afunction in leaf movement. To this aim, we analyzedthe diurnal leaf movement of two independent ACS2loss-of-function alleles (acs2-1 and acs2-2). Indeed, bothmutants showed reduced leaf movement amplitudeunder control and 2DIF conditions (Fig. 4, D and E).The phase of leaf movement under control conditionswas delayed in the two mutants compared with thewild type under control +DIF and was comparable to

the phase of wild-type plants under 2DIF (Fig. 4F). Inresponse to 2DIF, the phase in the acs2 mutants didnot change in contrast to the 5-h phase shift in wild-typeleaves under 2DIF (Fig. 4F). Analysis of the projectedand absolute leaf lengths confirmed that both mu-tants had reduced leaf lengths under both conditions(Supplemental Fig. S4, B–E). These results indicatethat ACC production by ACS2 in the petiole stronglycontributes to leaf elongation and the phase and am-plitude of diurnal leaf movement and that the effectsof 2DIF on growth are mediated through effects onACS2 activity.

In order to relate ACS2 activity to local ACC levels,petioles of both wild-type and acs2-1 plants developedin control and 2DIF conditions were analyzed forACC content. In wild-type plants, the ACC levels inthe petioles were reduced in response to 2DIF. Com-pared with the wild type, the acs2-1 mutant had re-duced ACC levels under control conditions, and theselevels were not further affected by 2DIF conditions,

Figure 3. Ethylene restores growth and movement under 2DIF. A, Average amplitudes of days 2 to 6 calculated from theprojected oscillations (in B and C) of leaves treated with mock (white bar), increasing concentrations of ethephon (gray to blackbars) and in Col-0 (light gray bar), etr1-7 (black bar), and etr1-6 etr2-3 ein4-4 (gray bar) developing under 2DIF conditions.Error bars represent SE; n = 8. B, Ethephon dose response of projected oscillations for Col-0 leaves developing on plants sprayedwith increasing ethephon concentrations (0.25–1 mM; gray to black lines) compared with mock (0.0 mM; dashed line) at the start(t = 0 h) of2DIF treatment. Error bars represent SE; n = 8. Gray areas indicate the dark period. C, Projected oscillations of leavesof the constitutive ethylene signaling mutant etr1-7 (black line) compared with the Col-0 wild type (dashed line) under 2DIFconditions. Error bars represent SE; n = 8. Gray areas indicate the dark period. D, Phase shifts of leaf movements between controland 2DIF conditions for leaves treated with mock or 1 mM ethephon and for Col-0, etr1-7, and etr1-6 etr2-3 ein4-4. Significantphase shifts (P , 0.05) are indicated with arrows, and nonsignificant shifts are indicated with bars. Each arrow depicts thedirection and strength of the shift in phase: the start of the arrow indicates average phase during days 2 to 6 under controlconditions, and the arrowhead indicates the average phase for 2DIF (days 2–6). Error bars represent SE; n = 8.


Bours et al.




Figure 4. Ethylene biosynthesis in the petiole is reduced under 2DIF conditions. A, Diurnal ethylene emissions of Col-0 rosetteplants growing under control (solid line, black triangles) or 2DIF (dashed line, white circles) conditions. Error bars represent SE;n $ 10. Gray areas indicate the dark period. B, Typical ACS2 promoter activity pattern obtained using histochemical GUS





indicating that 2DIF is not affecting other ACS activi-ties in the petiole (Fig. 4G). Taken together, we concludethat ACS2 is the main2DIF-responsive ACS gene in thepetiole.

These results suggest that ethylene is able to exertlocal growth effects even though ethylene is expectedto diffuse rapidly throughout the leaf tissue. We testedwhether the site of application of ACC is important forleaf movement responses by placing lanolin pastecontaining ACC (0.5% [w/w]) at either the base of thepetiole or the center of the leaf blade. ACC paste ap-plied to the petiole restored both the amplitude andthe phase of leaf movement under 2DIF, whereasACC applied to the blade only minimally affected leafmovement patterns in these conditions (Fig. 4, H and I,respectively; Supplemental Fig. S4, F and G). Com-bined, these results indicate that local, tissue-specificACS2-dependent ACC production in the petiole con-tributes to leaf elongation and regulates both the am-plitude and phase of leaf movement in response todiurnal light and temperature cycles.

2DIF Reduces the Ethylene Sensitivity of Arabidopsis

Besides differences in ethylene production, sensitivityto the hormone may play a role in controlling leafmovement under 2DIF. Therefore, we assayed the ef-fect of 2DIF on ethylene sensitivity using the classicalseedling triple-response assay (Guzmán and Ecker, 1990).Seedlings were germinated for a full 2 d (48 h) in eithercontrol or 2DIF conditions and then kept for 5 d in thedark, while the temperature cycling was maintained. Inthese conditions, seedlings entrained to control lightand temperature cycles before etiolating displayed thetypical dose-dependent triple response to ACC, whichincludes short, thickened hypocotyls (embryonic stems)and apical hook curvature with closed cotyledons (Fig. 5A).In contrast, in seedlings exposed to2DIF conditions, theinhibition of hypocotyl elongation in response to ACCwas absent up to a reasonably high concentration of

ACC (1 mM), and only at the highest tested ACC con-centrations was a strong inhibition of hypocotyl elon-gation induced. As a negative control, we used theein2-1 ethylene-insensitive mutant, which did not re-spond to any of the treatments (Fig. 5B). Combined,our results suggest that limiting ACC levels in thepetiole, possibly enhanced by reduced ethylene sensi-tivity, results in reduced leaf movement and elonga-tion under 2DIF.

