Reactive Oxygen Species Tune Root Tropic Responses1[OPEN]...these tropic responses. We note that strong DHR ﬂuo-rescence is detected in the root vasculature above the CEZ at all

Reactive Oxygen Species Tune Root TropicResponses1[OPEN]

Gat Krieger2, Doron Shkolnik2, Gad Miller, and Hillel Fromm*

Department of Molecular Biology and Ecology of Plants, Faculty of Life Sciences, Tel Aviv University, TelAviv 69978, Israel (G.K., D.S., H.F.); and Mina and Everard Goodman Faculty of Life Sciences, Bar-IlanUniversity, Ramat-Gan 5290002, Israel (G.M.)

ORCID IDs: 0000-0003-3498-4362 (D.S.); 0000-0002-7196-6529 (G.M.); 0000-0003-1977-1211 (H.F.).

The default growth pattern of primary roots of land plants is directed by gravity. However, roots possess the ability to sense andrespond directionally to other chemical and physical stimuli, separately and in combination. Therefore, these root tropicresponses must be antagonistic to gravitropism. The role of reactive oxygen species (ROS) in gravitropism of maize andArabidopsis (Arabidopsis thaliana) roots has been previously described. However, which cellular signals underlie theintegration of the different environmental stimuli, which lead to an appropriate root tropic response, is currently unknown.In gravity-responding roots, we observed, by applying the ROS-sensitive fluorescent dye dihydrorhodamine-123 and confocalmicroscopy, a transient asymmetric ROS distribution, higher at the concave side of the root. The asymmetry, detected at thedistal elongation zone, was built in the first 2 h of the gravitropic response and dissipated after another 2 h. In contrast,hydrotropically responding roots show no transient asymmetric distribution of ROS. Decreasing ROS levels by applying theantioxidant ascorbate, or the ROS-generation inhibitor diphenylene iodonium attenuated gravitropism while enhancinghydrotropism. Arabidopsis mutants deficient in Ascorbate Peroxidase 1 showed attenuated hydrotropic root bending.Mutants of the root-expressed NADPH oxidase RBOH C, but not rbohD, showed enhanced hydrotropism and less ROS intheir roots apices (tested in tissue extracts with Amplex Red). Finally, hydrostimulation prior to gravistimulation attenuated thegravistimulated asymmetric ROS and auxin signals that are required for gravity-directed curvature. We suggest that ROS,presumably H2O2, function in tuning root tropic responses by promoting gravitropism and negatively regulating hydrotropism.

Plants evolved the ability to sense and respond tovarious environmental stimuli in an integrated fashion.Due to their sessile nature, they respond to directionalstimuli such as light, gravity, touch, and moisture bydirectional organ growth (curvature), a phenomenontermed tropism. Experiments on coleoptiles conductedby Darwin in the 1880s revealed that in phototropism,the light stimulus is perceived by the tip, from which asignal is transmitted to the growing part (Darwin andDarwin, 1880). Darwin postulated that in a similarmanner, the root tip perceives stimuli from the envi-ronment, including gravity and moisture, processesthem, and directs the growthmovement, acting like “thebrain of one of the lower animals” (Darwin and Darwin,

1880). The transmitted signal in phototropism and gravi-tropism was later found to be a phytohormone, and itsredistribution on opposite sides of the root or shoot washypothesized to promote differential growth and bendingof the organ (Went, 1926; Cholodny, 1927). Over the years,the phytohormone was characterized as indole-3-aceticacid (IAA, auxin; Kögl et al., 1934; Thimann, 1935), andthe ‘Cholodny-Went’ theory was demonstrated forgravitropism and phototropism (Rashotte et al., 2000;Friml et al., 2002). In addition to auxin, second messen-gers such as Ca2+, pH oscillations, reactive oxygen species(ROS) and abscisic acid (ABA) were shown to play anessential role in gravitropism (Young and Evans, 1994;Fasano et al., 2001; Joo et al., 2001; Ponce et al., 2008).Auxin was shown to induce ROS accumulation duringroot gravitropism, where the gravitropic bending is ROSdependent (Joo et al., 2001; Peer et al., 2013).

ROS such as superoxide and hydrogen peroxidewere initially considered toxic byproducts of aerobicrespiration but currently are known also for their es-sential role in myriad cellular and physiological pro-cesses in animals and plants (Mittler et al., 2011). ROSand antioxidants are essential components of plant cellgrowth (Foreman et al., 2003), cell cycle control, andshoot apical meristem maintenance (Schippers et al.,2016) and play a crucial role in proteinmodification andcellular redox homeostasis (Foyer and Noctor, 2005).ROS function as signal molecules by mediating bothbiotic- (Sagi and Fluhr, 2006; Miller et al., 2009) and

1 This work was financially supported by the I-CORE Program ofthe Planning and Budgeting Committee and The Israel Science Foun-dation (grant no. 757/12).

2 These authors contributed equally to the article.* 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:Hillel Fromm ([email protected]).

G.K. and D.S. designed and performed experiments, analyzed thedata, and wrote the manuscript; H.F. and G.M. supervised experi-ments and participated in writing the manuscript.

[OPEN] Articles can be viewed without a subscription.www.plantphysiol.org/cgi/doi/10.1104/pp.16.00660

Plant Physiology�, October 2016, Vol. 172, pp. 1209–1220, www.plantphysiol.org � 2016 American Society of Plant Biologists. All Rights Reserved. 1209

https://plantphysiol.orgDownloaded on January 7, 2021. - Published by Copyright (c) 2020 American Society of Plant Biologists. All rights reserved.

http://orcid.org/0000-0003-3498-4362

http://orcid.org/0000-0002-7196-6529

http://orcid.org/0000-0003-1977-1211

mailto:[email protected]

http://www.plantphysiol.org

mailto:[email protected]

http://www.plantphysiol.org/cgi/doi/10.1104/pp.16.00660

https://plantphysiol.org

abiotic- (Kwak et al., 2003; Sharma and Dietz, 2009)stress responses. Joo et al. (2001) reported a transientincrease in intracellular ROS concentrations early in thegravitropic response, at the concave side of maize roots,where auxin concentrations are higher. Indeed, thisasymmetric ROS distribution is required for gravitropicbending, since maize roots treated with antioxidants,which act as ROS scavengers, showed reduced gravi-tropic root bending (Joo et al., 2001). The link betweenauxin and ROS production was later shown to involvethe activation of NADPH oxidase, a major membrane-bound ROS generator, via a PI3K-dependent pathway(Brightman et al., 1988; Joo et al., 2005; Peer et al., 2013).Peer et al. (2013) suggested that in gravitropism, ROSbuffer auxin signaling by oxidizing the active auxinIAA to the nonactive and nontransported form, oxIAA.

Gravitropic-oriented growth is the default growthprogram of the plant, with shoots growing upwards androots downward. However, upon exposure to specificexternal stimuli, the plant overcomes its gravitropicgrowth program and bends toward or away from thesource of the stimulus. For example, as roots respond tophysical obstacles or water deficiency. The ability of rootsto direct their growth toward environments of higherwater potential was described byDarwin and even earlierand was later defined as hydrotropism (Von Sachs, 1887;Jaffe et al., 1985; Eapen et al., 2005).