PHYB Modulates Ethylene Levels and Signaling toControl Rhythmic Diurnal Leaf Growth

It has been reported that the Arabidopsis photorecep-tor PHYB mutant phyB1 is insensitive to growth inhibi-tion under 2DIF; however, the underlying mechanismfor this insensitivity remained unclear (Thingnaes et al.,2008). In agreement with the work of Thingnaes et al.(2008), elongation of phyB9 mutant leaves was not re-duced by our 2DIF treatment (Supplemental Fig. S5A).Given that ethylene becomes limiting for leaf movementunder 2DIF (Fig. 3), we investigated a possible link be-tween PHYB and ethylene. First, the leaf movementpatterns in the phyB9 null mutant (Reed et al., 1993)under control and 2DIF conditions were comparedwith the leaf movement of wild-type plants (Fig. 6,A and B). This revealed that the leaf movement am-plitude of phyB9 was strongly increased comparedwith the wild type in both conditions (Fig. 6C). Undercontrol conditions, the phase of phyB9 leaf movementwas advanced compared with the wild type (t ; 13 hand t = 16 h, respectively). In response to 2DIF, thephase of the phyB9 leaf oscillations shifted approxi-mately 9 h (from t ; 13 to t ; 22 h under 2DIF),compared with an approximately 5-h shift in wild-typeplants (Fig. 6D). This early phase of phyB9 undercontrol conditions is similar to that of wild-type plantstreated with ethephon (Fig. 3D).

Given that some features of leaf movement of thephyB9 mutant resemble those of ethylene-treated plants

Figure 4. (Continued.)staining in leaves of equal age developed under control or 2DIF conditions on 4-week-old plants. C, Promoter activity analysisof ACS2 using histochemical GUS staining in leaves developed during 10 d of control or2DIF conditions in 4-week-old plants.D, Average amplitudes of acs2-1 and acs2-2 leaves developing under control (black bars) or 2DIF (gray bars) conditionscompared with the Col-0 wild type, calculated from days 2 to 6 of the projected oscillations (in E). Error bars represent SE; n = 8.E, Projected oscillations of acs2-1 (dark blue line) and acs2-2 (light blue line) leaves compared with the Col-0 wild type (blackdashed line) under control (top) and 2DIF (bottom) conditions. Error bars represent SE; n = 8. Gray areas indicate the darkperiod. F, Phase shifts of leaf movement between control and 2DIF conditions for acs2-1, acs2-2, and Col-0 wild-type leaves.A significant phase shift (P , 0.05) is indicated with the arrow, and nonsignificant shifts are indicated with bars. The arrowdepicts the direction and strength of the shift in phase: the start of the arrow indicates the average phase during days 2 to 6 undercontrol conditions, and the arrowhead indicates the average phase for 2DIF (days 2–6). Error bars represent SE; n = 8. G, ACCcontent of acs2-1 and Col-0 wild-type petioles harvested on the peak of ethylene emission (in A) under 2DIF or controlconditions. Error bars represent SE; n = 8. FW, Fresh weight. H, Average amplitudes of days 2 to 6 calculated from the projectedoscillations (Supplemental Fig. S4, F and G) of leaves treated with 0.5% (w/w) ACC (dark bars) or mock (light bars) lanolin paste(approximately 1 mg) on the petiole (solid bars) or the leaf blade (dashed bars) under 2DIF conditions. Error bars represent SE;n = 8. I, Average phase of days 2 to 6 calculated from the projected oscillations (Supplemental Fig. S4, F and G) of leaves treatedwith 0.5% (w/w) ACC (dark bars) or mock (light bars) lanolin paste (approximately 1 mg) on the petiole (solid bars) or the blade(dashed bars) under 2DIF conditions. Error bars represent SE; n = 8.


Bours et al.







(Supplemental Fig. S5B), ethylene production could beincreased in the phyB9 mutant. Therefore, ethyleneemissions from phyB9 plants under the contrasting day/night temperature difference conditions were measured.Peak ethylene evolution was indeed about 2-fold higherin phyB9 than in wild-type plants under both controland 2DIF conditions (Fig. 6, E and F, respectively). Thephase of ethylene emission in the phyB9 mutant, how-ever, was comparable with the wild type, including thetypical phase shift between control and2DIF conditions(Fig. 6. E and F). Because of the specific role of ACS2 inethylene production for leaf movement (Fig. 4), wenext introduced ACS2::GUS in the phyB9 backgroundby crossing and examined ACS2 promoter activity.ACS2::GUS activity strongly increased in leaves of thephyB9 mutant compared with the wild type undercontrol conditions, while (in contrast to the wild type)2DIF did not reduce the ACS2::GUS activity in thephyB mutant (compare Figs. 4C and 6G). This indicatesthat PHYB suppresses ACS2 expression and thatdiurnal light/temperature control of this repression isan important regulator of leaf movement and growth.Because 2DIF reduced ethylene sensitivity in the wild

type, we next tested the ethylene-sensing capacity inphyB9with a triple response assay. Compared with thewild type, the phyB9 mutant displayed an enhancedtriple response under control conditions. In contrast tothe wild type, under 2DIF the triple response of phyB9was similar to control +DIF (Fig. 5H; SupplementalFig. S5, C and D), indicating that increased ethylenesensitivity is apparent in phyB9 mutants independentof the day/night temperature difference conditions.Combined, these results suggest that the insensitivityof phyB9 to 2DIF relates to both increased ethyleneproduction and enhanced ethylene sensitivity.

To further test the role of ethylene signaling in theresponse of the phyB9 mutant to the day/night tem-perature difference phenotypes, we constructed a phyB9ein2-1 double mutant (Supplemental Fig. S6A), becauseein2-1 confers insensitivity to ethylene (Guzmán andEcker, 1990). In response to 2DIF, the phyB9 ein2-1 dou-ble mutant displayed visually lower leaf angles com-pared with phyB9 (Fig. 7A). The amplitude of the leafmovement of phyB9 ein2-1 under both control and 2DIFconditions was decreased compared with phyB9 but washigher than in the ein2-1 mutant (Fig. 7). This confirmsthat the phyB9 leaf movement phenotype is in partcaused by increased ethylene signaling. The strongerleaf movement of the phyB9 ein2-1 double mutant com-pared with the ein2-1 single mutant, however, suggeststhat additional EIN2 independent mechanisms may alsooperate in the regulation of leaf movement in the phyB9background. The absolute lengths of phyB9 ein2-1 leaveswere strongly reduced in response to 2DIF comparedwith control conditions (Supplemental Fig. S6B). Thisis in contrast to the phyB9 single mutant, in which leaflength is not reduced by2DIF (compare SupplementalFigs. S5C and S6B).

In conclusion, the leaf movement and growth re-sponses to 2DIF directly relate to ACC production inthe petiole, which is mostly mediated by ACS2 activityand depends on ethylene signaling capacity throughEIN2, both of which act downstream of the photore-ceptor PHYB.

DISCUSSION

2DIF Limits Leaf Growth and Movement during the Day

Previous studies have provided a detailed under-standing of how hormone and light signaling path-ways are integrated and synchronized with daylengthto fine-tune plant growth in response to diurnal orseasonal changes (Nozue et al., 2007; Michael et al.,2008; Zhong et al., 2012). However, in the natural en-vironment, light is accompanied by temperature cycles.Here, we performed detailed analysis of leaf growthand movement in response to both natural (+DIF) andartificially imposed (2DIF) light/temperature conditions.Whereas control diurnal conditions result in continuousleaf elongation, accompanied by continuous diurnal leafmovement, the 2DIF condition results in the cessation

Figure 5. 2DIF reduces the ethylene sensitivity of Arabidopsis.A, Responses of 5-d-old etiolated Col-0 seedlings to various concen-trations of ACC. After 48 h of germination under control or 2DIFconditions, seedlings were kept in the dark with temperature cycles for72 h. B, Hypocotyl lengths of 5-d-old Col-0 wild-type (black lines) orein2-1 (red lines) seedlings in response to various concentrations ofACC after 48 h of germination under control or 2DIF conditions.Seedlings were then kept in the dark with temperature cycles for 72 hon various ACC concentrations. Error bars represent SE; n $ 60. [Seeonline article for color version of this figure.]













of leaf movement and reduced elongation during thephotoperiod (Fig. 1; Supplemental Fig. S1). We foundthat the effect of 2DIF on leaf movement can be tracedback to an effect on ethylene levels and ethylene sig-naling, which become limiting under 2DIF, but can be

complemented by the application of ethephon or ACC(Figs. 3A and 4H).

Ethylene levels and signaling affect the phase andamplitude of thermoperiodic leaf movement, and thecharacterizations of phase and amplitude after different

Figure 6. PHYB represses ethylene-dependent leaf movement. A and B,Projected oscillations for phyB9 (solidblue line) and Col-0 (dotted black line)leaves developing under control (A) or2DIF (B) conditions. Error bars repre-sent SE; n = 8. Gray areas indicate thedark period. C, Average amplitudes ofCol-0 and phyB9 leaves developingunder control (black bars) or 2DIF(gray bars) conditions, calculated fromdays 2 to 6 of the projected oscillations(in A and B). Error bars represent SE;n = 8. D, Phase shifts of leaf movementbetween control and 2DIF conditionsfor phyB9 and Col-0 wild-type leaves.Significant phase shifts (P , 0.05) areindicated with arrows. Each arrow de-picts the direction and strength of theshift in phase: the start of the arrowindicates average phase during days2 to 6 under control conditions, and thearrowhead indicates the average phasefor 2DIF (days 2–6). Error bars repre-sent SE; n = 8. E and F, Ethylene emis-sions of phyB9mutants (blue diamonds,solid lines) compared with Col-0 wild-type emissions (white circles, dashedlines) under control (E) and 2DIF (F)conditions. Error bars represent SE; n $

10. Gray areas indicate the dark period.FW, Fresh weight. G, Promoter activityanalysis of ACS2 using histochemicalGUS staining in phyB9 leaves developedduring 10 d of 2DIF (top) or control(bottom) conditions on 4-week-oldplants. H, Relative hypocotyl lengths of5-d-old phyB9 (solid blue line) or Col-0wild-type (dotted black line) seedlingsin response to various concentrations ofACC. After 48 h of germination undercontrol (triangles) or 2DIF (circles)conditions, seedlings were kept in thedark with temperature cycles for 72 hon various ACC concentrations. Absolutelengths (as depicted in SupplementalFig. S5D) were normalized for thelength at 0 mM ACC (100%). Error barsrepresent SE; n = 60.


Bours et al.






chemical treatments and in various mutants are sum-marized in Figure 8. In general, high ethylene levels orincreased signaling correlate with an early phase andhigh amplitude of leaf movement, while low ethylenelevels or reduced signaling correlate with a late phaseand low amplitude of leaf movement (Fig. 8, A and B).Combined, our results demonstrate that leaf movementamplitude and leaf movement phase do not correlate tototal ethylene emissions; rather, they are a function oflocal ACC production by ACS2 (Fig. 4). In contrast tothe lower activity of ACS2:GUS under 2DIF, ACS8:GUS activity was increased under 2DIF conditions,and GUS presence increased mostly in the transverseedges of the blade (Supplemental Fig. S4A). Previously,rhythmic expression of ACS8 was reported to corre-spond with rhythmic ethylene emissions, which arecontrolled by the circadian clock (Thain et al., 2004).Interestingly, ACS8 transcription is negatively regulatedby ethylene signaling (Thain et al., 2004). In 2DIF, notonly local ethylene production decreased (Fig. 4G) butalso ethylene sensitivity was reduced (Fig. 5). A re-duced negative feedback on ACS8 expression is inagreement with the increased ACS8 activity under2DIFand could explain why overall ethylene emissions arenot reduced in 2DIF.