In Arabidopsis (Arabidopsis thaliana), wild-type seed-lings respond to moisture gradients (hydrostimulation)by bending their primary roots toward higher water po-tential. Upon hydrostimulation, amyloplasts, the starch-containing plastids in root-cap columella cells, whichfunction as part of the gravity sensing system, are de-graded within hours and recover upon water replenish-ment (Takahashi et al., 2003; Ponce et al., 2008; Nakayamaet al., 2012). Moreover, mutants with a reduced responseto gravity (pgm1) and to auxin (axr1 and axr2) exhibithigher responsiveness to hydrostimulation, manifestedas accelerated bending compared to wild-type roots(Takahashi et al., 2002, 2003). Recently, we have shownthat hydrotropic root bending does not require auxin re-distribution and is accelerated in the presence of auxinpolar transport inhibitors and auxin-signaling antagonists(Shkolnik et al., 2016). These results reflect the competi-tion, or interference, between root gravitropism and hy-drotropism (Takahashi et al., 2009). However, whichcellular signals participate in the integration of the dif-ferent environmental stimuli that direct root tropic cur-vature is still poorly understood.Herewe sought to assessthe potential role of ROS in regulating hydrotropism andgravitropism in Arabidopsis roots.

RESULTS

Different Spatial and Temporal ROS Patterns Occur inRoots in Response to Hydrostimulationand Gravistimulation

To investigate the role of ROS signals in tropic re-sponses we first assessed the spatial distribution of

ROS in Arabidopsis roots responding to gravitropicstimulation. Wild-type Arabidopsis seedlings grownvertically on agar-based medium (“Materials andMethods”) were gravistimulated by a 90° rotationand monitored for their ROS distribution by apply-ing dihydrorhodamine-123 (DHR), a rhodamine-basedfluorescent probe mostly sensitive to H2O2 (Gomeset al., 2005) that is often used in monitoring intracellu-lar, cytosolic ROS (Royall and Ischiropoulos, 1993;Crow, 1997; Douda et al., 2015). DHR staining wasdetected in the columella, lateral root cap, epidermallayer of elongation zone (EZ), and the vasculature andwas weaker at the meristematic zone (Fig. 1). Thispattern is similar to previously reported staining pat-terns obtained byH2O2-specific dyes in primary roots ofArabidopsis (Dunand et al., 2007; Tsukagoshi et al.,2010; Chen andUmeda, 2015) and of other plant species(Ivanchenko et al., 2013; Xu et al., 2015). One to twohours post gravistimulation, ROS asymmetric distri-bution, higher at the concave (bottom side of the root)was apparent in the epidermal layer of the distal elon-gation zone (DEZ), where the bending initiates (Fig.1A). The asymmetric ROS distribution dissipated afteranother 2 h (Fig. 1, A and D), in accordance with pre-vious reports (Joo et al., 2001; Peer et al., 2013).

To study ROS dynamics during hydrotropic growth,wild-type seedlings were introduced into a moisturegradient in a closed CaCl2-containing chamber (hereinreferred to as the CaCl2 / dry chamber) as previouslydescribed (Takahashi et al., 2002; Kobayashi et al., 2007;Shkolnik et al., 2016). Under this system, root bendingupon hydrostimulation initiates at a region more dis-tant from the root tip compared to root bending bygravitropism. The distances of curvature from the roottip for hydrotropism and gravitropism were 601.2 618.1 mm and 365.16 13.1 mm, respectively (mean6 SE),2 h post stimulation (n = 29). We therefore designatedthe region of gravitropic bending initiation as the DEZand the region of hydrotropic bending initiation as thecentral elongation zone (CEZ), in accordance withprevious definitions (Fasano et al., 2001; Massa andGilroy, 2003). Furthermore, during the hydrotropic re-sponse, the root tip keeps facing downward in responseto gravity, where a slight curvature is detected in theDEZ (Fig. 1B, 1, 2, and 4 h, concave side is indicated).Interestingly, during hydrotropic growth, ROS do notform an asymmetric distribution pattern at the DEZ, incontrast to the gravity-induced ROS asymmetric dis-tribution (Fig. 1, B and D). However, asymmetric dis-tribution of ROS appears at the CEZ, where thehydrotropic root curvature takes place and detectedROS levels are lower (Fig. 1, B and D). This unequaldistribution of ROS appears, however, also in roots thatwere subjected to nonhydrostimulating conditions(obtained by adding distilled water to the bottom thechamber), which do not undergo hydrotropic bending(Fig. 1C). Under these experimental conditions, a higherROS level was measured at the side of the root facingthe agar medium (Fig. 1C, arrowhead). The CEZ-located asymmetric distribution is not dynamic, and is

1210 Plant Physiol. Vol. 172, 2016

Krieger et al.



Figure 1. ROS spatial and temporal distribution patterns during root gravitropism and hydrotropism. A, B, C and E) Confocalmicroscopy of 5-d old seedlings stained with DHR, a ROS-sensitive fluorescent dye. Images were pseudo-colored; red indicateshigher ROS-dependent fluorescence intensity. Scale bars = 100 mm. White lines next to the root mark defined root zones(designated according to Fasano et al., 2001). g represents gravity vector, C represents water potential gradient. Concave and

Plant Physiol. Vol. 172, 2016 1211

ROS Tune Root Tropisms



maintained throughout the first 4 h of the hydrotropicresponse without a significant change in the ratio levelbetween the two sides of the root (Fig. 1, B and D). Wesuspected that this asymmetric distribution of ROSmaybe caused by the mechanical tension formed as the rootbends around the agar bed. To further test this, weused the split-agar / sorbitol system (“Materials andMethods”) for assessing ROS distribution during hy-drotropism. In this experimental system, no asymmet-ric ROS distribution could be detected in response tohydrostimulation in the DEZ or CEZ (Fig. 1, E and F).Moreover, we detected no changes in the overall in-tensity of DHR fluorescence at the indicated time pointsin both hydrostimulated and gravistimulated roots(Supplemental Fig. S1). Collectively, these results de-pict distinct dynamics and spatial patterns of ROS

distribution during gravitropic and hydrotropic re-sponses, which may imply different roles of ROS inthese tropic responses. We note that strong DHR fluo-rescence is detected in the root vasculature above theCEZ at all time points, similar to previous reports(Tsukagoshi et al., 2010; Chen and Umeda, 2015).

ROS Tune Root Tropic Responses

To assess the possible role of ROS in hydrotropismcompared to gravitropism, we tested whether ROSscavenging molecules or ROS-generation inhibitors af-fect hydrotropic growth. As described previously, theantioxidant ascorbic acid (ascorbate) has an inhibitoryeffect on root gravitropism (Joo et al., 2001; Peer et al.,

Figure 1. (Continued.)convex sides of the root are indicated. Arrowheads point to regions where the fluorescence signal distributes unevenly betweenthe two sides of the root. A, Under gravistimulation, an asymmetric distribution of ROS was apparent 2 h post stimulation anddissipated after another 2 h. This asymmetry was detected at the DEZwhere higher ROS levels were observed at the concave sideof the root. B, Under hydrostimulation, ROS distribute asymmetrically at the CEZ but maintain symmetric distribution at the DEZ.C, The asymmetric ROS pattern at the CEZ was also observed in roots that were exposed to nonhydrostimulating conditions anddo not bend hydrotropically. The higher ROS level was observed at the side that is in contact with the agar medium. D, Quan-tification of DHR fluorescence, measured at the epidermal layer in two regions of the root EZ (in the DEZ of gravistimulated rootsand in the DEZ and CEZ of hydrostimulated roots). The data are presented as the ratio between the signal at the concave and theconvex sides of the root. Error bars represent mean 6 SE (three biological independent experiments, 14 , n , 23). **P , 0.01,Student’s t test versus start time. E, Roots were hydrostimulated for the indicated times using the split-agar / sorbitol system. F,Quantification of DHR fluorescence, measured at the DEZ epidermal layer (200 mm above apex) and CEZ (600 mm above apex).The data are presented as the ratio between the signal at the concave and the convex sides of the root. Error bars represent mean6 SE

(three biological independent experiments, n = 20). No significant difference was found among different hydrostimulation times(Tukey-HSD post hoc-test, P , 0.05).

Figure 2. Ascorbate accelerates root hydrotropic growth, and a mutant deficient in APX1 shows attenuated hydrotropic bending.A, Seedlings performing hydrotropic bending 2.5 h post hydrostimulation in the presence or absence of 1mM sodium ascorbate. Inboth A and C, g represents gravity vector,C represents water potential gradient, Scale bar = 1 mm. B, Root curvature kinetics andgrowth rate of ascorbate-treated hydrostimulated seedlings. Root curvature was measured at 1-h intervals for 7 h followinghydrostimulation. Root growth rate was determined by measuring the length at the beginning and at the end of the experiment.Error bars represent mean 6 SE (three biological independent experiments, 10 seedlings each). *P , 0.05, Student’s t test forindependent measurements. C, Root hydrotropic bending of wild type and apx1-2, 5 h post hydrostimulation. D, Root curvaturekinetics and growth rate of apx1-2 and wild type hydrostimulated seedlings. Root curvature and root growth rate were measuredas described in B.


Krieger et al.


http://www.plantphysiol.org/cgi/content/full/pp.16.00660/DC1


2013). Indeed, our results show gravitropic bendinginhibition in the presence of 1 mM ascorbate, a concen-tration that we found to significantly reduce ROS levelat the root tip (Supplemental Fig. S2). Root curvature incontrol conditions was 64.9 6 2.6 degrees, whereas inthe presence of ascorbate only 49.1 6 5.2 degrees(mean 6 SE) 8 h post gravistimulation (P = 0.011, Stu-dent’s t test for independent measurements), withoutdifferences in root growth rates (Supplemental Fig. S3).In contrast, application of 1 mM ascorbate acceleratedhydrotropic root bending. Root curvature in theCaCl2 / dry chamber was 27.26 2.6 degrees in controlconditions, whereas in the presence of ascorbatecurvature was 39.36 3.5 degrees (mean6 SE) 2 h posthydrostimulation (P = 0.01, Student’s t test for inde-pendent measurements) and reduced root growthrate by 29.4% (Fig. 2, A and B). The same trend wasapparent when 1 mM of the antioxidant N-Acetyl-Cyswas applied (not shown).To further study the effect of ascorbate metabolism

on hydrotropism we tested mutants deficient in themost abundant cytosolic ascorbate peroxidase, Ascor-bate Peroxidase 1 (APX1; Davletova et al., 2005). apx1-2seedlings exhibited attenuated hydrotropic bendingcompared to wild type. Root curvature in the CaCl2 /dry chamber of wild type was 72.0 6 2.8 degrees,whereas that of apx1-2was 55.86 3.5 degrees (mean6 SE)5 h post hydrostimulation (P = 9.6 * 10

24, Student’s ttest for independent measurements), with no differ-ences in their growth rates (Fig. 2, C and D). These

results were reproduced using the split-agar / sorbitolsystem in which the ascorbate was supplemented to thesorbitol agar slice, allowing diffusion of the chemi-cals toward the root tip so that the exposure to as-corbate occurs while a water potential gradient isformed (Takahashi et al., 2002; Antoni et al., 2016;Supplemental Fig. S4, A and B). These data stronglysuggest that the reduced ability to scavenge cytosolicH2O2 inhibited root hydrotropic bending. Unlikeascorbate-treated seedlings, gravitropic bending wasnot impaired or promoted in the apx1-2 mutant(Supplemental Fig. S7).

ROS Generation by NADPH Oxidase Has Opposite Effectson Different Root Tropic Responses

To further study the roles of ROS in root tro-pisms, we tested the effects of diphenylene iodonium(DPI), an inhibitor of NADPH oxidase and otherflavin-containing enzymes (Foreman et al., 2003), onhydrotropic- and gravitropic-bending kinetics and thecorresponding ROS distribution patterns in primaryroots. NADPH oxidase is a plasma membrane-boundenzyme that produces superoxide (O2

c–) to the apoplast(Sagi and Fluhr, 2006). Superoxide is rapidly convertedto H2O2, which may enter the cell passively or throughaquaporins (Miller et al., 2010; Mittler et al., 2011).Application of DPI accelerated hydrotropic root bend-ing but attenuated gravitropic root bending (Fig. 3). In

Figure 3. Application of DPI, an NADPHoxidase inhibitor, accelerates hydrotropismwhile delaying gravitropism. A, Applica-tion of 10 mM DPI to the growing mediumpromotes hydrotropic curvature (first twoleft panels), and impedes gravitropic cur-vature (two right panels). Images weretaken 2 h post hydrostimulation (scale bar,1 mm) and 12 h post gravistimulation (scalebar = 5 mm). g represents gravity vector, Crepresents water potential gradient. B, Rootcurvature was measured at 1-h intervalsfor 6 h following hydrostimulation andat 2-h intervals for 12 h following gravi-stimulation. Error bars represent mean6 SE

(three biological independent experiments,10 seedlings each). C, DPI inhibits rootgrowth in both physiological assays. Rootgrowth rate was determined by measuringlength at the beginning and at the end of theexperiment. Error bars represent mean6 SE

(three biological independent experiments,10 seedlings each). **P , 0.01, t test forindependent measurements.









response to hydrostimulation, root bending was accel-erated in the presence of DPI, showing 86.3 6 2.1 de-grees curvature (mean6 SE) in the CaCl2 / dry chamberafter only 4 h, even though root growth rate wasinhibited by 65.3% (Fig. 3). This result was reproducedusing the split-agar / sorbitol system (SupplementalFig. S4A).