PHYB Control of Ethylene Production and Sensitivity

Our results confirm previous observations that PHYBaffects ethylene production (Vandenbussche et al., 2003;

Pierik et al., 2004). In our conditions, the phyB9 mutantshowed increased ethylene production compared withthe wild type (Fig. 6, E and F). PHYB also repressedethylene sensitivity, as the phyB9 mutant showed in-creased sensitivity to ethylene in the triple response(Fig. 6H). The increased ethylene emission and sensi-tivity of the phyB9 mutant tentatively result in en-dogenous ethylene signaling levels that are no longerlimiting during 2DIF. The ethylene signaling towardthe leaf movement response is likely mediated byEIN2, as mutations in this gene strongly reduce the leafmovement amplitude of the phyB9 mutant (Fig. 7B). Ithas been reported that the role of ethylene in growth isdependent on light (for review, see Pierik et al., 2006).For example, in dark-grown seedlings, ethylene in-hibits hypocotyl elongation as part of the triple re-sponse (Guzmán and Ecker, 1990), but in continuouslight, ethylene stimulates hypocotyl elongation (Smalleet al., 1997). This phenomenon was recently explainedby the activation of two contrasting pathways by theethylene signaling transcription factor EIN3. In thelight, EIN3 activates the growth-stimulating transcriptionfactor PHYTOCHROME-INTERACTING FACTOR3. Incontrast, in the dark, EIN3 activates the ETHYLENERESPONSE FACTOR1-mediated growth-inhibiting path-way (Zhong et al., 2012). The role of ethylene in thecontrol of elongation is thus condition dependent,which could explain why, in a previous study, per-formed under continuous light and temperature, eth-ylene appeared to have no role in controlling the phase

Figure 7. Dissection of leaf movementin the phyB9 ein2-1 double mutant.A, Representative 36-d-old phyB9,ein2-1, and phyB9 ein2-1 rosette plantsdeveloping under 2DIF conditions atthe end of the fourth photoperiod.B, Average amplitudes of phyB9, ein2-1,and phyB9 ein2-1 leaves developingunder control (black bars) or 2DIF(gray bars) conditions, calculated fromdays 2 to 6 of the projected oscillations(in C and D). Error bars represent SE;n = 8. C and D, Projected leaf oscilla-tions of phyB9, ein2-1, and phyB9ein2-1 double mutant developing un-der control (C) or 2DIF (D) conditions.Error bars represent SE; n = 8. Grayareas indicate the dark period. [Seeonline article for color version of thisfigure.]






of circadian leaf movement (Thain et al., 2004). Thisillustrates that the characterization of circadian pro-cesses under constant conditions is not always easilytranslated to natural (or economically relevant) diurnallight and temperature conditions.

PHYB Constrains the Phase Shifts of Clock-RegulatedProcesses in Response to 2DIF

Previously, it was shown that multiple genes in-volved in cell elongation are regulated by the circadianclock and that the sequential transcription of key stepsin cell elongation remained orchestrated under constantconditions (Harmer et al., 2000). Our results show thatfor two principally circadian clock-regulated processes,leaf movement (McClung, 2006) and ethylene produc-tion (Thain et al., 2004), the phase of activity is shiftedunder the two diurnal conditions, 2DIF and +DIF, butthe phase shift for each of these processes is different

(e.g. 5 h forward for leaf movement, 4 h backward forethylene production; Fig. 8, B and D). Rhythmic eth-ylene evolution is regulated by the circadian clock,resulting in a specific phase of emission under diurnalentrainment by both light and temperature cycles(Finlayson et al., 1998; Thain et al., 2004). In sorghum(Sorghum bicolor), temperature was the dominant en-trainment signal, setting the phase of ethylene pro-duction independently of light cycles (Finlayson et al.,1998). In our experiments with Arabidopsis, the phaseof ethylene emission remained within the photoperiodin response to2DIF (Fig. 4A). Under 2DIF conditions,the phase of leaf movement of the wild type is shifted5 h forward compared with control conditions, but thephase remained within the dark (warm) period. Incontrast to the wild type, the phase of leaf movementof the phyB9 null mutant largely adapted to the temper-ature cycles: phyB9 leaves reached the highest position atthe start of the cold period independently of the lightcycles (Fig. 5D). This indicates that the 2DIF-induced

Figure 8. Amplitudes and phase shifts of leaf movement and ethylene emission between control and 2DIF conditions inmutants and treatments used in this study. A, Averaged amplitudes (measured during days 2–6) of leaf movements. Solid barsmark the wild-type response under control conditions, and dashed lines mark the2DIF wild-type response. Error bars representSE; n = 8. B, Phase shifts of leaf movements between control and 2DIF conditions compared with the wild-type (WT) response.Significant phase shifts (P , 0.05) are indicated with arrows, and nonsignificant shifts are indicated with bars. Each arrowdepicts the direction and strength of the shift in phase: the start of the arrow indicates average phase during days 2 to 6 undercontrol conditions, and the arrowhead indicates the average phase for 2DIF (days 2–6). Error bars represent SE; n = 8. C and D,Average peak amplitudes of ethylene emission (C) and phase shifts of ethylene emission (D) between control and 2DIF con-ditions for phyB9 mutants and the Col-0 wild-type. Error bars represent SE; n $ 10.


Bours et al.



phase shift of leaf movement in the wild type is partlycompensated by PHYB signaling. However, the effectof PHYB on the phase is not the same for all clock-controlled processes: loss of PHYB function had no ef-fect on the phase shift of ethylene emissions (Fig. 8D). Ithas been reported that 2DIF differentially entrains thephase of expression (measured under constant condi-tions) of two other clock-controlled genes, CHLORO-PHYL A/B BINDING2 and CATALASE3 (Michael et al.,2003). A differential effect of 2DIF on the rhythmic ex-pression (phase/amplitude) of all clock-regulated geneswould reduce the coordination of growth- and elongation-related processes and thus could explain the compactnessof plants exposed to 2DIF.