Fluorescent ROS staining of DPI-treated rootsrevealed elimination of ROS from the epidermal layerof the EZ and further along the root, where ROS at theouter layers (epidermis and cortex) seemed to dropdown and the remaining ROS appeared in the vascu-lature and its surrounding layers (Fig. 4, A and B). ROSelimination at the outer root cell layers was previouslydescribed for hydroxyphenyl fluorescein-staining uponDPI treatment (Dunand et al., 2007). Along with de-creased fluorescence at the EZ, we detected an increaseof DHR fluorescence intensity at the meristematic zoneof DPI-treated roots (Fig. 4). Dunand et al. (2007) usedNitroblue tetrazolium (NBT) for assessing extracellular

O2c– levels in Arabidopsis root tips and detected a de-

crease in NBT intensity upon DPI treatment. Since theDHR probe is mostly sensitive to cytosolic H2O2(Gomes et al., 2005), our results do not contradict pre-viously reported results.

Gravistimulated seedlings that were pretreated for2 h with DPI showed less ROS accumulation and con-sequently no ROS asymmetric distribution in theepidermal layer of the EZ, resulting in a delayedgravitropic response (Fig. 4C). Similarly, seedlings thatwere hydrostimulated in the presence of DPI showedelimination of ROS from the epidermal layer at thebending region, which became more proximal to theroot tip (Fig. 4D). Interestingly, the gravity-directedcurvature of the root tip, which occurs during hy-drotropic root bending, appeared to be attenuated inascorbate- and DPI-treated seedlings (Figs. 2A and 4D).This finding demonstrates again the negative effect ofROS elimination on root gravitropism, also in combi-nation with a hydrotropic response.

Figure 4. DPI eliminates ROS levels at the epidermal layer of the root EZ and elevates ROS levels at the meristematic zone. A, C,andD, DHR fluorescence (in A, over bright field, in C andD, fluorescent channel only) of seedlings treated for 2 hwith 10mMDPIor DMSO for control. Scale bars = 100 mm in all confocal images. g represents gravity vector, C represents water potentialgradient. A) Images of unstimulated roots, pretreated for 2 hwith 10mMDPI or DMSO.DHR signal is more intense and penetratesto the deeper root layers due to longer incubation in the dye (5 min). Images are representatives of n = 23 seedlings. B, DHRfluorescence intensity of seedlings treated with DPI or DMSO for 2 h, measured at the epidermal layer of the EZ and at themeristematic zone. Error bars represent mean 6 SE (three biological independent experiments, n = 23 seedlings in total). *P ,0.05, **P , 0.01, t test for independent measurements. C, Seedlings pretreated with DPI for 2 h were gravistimulated and showless ROS accumulation and asymmetrical distribution at the EZ. Images shown here are of a more extraneous section of the root,where the differences betweenDPI treatment and control are highly detectable. Images are representatives of n = 11 seedlings. D,Seedlings that were hydrostimulated for 2 h on a DPI-containing medium showing elimination of the signal from the epidermallayer at the bending region, which became more proximal to the root tip. Images are representatives of n = 20 seedlings.


Krieger et al.





Hydrotropism Is Affected by Root NADPH Oxidase

To further assess the inhibitory effect of ROS gen-eration by NADPH oxidase on root hydrotropismwe tested transposon-insertion mutants of the plantNADPH oxidase - RBOH (Respiratory Burst OxidaseHomolog) gene family, which consists of 10members inArabidopsis. These can be divided into three classesbased on their tissue specificity: RBOH D and F arehighly expressed throughout the plant, RBOHA-G andI are expressed mostly in roots, and RBOH H and Jexpress specifically in pollen (Sagi and Fluhr, 2006).RBOHChas been intensively studied, and its activity inROS production in trichoblasts is essential for root hairelongation and mechanosensing (Foreman et al., 2003;Monshausen et al., 2009). It is expressed in trichoblastsand in the epidermal layer of the EZ (Foreman et al.,2003), though its role in the EZ is still unclear(Monshausen et al., 2009). When hydrostimulated inthe CaCl2 / dry chamber or in the split-agar / sorbitolsystems, rbohC seedlings exhibited accelerated hydro-tropic bending. Measured in the CaCl2 / dry chamber,root curvature in wild type was 46.4 6 3.1 degreescompared to 64.26 3.5 degrees in rbohC (mean6 SE) 2 h

post hydrostimulation (P = 5.1 * 1024, Student’s t test

for independent measurements) with no difference ingrowth rate compared to wild type (Fig. 5, A and B;Supplemental Fig. S4C; Supplemental Movie S1). Wethen examined the hydrotropic response of seedlingsdeficient in RBOH D, which has the highest expressionlevels among the RBOHs. RBOH D is expressed in allplant tissues but mainly in stems and leaves and isknown as a key factor in ROS systemic signaling (Sagiand Fluhr, 2006; Miller et al., 2009; Suzuki et al., 2011).Interestingly, rbohD seedlings did not exhibit signifi-cantly different hydrotropic bending kinetics or rootgrowth rates compared to wild type (Fig. 5, A and B;Supplemental Fig. S4C; Supplemental Movie S2). DHRstaining revealed no significant difference in ROS spa-tial patterns in gravistimulated nor hydrostimulated(using the CaCl2 / dry chamber or split-agar / sorbitolsystem) roots of the RBOH mutants compared to wildtype (Supplemental Fig. S5–S8). Therefore, to bettercharacterize endogenous ROS levels in root tissues ofwild type and rbohC and rbohD mutants, we appliedAmplex red for determination of H2O2 content in tissueextracts (“Materials and Methods”). When examiningextracts from whole seedlings, we observed a 68% and

Figure 5. rbohC, but not rbohD, show ac-celerated hydrotropic bending and lowerROS levels in the root apex. A, Root hy-drotropic growth of wild type, rbohC, andrbohD, 2 h post hydrostimulation. Scalebar = 1 mm. g represents gravity vector, Crepresents water potential gradient. B, Rootcurvature kinetics and growth rates. Rootcurvature was measured at 1-h intervalsfor 7 h following hydrostimulation. Rootgrowth rate was determined by measur-ing length at the beginning and at the endof the experiment. Error bars representmean 6 SE (three biological independentexperiments, 10 seedlings each). Statisticaldifference in root curvature was tested for2 and 5 h post hydrostimulation. C, Deter-mination of H2O2 content in root apices(1-2 mm from tip) and whole seedlings (D)of wild type, rbohD, and rbohC, measuredby the Amplex red assay (“Materials andMethods”). The fluorescent reads werenormalized to the amount of extractedprotein, measured by the Bradford assay.Error bars represent mean 6 SD (three bio-logical repeats with two technical repli-cates, for root apices, n = 60, for wholeseedlings, n= 20). The higher y-scale in C isa result of normalization to 10-fold lowerprotein level extracted from root apices. InB, C ,and D, meanswith different letters aresignificantly different (P , 0.05, TukeyHSD adjusted comparisons).










77% reduction in H2O2 levels in rbohD and rbohC, re-spectively, compared to wild type (Fig. 5D). We then ex-amined extracts from excised root apices (1–2 mm fromtip) and observed a relatively similar H2O2 content inwild-type and rbohD roots,while rbohCmutants showed a57% reduction in H2O2 content compared to wild type(Fig. 5C). These results are consistent with the tissue-specific expression pattern of the two RBOHs, as RBOHC is highly expressed in roots, while RBOHD is not (Sagiand Fluhr, 2006) and with the accelerated hydrotropicphenotype of rbohC compared to rbohD and wt. Theirdifferent expression patterns could also be visualized inthe high-resolution spatiotemporal map (Brady et al.,2007) of the eFP browser (Winter et al., 2007).

The acceleration in hydrotropic root bending of rbohCis, however, weaker compared with that of DPI-treatedwild-type seedlings (measured in the CaCl2 / drychamber, root curvature in rbohC was 75.41 6 2.19 de-grees and root curvature of DPI-treated seedlings was86.31 6 2.11 degrees after 4 h of hydrostimulation,while wild type and dimethyl sulfoxide (DMSO)-treatedwild-type roots exhibited 63.27 6 2.38 and 62.67 63.17 degrees in that time, respectively). These resultsmay indicate partial functional redundancy with otherroot-expressed RBOHs, or involvement of other DPI-sensitive enzymes in this tropic growth. When treatedwith DPI, rbohC roots presented the same hydrotropic-bending kinetics as wild-type roots (not shown). UnlikeDPI-treated seedlings, RBOHC- and RBOHD-deficientmutants did not show inhibition or acceleration in theirgravitropic growth (Supplemental Fig. S8) or weakenedgravity-directed curvature of the root tip during hy-drotropic growth (Fig. 5) and gravitropic ROS asym-metric distribution as in the wild type (SupplementalFig. S7). These results may be explained by functionalredundancy between the root-expressed RBOH familymembers as well as by compensation of ROS signalingby mechanisms involved specifically in gravitropism.

Hydrotrostimulation Attenuates the Gravitropic ROS andAuxin Signals

To test a possible direct link between hydrotropismand gravitropism through ROS, we challenged wild-type seedlings with combined stimuli using the split-agar / sorbitol method (Fig. 6A). The split-agar systemallows slow and controlled exposure of the root tips toincreasing osmotic pressure, and by rotation of thechamber allows changes in the gravity vector (Fig. 6A).After 0 to 2 h of hydrostimulation, 1 h of gravi-stimulation induced a clear asymmetric ROS distribu-tion at the bending EZ. After 3 h of hydrostimulation,1 h of gravistimulation generated a weak asymmetric

Figure 6. Hydrotropism abrogates the gravitropic ROS signal. A,Schematic presentation of the assay applied to test ROS distribution atroot tips of hydrostimulated seedlings and a combined gravistimulationwith hydrostimulation. B, Roots were hydrostimulated for the indicatedtimes and then gravistimulation for 1 h, stained with DHR and imagedusing a confocal microscope (“Materials and Methods”). Images arepresented as pseudo color. Scale bar, 100mm.C,Quantification of DHRfluorescence, measured at the DEZ epidermal layer (200 mm aboveapex). The data are presented as the ratio between the signal at theconcave and the convex sides of the root. Error bars represent mean6 SE

(three biological independent experiments, n = 20). D, Root curvatureof 1 h gravistimulated seedlings following hydrostimulation for theindicated times. The 1 h gravitropic curvature following 0, 2, 3, and4 h hydrosimulation was 14.42˚ 6 1.27, 9.16˚ 6 0.76, 6.33˚ 6 0.78and23.14˚6 2.03, respectively. Error bars represent mean6 SE (threebiological independent experiments, n = 15). Negative value meanscurvature against the gravity vector direction. In A and B, C and grepresent the water potential gradient and gravity vector, respectively.

ROS images of hydrostimulated roots for the same indicated times,without gravistimulation are shown in Figure 1 E. In C and D, lettersabove bars represent statistically significant differences by Tukey-HSDpost hoc-test (P , 0.05).


Krieger et al.






ROS distribution (Fig. 6, B and C). Strikingly, follow-ing 4 h of hydrostimulation, 1 h of gravistimulationfailed to generate an asymmetric ROS distribution, andgravity-directed root bending was not observed (Fig. 6,B–D). These results indicate that as the osmotic stressstimulus increases and promotes hydrotropic curva-ture, gravistimulation is not sufficient to evoke typicalROS asymmetric distribution, and growth towardhigher water potential is favorable. Indeed, with in-creasing hydrostimulation time from 0 to 4 h prior togravistimulation, gravitropic curvature decreased (Fig.6D). Four hours of hydrostimulation prevented gravi-tropic curvature as roots responded only to the hy-drotropic stimulus (depicted as a negative curvatureangle in Fig. 6D).Subsequently, to assess whether the attenuation of

the ROS signal of gravistimulated roots followinghydrostimulation is associated with the attenuation ofauxin distribution, roots of DII-VENUS-expressingtransgenic seedlings (Brunoud et al., 2012) were gravi-stimulated for 1 h following exposure to an osmoticgradient for 0, 2, or 4 h (Supplemental Fig. S9).With thisauxin reporter, lower levels of DII-VENUS fluorescenceindicate higher levels of auxin. In agreement with theROS signal dynamics, we observed asymmetric auxindistribution in the lower part of the root tip (concave)in roots that were gravistimulated with no priorhydrostimulation, or following 2 h of hydrostimulation(Supplemental Fig. S9), as previously demonstrated ingraviresponding roots (Band et al., 2012). However,hydrostimulation for 4 h prior to gravistimulation im-paired the generation of an auxin gradient across theroot tip (Supplemental Fig. S9). Based on the knownrelationship between auxin and ROS in gravistimulation,these results may suggest that hydrotropic stimula-tion attenuates the gravitropic ROS signal throughthe interruption of auxin distribution. However, wecannot exclude the possibility that hydrostimulationattenuates gravistimulated ROS and auxin distribu-tion through independent signaling pathways thatare yet to be elucidated.

DISCUSSION

To perform hydrotropic bending, a root must over-come its gravity-directed growth (Eapen et al., 2005;Takahashi et al., 2009). Our results suggest oppositeroles for ROS in hydrotropic and gravitropic growthbehaviors. When treated with ascorbate, an antioxi-dant, or DPI, an inhibitor of NADPH oxidase and otherflavin-containing enzymes (Foreman et al., 2003), Ara-bidopsis primary roots exhibit opposite changes in theirbending kinetics in response to the different stimula-tions, namely, delay in gravitropism and acceleration inhydrotropism (Figs. 2 and 3; Supplemental Figs. S3 andS4). The antagonism between these two responses wasshown previously for the agravitropic pea mutant(ageotropum), whose lack of gravity response contrib-utes to its hydrotropic responsiveness (Takahashi and

Suge, 1991). Amyloplast degradation at early stages of ahydrotropic response may also be a mechanism bywhich the root eliminates its sense of gravity in order toperform nongravitropic growth (Takahashi et al., 2003;Ponce et al., 2008). When examining the ROS and auxinpatterns in response to combined stimuli by first ap-plying hydrostimulation and afterward applying bothhydro- and gravistimulation, we observed a reductionin gravity-directed ROS-asymmetry and auxin-gradientwhen the duration of hydrostimulation is increased(Fig. 6; Supplemental Fig. S9). We therefore concludethat during hydrotropic growth, the root actively at-tenuates gravitropic auxin and ROS signaling to over-come gravitropic growth.