Factors Downstream of Ethylene That Limit Growthduring the Cold Photoperiod

Previously, the average expansion rate of rosette leavesunder diurnal conditions was dissected by Pantin et al.(2011). It was shown that the expansion rate during thephotoperiod is approximately 50% of that during thedark period. If 2DIF would completely abolish the ex-pansion rate during the day, those results would predictan approximately 33% reduction in leaf size under2DIF.However, 2DIF results in an approximately 40% re-duction in leaf size (Supplemental Fig. S1B), suggest-ing that under2DIF also the expansion rate during thenight is slightly reduced compared with control con-ditions. Two factors were identified to be of impor-tance for leaf expansion rate: for emerging leaves, theexpansion rate of rosette leaves may be limited bystarch availability in the early night, while for biggerleaves, the expansion during the day was shown to belimited by reduced hydraulic pressure (Pantin et al.,2011). The leaf movement analysis of our OSCILLATOR-dependent experiments started with leaves 2 d afteremergence, in which the expansion rate is mainly lim-ited by hydraulic pressure (Pantin et al., 2011). How-ever, it is possible that the 2DIF treatment also altersclock-regulated carbohydrate metabolism such that, forthe later stages of leaf development, the required car-bohydrates become limiting for cell elongation and leafmovement. It would be interesting to assess the effectsof 2DIF on the phase and amplitude of expression ofcore clock components and examine whether carbohy-drate availability relates to the altered leaf movementpatterns.

MATERIALS AND METHODS

Plant Material and Growth Conditions

Arabidopsis (Arabidopsis thaliana) mutants were either from the Notting-ham Arabidopsis Stock Centre (www.arabidopsis.info), with accession num-bers in parentheses, or from the authors who described the mutant. Alltransgenic lines and mutants are in the Col-0 (N1092) background: ACS2::GUS/GFP (N31380), ACS4::GUS (N31381), ACS5::GUS (N31382), ACS6::GUS/GFP(N31383), ACS8::GUS/GFP (N31385) ACS9::GUS/GFP (N31386), ACS11::GUS/GFP (N31387), acs2-1 (N16564), acs2-2 (N16565), ein2-1 (N3071; Guzmán and

Ecker, 1990), etr1-1 (N237; Bleecker et al., 1988), etr1-7, etr1-6, ein4-4, etr2-3 (Huaand Meyerowitz, 1998), and phyB9 (Reed et al., 1993). The phyB9 ein2-1 mutantand the phybB9 ACS2::GUS/GFP and phybB9 ACS8::GUS/GFP lines were gen-erated by crossing the phyB9 mutant with ein2-1 or ACS2::GUS and ACS8::GUS(Col-0), respectively. Plants with the phyB9 phenotype were selected in the F2and selfed (F3). The phyB9 ein2-1 double mutant was detected in the F3 byselecting for ethylene insensitivity (lack of the triple response phenotype in thepresence of 10 mM ACC). The phyB9 ACS::GUS/GFP line was selected basedon GUS presence. The presence of the phyB allele was confirmed by testinghypocotyl elongation under continuous red (630 nm) light-emitting diodelight. The selected lines were selfed, and the F4 progeny displayed no phe-notypic segregation and were used in this study.

Plant growth under control (+DIF) conditions was performed as describedpreviously (Bours et al., 2012). All experiments were performed in automatedclimate-controlled Weiss (http://www.wkt.com) cabinets (12/12-h light/darkcycle). Relative humidity was kept constant at 60% (v/v), and photosyntheticactive radiation was 150 mmol m22 s21 from white fluorescents tubes (Philips;type T5, color code 840). Ambient temperature cycles for growth under controlconditions were 22°C (photoperiod) and 12°C (dark period), with a temper-ature ramp of 0.33°C min21. Measurements showed that soil temperaturelagged approximately 20 min behind ambient air temperature. Plants weregrown in fertilized peat/perlite-based soil in square (53 53 5 cm) plastic potswith different genotypes placed at random positions in the growth cabinet onan irrigation mat that was watered automatically to saturation through poroustubing from a basin containing tap water every 3 d at the start of the photo-period. After 20 d, plants were irrigated once with one-half-strength Hoaglandnutrient solution (Hoagland and Arnon, 1950) instead of water to provideextra nutrients. Five days later, plants were transferred to a second climatecabinet for imaging with similar conditions and an infrared camera system(OSCILATOR; see below) with infrared lights and allowed to acclimate for7 d before the onset of imaging. For2DIF treatment, plants were grown undercontrol (+DIF) conditions, and at the start of the photoperiod, the temperaturecycles were reversed to 2DIF 12°C (photoperiod) and 22°C (dark period) witha temperature ramp of 0.33°C min21 in the same growth cabinet. All othergrowth parameters were kept equal to control conditions.

Plant Growth Analysis Based on the OSCILLATORGrowth Monitoring System

Plant growth imaging, image data analysis, and extraction of parameters ofleaf growth oscillations using the OSCILLATOR setup were performed asdescribed previously (Bours et al., 2012). OSCILLATOR uses a continuous top-down imaging of mature Arabidopsis rosette plants with an infrared camerasystem and enables extraction and wavelet analysis of oscillating leaf move-ments. The distance between the leaf tip and the plant apex for each image iscalculated and plotted as projected length. Subsequently, a best-fit second-degree polynomial trend line is automatically calculated for each individualprojected leaf length curve, and the residual values are subtracted from thisline. The resulting residual is inverted to allow maximum upright leaf positionto correspond to maximum peak height. These raw projected oscillations arethen smoothed using Waveclock script (Price et al., 2008), embedded in theOSCILLATOR script providing the (smoothed) projected oscillations fromwhich the amplitude and phase for each day for single leaves are automati-cally calculated. Calculation of phase and amplitude is based on the selectionof the data point with the highest value (peak), which has to be flanked by twolower value data points for each individual day (24 h). For this “peak,” thex coordinate represents the phase and the y coordinate represents the ampli-tude. Averages of phase and amplitude were calculated from the values forindividual leaves for independent days. All plants analyzed were in the samedevelopmental stage, and for each plant, rosette leaves 12 and 13 (in order ofemergence) were used for leaf movement analysis. Two leaves per plant wereanalyzed, with a minimum of four plants (total n = 8–10, referring to thereplicate number of leaves included in the analysis). Statistical comparisonsbetween treatments were performed by two-tailed Student’s t test (Excel). Allexperiments were conducted at least twice using independent trials withsimilar setups and similar outcomes.

Analysis of Leaf Morphology

All leaves present at t = 0 h (less than 7 mm) were minimally marked withred paint (Vingerverf Creal). Unmarked leaves that developed during thetreatment were harvested and subsequently scanned using a Canon flatbed





http://www.arabidopsis.info

http://www.wkt.com


scanner (CanoScan 5600F). Morphological parameters were calculated usingImageJ freeware (rsb.info.nih.gov/ij/).

Histochemical GUS Staining

Histochemical GUS staining of ACS::GUS leaves was performed as de-scribed previously (Kim et al., 2006). All leaves present at t = 0 h (more than7 mm) were minimally marked with red paint (Vingerverf Creal), and plantswere kept under control or 2DIF conditions. After 10 d, whole rosettes wereharvested, and marked leaves that developed before the treatment were re-moved. After staining, fixation, and ethanol clearing of the intact rosetteplants, the individual leaves were separated and scanned with a Canon flat-bed scanner (CanoScan 5600F). Backgrounds of the scanned images were re-moved using Adobe Photoshop 3.1.

Pharmacological Experiments

All plant growth regulators used were dissolved in deionized water with0.005% (v/v) surfactant Tween 20 (Sigma-Aldrich) and applied to aerial parts ofrosette plants by spraying (airbrush) at the start of the experiment (t = 0 h).Mock plants were treated with identical solutions that lacked the activecomponents. For ethephon treatments of rosette plants, Ethrel-A (Luxan),which contains 2-chloroethylphosphonic acid (480 g L21) stabilized in apotassium phosphate buffer, was diluted to a desired concentration (0, 0.25,0.5, or 1 mM) before application. STS was prepared by mixing 80% (v/v) 0.1 M

STS (Sigma-Aldrich) with 20% (v/v) 0.1 M silver nitrate (Sigma-Aldrich) asdescribed in Sigma protocols and subsequently diluted to 50 mM STS.

Hypocotyl Elongation Response to ACC

Seeds were sterilized in 2.5% (v/v) sodium hypochlorite solution for 5 min,washed three times with water, sown on petri dishes containingMurashige andSkoog enriched plant agar (8 g L21 plant agar [Duchefa] and 2.2 g L21 Murashigeand Skoog agar [Duchefa]) and different concentrations (0, 0.1, 1, 10, and100 mM) of ACC (Sigma-Aldrich), and stratified in the dark for 4 d (4°C) tosynchronize germination. Seeds were then exposed to either control or 2DIFconditions for 48 h (two photoperiods) until radicle protrusion. The timing ofradical protrusion was equal between control and 2DIF conditions. Theseedling plates were then transferred to darkness (covered with aluminumfoil) at the start of the third photoperiod or left uncovered under the diurnalexperimental conditions (+DIF/2DIF) for two cycles (72 h). Seedlings wereplaced in a horizontal position, photographed, and hypocotyl lengths weremeasured using ImageJ software.

Ethylene Emission Analysis

Ethylene releasewasmeasured in real time as described previously (Millenaaret al., 2009) using a laser-driven photoacoustic ethylene detector (Sensor Sense)for 4- to 5-week-old Col-0 and phyB9 plants growing under experimental con-ditions with the following modifications. Single plants including the pot wereplaced in 350-mL gas-tight translucent tissue culture pots equipped with tubes toaccommodate inflow and outflow under control or 2DIF conditions 1 d beforethe start of the measurement. Pots were flushed continuously with ethylene-freeair (1 L h21

flow rate). From day 2 onward, emissions were measured with 2-hintervals for a period of at least 48 h. Pots containing only soil without plantswere included in the cultivation procedure and included in the analysis asbackground emissions.

ACC Extraction, Detection, and Quantification by LiquidChromatography-Tandem Mass Spectrometry

ACC extraction was performed as described previously (Voesenek et al.,2003). Frozen leaf material (about 200 mg fresh weight) was ground in liquidnitrogen with a ball mill (MM400; Retsch) in a 2-mL Eppendorf tube and thenextracted with 1 mL of extraction solvent (water:ethanol, 20:80 [v/v]) andsonicated for 5 min at 22°C in a Branson 3510 ultrasonic bath (BransonUltrasonics). After centrifugation (10,000 rpm for 15 min at 4°C), the supernatantwas collected and the pellet was reextracted with 1 mL of extraction solvent.[2H2]ACC (OlChemim) was added to the second extraction solvent as an

internal standard. After centrifugation (10,000 rpm for 15 min at 4°C), thesupernatant was collected and evaporated to dryness in vacuo at 55°C(SpeedVacuum Savant SPD121P; Thermo Scientific). The residue was dissolved in2 mL of distilled water, 2 mL of dichloromethane was added, and after vortexing(10 s), the mixture was centrifuged at 2500 rpm for 5 min. The upper phase wascollected and evaporated to dryness in vacuo at 55°C. The residue was suspendedin 50 mL of methanol. After centrifugation (10,000 rpm for 15 min at 4°C), 10 mL ofthe supernatant was used for derivatization of the ACC for liquid chromatography-tandem mass spectrometry analysis as described previously (Hall et al., 1989), withthe modification that the Waters AccQ$Fluor Reagent Kit (for amino acidderivatization) was used (Waters). For ACC derivatization, 10 mL of ACCextract supernatant were mixed with 20 mL of reconstituted AccQ$Fluor rea-gent in 70 mL of borate buffer and heated at 55°C for 10 min.