In gravitropism, auxin is required for ROS produc-tion (Joo et al., 2005; Peer et al., 2013). In contrast, nei-ther auxin redistribution nor auxin signaling arerequired for hydrotropic bending (Shkolnik et al.,2016). Moreover, inhibition of polar auxin transportor Transport Inhibitor Response 1-dependent signalingaccelerates hydrotropism (Shkolnik et al., 2016). Con-sistent with these observations, asymmetric distribu-tion of ROS was not detected in the DEZ duringhydrotropism. In gravitropism, however, both an auxingradient at the lateral root cap, and ROS asymmetricdistribution at the DEZ is formed transiently. Collec-tively, these results demonstrate the antagonism be-tween hydro- and gravitropism with respect to auxinand ROS signaling.

Asymmetric ROS distribution was however ob-served in the CEZ of hydrostimulated roots in theCaCl2 / dry chamber system, and its asymmetry ratiolevel has not changed during the measured time points(Fig. 1, B and D). This asymmetric pattern, that is,higher ROS levels at the side of the root that is in contactwith the agar medium, was also present in roots thatwere exposed to nonhydrostimulating conditions anddo not perform hydrotropic bending (Fig. 1, C and D).Therefore, this nontransient unequal distribution ofROS in the CEZ may be a result of mechanosensing-induced ROS (Monshausen et al., 2009) at the regionwhere the root detaches from the agar medium. Indeed,no ROS asymmetry was observed in roots exposed toa water-potential gradient in the split-agar / sorbitolsystem (Fig. 1, E and D), where the root does not en-counter mechanical tension by the agar due to bending.Therefore, it is clear that hydrotropism does not involveasymmetric distribution of ROS. Yet it attenuatesgravity-directed asymmetric ROS distribution.

In addition to their roles as intracellular signalingmolecules, ROS function in several apoplastic pro-cesses, including cell wall rigidification that is thoughtto restrict cell elongation (Hohl et al., 1995;Monshausenet al., 2007). It is tempting to hypothesize that in gravi-tropism, the higher levels of ROS in the concave side ofthe root promote root bending by inhibition of cellelongation at this side. However, this hypothesis fails toexplain the opposite effects of antioxidants and ROS-generator inhibitors on gravi- and hydrotropism, asdifferential cell elongation is needed in both cases.











In this study, we show that ROS, presumably cyto-solic H2O2 in the epidermal layer of the root EZ, nega-tively regulate hydrotropic bending. The activity ofRBOH Cwas characterized as essential for this process,since rbohC mutants showed accelerated hydrotropicroot bending and lower levels of H2O2 in the root apex(Fig. 5). This, however, does not exclude the possiblecontribution of other root-expressed RBOHs or otherflavin-containing enzymes to the process. The locali-zation of ROS-generating enzymes of the RBOH familyhas substantial effects on the tissue-specific ROS levelsand the consequent hydrotropic root curvature, as itappears that in mutants deficient in RBOH D, which isexpressed throughout the plant but mostly in leavesand stems (Suzuki et al., 2011) ROS levels in the rootapex and hydrotropic curvaturewere similar to those ofwild type (Fig. 5; Supplemental Fig S3). As for ROSscavenging enzymes, we detected a weak hydrotropicroot bending in apx1-2 mutants (Fig. 2; SupplementalFig S4), which lack the function of the abundant cyto-solic H2O2-scavenging enzyme APX1 and are thus ex-pected to accumulate higher H2O2 levels in all planttissues. Peroxidases were shown to play an importantrole in root development and growth control (Dunandet al., 2007) by modifying O2

c– to H2O2 at the transition-to- EZ (Tsukagoshi et al., 2010). Our observations areconsistent with this ROS type-specific accumulationpattern, and add a new aspect to the role of H2O2 at theroot EZ.

The phytohormone ABA was previously reportedas a positive regulator of root hydrotropism. Arabi-dopsis mutants deficient in ABA sensitivity (abi2-1)and ABA biosynthesis (aba1-1) were reported as less re-sponsive to hydrostimulation, whereas ABA treatmentrescued the delayed hydrotropic phenotype of aba1-1(Takahashi et al., 2002). ABA signaling involves theactivation of Pyrabactin Resistance/PYR1-like (PYR/PYL) receptors that mediate the inhibition of clade Aphosphatases type 2C (PP2C), which are negativeregulators of the pathway (Antoni et al., 2013). Theinvolvement of this pathway in root hydrotropismwas demonstrated recently, as a pp2c-quadruple mu-tant exhibited an ABA-hypersensitive phenotype andconsequently enhanced hydrotropic response, while amutant deficient in six PYR/PYL receptors exhibitedinsensitivity to ABA treatment and to hydrotropicstimulation (Antoni et al., 2013). Since ABA was shownto induce stomata closure through the activation of theNADPH oxidases RBOH D and RBOH F (Kwak et al.,2003), it is tempting to hypothesize that ABA activatesROS production in root-expressed NADPH oxidasesduring hydrotropic growth. A candidate mediator forthis process may be PYL8, since PYL8-deficientmutants(pyl8-1 and pyl8-2) exhibited a nonredundant ABA-insensitive root growth when treated with ABA, andtranscriptional fusion of PYL8 (ProPYL8:GUS) revealedits expression in the stele, columella, lateral root cap,and root epidermis cells (Antoni et al., 2013). The latterexpression region overlaps with that of RBOH C(Foreman et al., 2003). However, distinguished from

their role in stomata closure, ROS negatively regulatehydrotropism and thus may function in a negativefeedback to ABA signaling. Antagonism between ROSand ABA also appears in seed germination, as H2O2breaks ABA-induced seed dormancy in several plantspecies (Sarath et al., 2007).

In the context of integration of environmental stimuliby the root tip (Darwin and Darwin, 1880), we suggestthat ROS, presumably cytosolic hydrogen peroxide,fine tune root tropic responses by acting as positiveregulators of gravitropism and as negative regulators ofhydrotropism. Root hydrotropism and gravitropismdiffer in several aspects, such as the time of response(Eapen et al., 2005), the region of bending initiation(reported in this study), the involvement of auxin(Kaneyasu et al., 2007; Shkolnik et al., 2016), and theeffect of ROS on the response kinetics. To elucidate theeffects of ROS on tropic responses, their downstreameffectors in gravitropism and hydrotropism need to becharacterized.