Analysis of ACC in Arabidopsis leaf extracts was performed by comparingretention times and mass transitions with those of standard ACC (Sigma) as de-scribed previously (Kohlen et al., 2011) with the following modifications. AWatersXevo tandem mass spectrometer equipped with an electrospray ionization sourceand coupled to an Acquity ultra-performance liquid chromatography system(Waters) was used. Chromatographic separation was achieved on an AcquityUPLC HSS T3 column (100 3 2.1 mm, 1.8 mm; Waters) by applying a methanol-water gradient to the column, starting from 0.1% (v/v) methanol for 0.1 min andrising to 9.1% (v/v) methanol at 5.74 min, followed by a 2.0-min gradient to21.2% (v/v) methanol, followed by a 0.3-min gradient to 25% (v/v) methanol,which was maintained for 0.56 min, before going back to 0.1% (v/v) methanolusing a 0.15-min gradient. The column was equilibrated for 2.25 min, using thissolvent composition prior to the next run. The run time was 11 min. The columnwas operated at 55°C with a flow rate of 0.7 mL min21. Sample injection volumewas 5 mL. The mass spectrometer was operated in positive electrospray ionizationmode. Cone and desolvation gas flows were set to 50 and 1,000 L h21, respectively.The capillary voltage was set at 3.0 kV, the source temperature at 150°C, and thedesolvation temperature at 650°C. The cone voltage was optimized for derivatizedACC and [2H2]ACC standard using the IntelliStart MS Console (Waters). Argonwas used for fragmentation by collision-induced dissociation. Multiple reactionmonitoring (MRM) was used for ACC identification and quantification. Parent-daughter transitions were optimized for derivatized ACC and [2H2]ACC standardsusing the IntelliStart MS Console. For identification, the followingMRM transitionswere selected: mass-to-charge ratio (m/z) 272.20 . 170.98 at a collision energy of24 eV andm/z 272.20. 115.97 at 46 eV; and for [2H2]ACC,m/z 276.20. 170.96 at acollision energy of 22 eV andm/z 276.20. 115.97 at 46 eV. Cone voltage was set to28 eV. ACC was quantified using a calibration curve with known amounts ofstandards and based on the ratio of the area of the MRM transition m/z 272.20 .170.96 for ACC to the MRM transition m/z 276.20 . 170.96 for [2H2]ACC. Dataacquisition and analysis were performed using MassLynx 4.1 software (Waters).The summed area of all the correspondingMRM transitions was used for statisticalanalysis.

Supplemental Data

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

Supplemental Figure S1. Projected and absolute leaf lengths decrease un-der 2DIF conditions.

Supplemental Figure S2. Ethylene signaling controls the amplitude andphase of diurnal leaf movement.

Supplemental Figure S3. Effect of ethylene on leaf growth and movement.

Supplemental Figure S4. ACS::GUS activities in Arabidopsis leaves, andleaf lengths and oscillations of acs2-1 and acs2-2 mutants.

Supplemental Figure S5. PHYB affects ethylene sensitivity and leaf growthand movement under diurnal light and temperature cycles.

Supplemental Figure S6. Hypocotyl and leaf elongation of the phyB9ein2-1 double mutant.

Supplemental Movie File S1. Infrared time-lapse recording of Col-0 rosetteplants developing under control or 2DIF conditions.

ACKNOWLEDGMENTS

We are grateful to T. Charnikova for setting up and supervising the ACCanalysis. We thank B. Binder for his advice and for providing seeds of the


Bours et al.










ethylene mutants and R. Welschen, F. Verstappen, M. Schreuder, C. MengFoong,and W. Kohlen for advice and technical assistance.

Received May 20, 2013; accepted August 22, 2013; published August 26, 2013.

LITERATURE CITED

Bleecker AB, Estelle MA, Somerville C, Kende H (1988) Insensitivity toethylene conferred by a dominant mutation in Arabidopsis thaliana.Science 241: 1086–1089

Bours R, Muthuraman M, Bouwmeester H, van der Krol A (2012) OSCILLA-TOR: a system for analysis of diurnal leaf growth using infrared photographycombined with wavelet transformation. Plant Methods 8: 29

Cancel JD, Larsen PB (2002) Loss-of-function mutations in the ethylenereceptor ETR1 cause enhanced sensitivity and exaggerated response toethylene in Arabidopsis. Plant Physiol 129: 1557–1567

Carvalho SMP, Heuvelink E, Cascais R, van Kooten O (2002) Effect of dayand night temperature on internode and stem length in chrysanthemum:is everything explained by DIF? Ann Bot (Lond) 90: 111–118

Finlayson SA, Lee IJ, Morgan PW (1998) Phytochrome B and the regulation ofcircadian ethylene production in sorghum. Plant Physiol 116: 17–25

Guzmán P, Ecker JR (1990) Exploiting the triple response of Arabidopsis toidentify ethylene-related mutants. Plant Cell 2: 513–523

Hall KC, Pearce DME, Jackson MB (1989) A simplified method for de-termining 1-aminocyclopropane-1-carboxylic acid (ACC) in plant-tissues using a mass selective detector. Plant Growth Regul 8: 297–307

Harmer SL, Hogenesch JB, Straume M, Chang HS, Han B, Zhu T, Wang X,Kreps JA, Kay SA (2000) Orchestrated transcription of key pathways inArabidopsis by the circadian clock. Science 290: 2110–2113

Hoagland DR, Arnon DI (1950) The water-culture method for growing plantswithout soil. Circular California Agricultural Experiment Station 347: 1–32

Hua J, Meyerowitz EM (1998) Ethylene responses are negatively regulatedby a receptor gene family in Arabidopsis thaliana. Cell 94: 261–271

Kende H (1993) Ethylene biosynthesis. Annu Rev Plant Physiol Plant MolBiol 44: 283–307

Kim K-W, Franceschi VR, Davin LB, Lewis NG, Salinas J, Sanchez-Serrano JJ (2006) Beta-glucuronidase as reporter gene: advantages andlimitations. Methods Mol Biol 323: 263–273

Kohlen W, Charnikhova T, Liu Q, Bours R, Domagalska MA, Beguerie S,Verstappen F, Leyser O, Bouwmeester H, Ruyter-Spira C (2011) Stri-golactones are transported through the xylem and play a key role inshoot architectural response to phosphate deficiency in nonarbuscularmycorrhizal host Arabidopsis. Plant Physiol 155: 974–987

Maas FM, van Hattum J (1998) Thermomorphogenic and photomorpho-genic control of stem elongation in fuchsia is not mediated by changes inresponsiveness to gibberellins. J Plant Growth Regul 17: 39–45

McClung CR (2006) Plant circadian rhythms. Plant Cell 18: 792–803Michael TP, Breton G, Hazen SP, Priest H, Mockler TC, Kay SA, Chory J