MATERIALS AND METHODS

Plant Material and Growth Conditions

Wild-type Arabidopsis (Arabidopsis thaliana) Col-0 and T-DNA/Transposoninsertion mutants: rbohC (rhd2), rbohD (Miller et al., 2009), and apx1-2(SALK_000249; Suzuki et al., 2013) were used in this research. For vapor ster-ilization, seeds were put inside a desiccator next to a glass beaker containing25mLwater, 75mL bleach, and 5mLHCl for 2 h. Sterilized seeds were sown on12 x 12 cm squared petri dishes, containing 2.2 g/L Nitsch and Nitsch medium(Duchefa Biochemie B.V., Haarlem, The Netherlands) titrated to pH 5.8, 0.5%(w/v) Suc supplemented with 1% (w/v) plant agar (Duchefa) and vernalizedfor 1 d in 4°C in dark. Plates were put vertically in a growth chamber at 22°Cand daylight (100 mE m22 sec21) under 16-h-light/8-h-dark photoperiod. Theroot hair-deficient phenotype of rbohC was observed when grown on pH5-titrated growthmedium. Treatments with 10 mMDPI (Diphenyleneiodoniumchloride, Sigma) dissolved in DMSO, 1 mM sodium ascorbate (Sigma) dissolvedin distilled water, and 1mM N-acetyl-Cys (Acros organics) dissolved in distilledwater were performed by applying these chemicals in the agar medium. As-corbate treatment for DHR staining was performed by transferring seedlingsonto 1 mm Whatman filter paper 0.253 Murashige and Skoog medium(Murashige and Skoog, 1962) and the indicated ascorbate concentrations.

Hydrotropic Stimulation Assays

ACaCl2 dry chamber was designed based on a previously described system(Takahashi et al., 2002; Kobayashi et al., 2007; Shkolnik et al., 2016) with thefollowing modifications: Plates were prepared as described in “Plant Materialand Growth Conditions” with or without supplemented chemicals, as indi-cated. The medium was cut 6 cm from the bottom and 5- to 7-d-old seedlingswere transferred to the cut medium, such that approximately 0.2 mm of theprimary root tip was bolting from the agar into air. Twelve mL of 40% CaCl2(w/v) (Duchefa) was put at the bottom of the plate, which was then closed,sealed with Parafilm and placed vertically under 30 mE m22 sec21 white light.As control, nonhydrostimulating conditions were achieved by adding 20 mL ofdistilled water to the bottom of the plate. In this system, the roots were exposedto the supplemented chemical at the beginning of the experiment. Hydro-stimulation was performed also using the previously described split-agarmethod (Takahashi et al., 2002; Antoni et al., 2016). Ascorbate, DPI or DMSO(control) were added directly to the sorbitol containing gel slice. Root tips wereimaged at indicated time points using Nikon D7100 camera equipped withAF-S DXMicro NIKKOR 85 mm f/3.5G ED VR lens (Nikon, Tokyo, Japan). Forroot curvature measurements and supplemental movies of the humidity-gradient system, plates were faced approximately 45° to the lens, and multi-ple photos with changing focus were obtained using Helicon remote software,and stacked using Helicon focus software (www.heliconsoft.com). Root


Krieger et al.





http://www.heliconsoft.com


curvature and growth were analyzed using ImageJ software 1.48V (WayneRasband, NIH).

Gravitropic Stimulation Assay

Five- to seven-day-old seedlings were transferred to a standard medium, orascorbate containing medium, following 1 h of acclimation at original growthorientation before the plateswere 90° rotated. ForDPI treatment, seedlingswerepretreated in DMSO or 10 mMDPI-containing media for 2 h, then transferred toanother plate containing standard medium, followed by 30 min acclimation atthe original growth orientation before the plates were rotated by 90°.

Confocal Microscopy

For ROS detection, seedlings were immersed in 86.5 mM [0.003% (w/v)]DHR (Sigma) dissolved in phosphate buffer saline (31, pH 7.4) for 2 or 5 min,after hydrotropic or gravitropic stimulation assays. Fluorescent signals in rootswere imaged with a Zeiss LSM 780 laser spectral scanning confocal microscope(Zeiss, http://corporate.zeiss.com), with a 103 air (EC Plan-Neofluar 103/0.30 M27) objective. Acquisition parameters were as follows: master gain wasalways set between 670 and 720, with a digital gain of 1, excitation at 488 nm(2%) and emission at 519 to 560 nm. Signal intensity was quantified as meangray value using ImageJ software. Confocal images were pseudo-colored usingthe RGB look-up table of the ZEN software, for easier detection of the fluo-rescent signal distribution in the root. Imaging of DII-VENUS expressing rootswas performed as previously described (Shkolnik et al., 2016).

Determination of H2O2 in Tissue Extracts

Whole seedlings (n= 20 seedlings) and root apices (1–2mm from root tip, n =60 seedlings) were frozen in liquid nitrogen and homogenized in phosphatebuffer saline (31, pH 7.4; 600mL forwhole seedlings and 150mL for root apices),centrifuged in 10,000 g for 5 min in 4°C, and the supernatant was used as thetissue extract. H2O2 levels in the extracts were measured using the Amplex redassay kit (Molecular Probes, Invitrogen) according to the manufacturer’s pro-tocol, with three biological repeats and two technical replicates. Samples weremeasured with a Synergy HT fluorescence plate reader (BioTek) using 530/590-nm excitation/emission filters. Protein levels in the extracts were deter-mined using the Bradford reagent (Bio-Rad). The absorbance was read in thesame plate reader using a 595-nm filter. Fluorescence reads were then nor-malized to the protein amount.

Statistical Analysis

Results were analyzed using MS Excel ToolPak and R version 3.1.1.

Supplemental Materials

The following supplemental materials are available.

Supplemental Figure S1. Relative DHR fluorescence intensity ingravistimulated and hydrostimulated roots.

Supplemental Figure S2. ROS level is reduced by ascorbate.

Supplemental Figure S3. The antioxidant ascorbate impedes root gravi-tropic response.

Supplemental Figure S4. Hydrostimulation using the split-agar / sorbitolmethod.

Supplemental Figure S5. ROS distribution during hydrotropic growth inwild type, rbohC, and rbohD mutants.

Supplemental Figure S6. ROS distribution in hydrostimulated wild type,rbohC, and rbohD mutants using the split-agar / sorbitol system.

Supplemental Figure S7. ROS distribution in gravistimulated wild type,apx1-2, rbhoC and rbhoD mutants.

Supplemental Figure S8. rbohC and rbohD exhibit normal gravitropicgrowth compared to wild type.

Supplemental Figure S9. Auxin distribution in gravistimulated root tipswith or without prior hydrostimulation.

Supplemental Movie S1. Hydrotropism of rbohC mutant compared towild type.

Supplemental Movie S2. Hydrotropism of rbohD mutant compared towild type.

ACKNOWLEDGMENTS

We thank Professor Robert Fluhr for critical reading of the manuscript, andlaboratory members Yosef Fichman and Roye Nuriel for helpful suggestions.

Received June 8, 2016; accepted August 15, 2016; published August 17, 2016.