(2008) A morning-specific phytohormone gene expression programunderlying rhythmic plant growth. PLoS Biol 6: e225

Michael TP, Salome PA, McClung CR (2003) Two Arabidopsis circadianoscillators can be distinguished by differential temperature sensitivity.Proc Natl Acad Sci USA 100: 6878–6883

Millenaar FF, Cox MCH, van Berkel YEMD, Welschen RAM, Pierik R,Voesenek LAJC, Peeters AJM (2005) Ethylene-induced differentialgrowth of petioles in Arabidopsis: analyzing natural variation, responsekinetics, and regulation. Plant Physiol 137: 998–1008

Millenaar FF, van Zanten M, Cox MC, Pierik R, Voesenek LA, Peeters AJ(2009) Differential petiole growth in Arabidopsis thaliana: photocontroland hormonal regulation. New Phytol 184: 141–152

Myster J, Moe R (1995) Effect of diurnal temperature alternations on plantmorphology in some greenhouse crops: a mini review. Sci Hortic (Am-sterdam) 62: 205–215

Nozue K, Covington MF, Duek PD, Lorrain S, Fankhauser C, Harmer SL,Maloof JN (2007) Rhythmic growth explained by coincidence betweeninternal and external cues. Nature 448: 358–361

Pantin F, Simonneau T, Rolland G, Dauzat M, Muller B (2011) Control ofleaf expansion: a developmental switch from metabolics to hydraulics.Plant Physiol 156: 803–815

Patil GG, Alm V, Moe R, Junttila O (2003) Interaction between phyto-chrome B and gibberellins in thermoperiodic responses of cucumber. JAm Soc Hortic Sci 128: 642–647

Pierik R, Tholen D, Poorter H, Visser EJW, Voesenek LACJ (2006) TheJanus face of ethylene: growth inhibition and stimulation. Trends PlantSci 11: 176–183

Pierik R, Whitelam GC, Voesenek LACJ, de Kroon H, Visser EJW (2004)Canopy studies on ethylene-insensitive tobacco identify ethylene as a novelelement in blue light and plant-plant signalling. Plant J 38: 310–319

Price TS, Baggs JE, Curtis AM, Fitzgerald GA, Hogenesch JB (2008)WAVECLOCK: wavelet analysis of circadian oscillation. Bioinformatics24: 2794–2795

Reed JW, Nagpal P, Poole DS, FuruyaM, Chory J (1993) Mutations in the gene forthe red/far-red light receptor phytochrome B alter cell elongation and physio-logical responses throughout Arabidopsis development. Plant Cell 5: 147–157

Smalle J, Haegman M, Kurepa J, Straeten DV Van Montagu M (1997)Ethylene can stimulate Arabidopsis hypocotyl elongation in the light.Proc Natl Acad Sci USA 94: 2756–2761

Stavang JA, Lindgård B, Erntsen A, Lid SE, Moe R, Olsen JE (2005)Thermoperiodic stem elongation involves transcriptional regulation ofgibberellin deactivation in pea. Plant Physiol 138: 2344–2353

Strader LC, Beisner ER, Bartel B (2009) Silver ions increase auxin effluxindependently of effects on ethylene response. Plant Cell 21: 3585–3590

Thain SC, Vandenbussche F, Laarhoven LJJ, Dowson-Day MJ, Wang ZY,Tobin EM, Harren FJM, Millar AJ, Van Der Straeten D (2004) Circadianrhythms of ethylene emission in Arabidopsis. Plant Physiol 136: 3751–3761

Thingnaes E, Torre S, Ernstsen A, Moe R (2003) Day and night tempera-ture responses in Arabidopsis: effects on gibberellin and auxin content,cell size, morphology and flowering time. Ann Bot (Lond) 92: 601–612

Thingnaes E, Torre S, Moe R (2008) The role of phytochrome B, D and E inthermoperiodic responses of Arabidopsis thaliana. Plant Growth Regul56: 53–59

Tsuchisaka A, Theologis A (2004) Unique and overlapping expressionpatterns among the Arabidopsis 1-amino-cyclopropane-1-carboxylatesynthase gene family members. Plant Physiol 136: 2982–3000

Tsuchisaka A, Yu GX, Jin HL, Alonso JM, Ecker JR, Zhang XM, Gao S,Theologis A (2009) A combinatorial interplay among the 1-aminocyclopropane-1-carboxylate isoforms regulates ethylene biosynthesis in Arabidopsis thaliana.Genetics 183: 979–1003

Vandenbussche F, Vriezen WH, Smalle J, Laarhoven LJJ, Harren FJM,Van Der Straeten D (2003) Ethylene and auxin control the Arabidopsisresponse to decreased light intensity. Plant Physiol 133: 517–527

van Zanten M, Basten Snoek L, van Eck-Stouten E, Proveniers MC, ToriiKU, Voesenek LA, Peeters AJ, Millenaar FF (2010) Ethylene-inducedhyponastic growth in Arabidopsis thaliana is controlled by ERECTA.Plant J 61: 83–95

Voesenek LACJ, JacksonMB, Toebes AHW, Huibers W, VriezenWH, ColmerTD (2003) De-submergence-induced ethylene production in Rumex palustris:regulation and ecophysiological significance. Plant J 33: 341–352

Went FW (1944) Plant growth under controlled conditions. II. Thermope-riodicity in growth and fruiting of the tomato. Am J Bot 31: 135–150

Yamagami T, Tsuchisaka A, Yamada K, Haddon WF, Harden LA, Theologis A(2003) Biochemical diversity among the 1-amino-cyclopropane-1-carboxylatesynthase isozymes encoded by the Arabidopsis gene family. J Biol Chem 278:49102–49112

Zhong S, Shi H, Xue C, Wang L, Xi Y, Li J, Quail P H, Deng X W, Guo H(2012) A molecular framework of light-controlled phytohormone actionin Arabidopsis. Curr Biol 22: 1530–1535





Documents

Antiphase Light and Temperature Cycles Affect … · Ethylene application restored leaf growth in 2DIF conditions, and constitutive ethylene signaling mutants maintain robust leaf