LITERATURE CITED

Antoni R, Dietrich D, Bennett MJ, Rodriguez PL (2016) Hydrotropism:analysis of the root response to a moisture gradient. Environ ResponsesPlants Methods Protoc 3–9

Antoni R, Gonzalez-Guzman M, Rodriguez L, Peirats-Llobet M, PizzioGA, Fernandez MA, De Winne N, De Jaeger G, Dietrich D, BennettMJ, et al (2013) PYRABACTIN RESISTANCE1-LIKE8 plays an impor-tant role for the regulation of abscisic acid signaling in root. PlantPhysiol 161: 931–941

Band LR, Wells DM, Larrieu A, Sun J, Middleton AM, French AP, BrunoudG, Sato EM, Wilson MH, Péret B, et al (2012) Root gravitropism is regu-lated by a transient lateral auxin gradient controlled by a tipping-pointmechanism. Proc Natl Acad Sci USA 109: 4668–4673

Brady SM, Orlando DA, Lee J-Y, Wang JY, Koch J, Dinneny JR, Mace D,Ohler U, Benfey PN (2007) A high-resolution root spatiotemporal mapreveals dominant expression patterns. Science 318: 801–806

Brightman AO, Barr R, Crane FL, Morré DJ (1988) Auxin-stimulatedNADH oxidase purified from plasma membrane of soybean. PlantPhysiol 86: 1264–1269

Brunoud G, Wells DM, Oliva M, Larrieu A, Mirabet V, Burrow AH,Beeckman T, Kepinski S, Traas J, Bennett MJ, et al (2012) A novelsensor to map auxin response and distribution at high spatio-temporalresolution. Nature 482: 103–106

Chen P, Umeda M (2015) DNA double-strand breaks induce the expressionof flavin-containing monooxygenase and reduce root meristem size inArabidopsis thaliana. Genes Cells 20: 636–646

Cholodny NG (1927) Wuchshormone und Tropismen bei den Pflanzen. BiolZent 47: 604–626

Crow JP (1997) Dichlorodihydrofluorescein and dihydrorhodamine 123 aresensitive indicators of peroxynitrite in vitro: implications for intracel-lular measurement of reactive nitrogen and oxygen species. Nitric Oxide1: 145–157

Darwin C, Darwin F (1880) The power of movement in plants. John Murray,London

Davletova S, Rizhsky L, Liang H, Shengqiang Z, Oliver DJ, Coutu J,Shulaev V, Schlauch K, Mittler R (2005) Cytosolic ascorbate peroxidase1 is a central component of the reactive oxygen gene network of Ara-bidopsis. Plant Cell 17: 268–281

Douda DN, Khan MA, Grasemann H, Palaniyar N (2015) SK3 channel andmitochondrial ROS mediate NADPH oxidase-independent NETosis in-duced by calcium influx. Proc Natl Acad Sci USA 112: 2817–2822

Dunand C, Crèvecoeur M, Penel C (2007) Distribution of superoxide andhydrogen peroxide in Arabidopsis root and their influence on root de-velopment: possible interaction with peroxidases. New Phytol 174: 332–341

Eapen D, Barroso ML, Ponce G, Campos ME, Cassab GI (2005) Hy-drotropism: root growth responses to water. Trends Plant Sci 10:44–50

Fasano JM, Swanson SJ, Blancaflor EB, Dowd PE, Kao TH, Gilroy S (2001)Changes in root cap pH are required for the gravity response of theArabidopsis root. Plant Cell 13: 907–921

Foreman J, Demidchik V, Bothwell JHF, Mylona P, Miedema H, TorresMA, Linstead P, Costa S, Brownlee C, Jones JDG, et al (2003) Reactiveoxygen species produced by NADPH oxidase regulate plant cell growth.Nature 422: 442–446

Foyer CH, Noctor G (2005) Redox homeostasis and antioxidant signaling: ametabolic interface between stress perception and physiological re-sponses. Plant Cell 17: 1866–1875




http://corporate.zeiss.com













Friml J, Wi�sniewska J, Benková E, Mendgen K, Palme K (2002) Lateralrelocation of auxin efflux regulator PIN3 mediates tropism in Arabi-dopsis. Nature 415: 806–809

Gomes A, Fernandes E, Lima JLFC (2005) Fluorescence probes used for de-tection of reactive oxygen species. J Biochem Biophys Methods 65: 45–80

Hohl M, Greiner H, Schopfer P (1995) The cryptic-growth response ofmaize coleoptiles and its relationship to H2O2-dependent cell wallstiffening. Physiol Plant 94: 491–498

Ivanchenko MG, den Os D, Monshausen GB, Dubrovsky JG, Bednárová A,Krishnan N (2013) Auxin increases the hydrogen peroxide (H2O2) concen-tration in tomato (Solanum lycopersicum) root tips while inhibiting rootgrowth. Ann Bot (Lond) 112: 1107–1116

Jaffe MJ, Takahashi H, Biro RL (1985) A pea mutant for the study of hy-drotropism in roots. Science 230: 445–447

Joo JH, Bae YS, Lee JS (2001) Role of auxin-induced reactive oxygen spe-cies in root gravitropism. Plant Physiol 126: 1055–1060

Joo JH, Yoo HJ, Hwang I, Lee JS, Nam KH, Bae YS (2005) Auxin-inducedreactive oxygen species production requires the activation of phospha-tidylinositol 3-kinase. FEBS Lett 579: 1243–1248

Kaneyasu T, Kobayashi A, Nakayama M, Fujii N, Takahashi H, MiyazawaY (2007) Auxin response, but not its polar transport, plays a role in hy-drotropism of Arabidopsis roots. J Exp Bot 58: 1143–1150

Kobayashi A, Takahashi A, Kakimoto Y, Miyazawa Y, Fujii N, HigashitaniA, Takahashi H (2007) A gene essential for hydrotropism in roots. ProcNatl Acad Sci USA 104: 4724–4729

Kögl F, Haagen-Smit AJ, Erxleben H (1934) Über ein neues Auxin (“Hetero-auxin”) aus Harn. 11. Mitteilung über pflanzliche Wachstumsstoffe. HoppeSeylers Z Physiol Chem 228: 90–103

Kwak JM, Mori IC, Pei ZM, Leonhardt N, Torres MA, Dangl JL, BloomRE, Bodde S, Jones JDG, Schroeder JI (2003) NADPH oxidase AtrbohDand AtrbohF genes function in ROS-dependent ABA signaling in Arabi-dopsis. EMBO J 22: 2623–2633

Massa GD, Gilroy S (2003) Touch modulates gravity sensing to regulatethe growth of primary roots of Arabidopsis thaliana. Plant J 33: 435–445

Miller EW, Dickinson BC, Chang CJ (2010) Aquaporin-3 mediates hy-drogen peroxide uptake to regulate downstream intracellular signaling.Proc Natl Acad Sci USA 107: 15681–15686

Miller G, Schlauch K, Tam R, Cortes D, Torres MA, Shulaev V, Dangl JL,Mittler R (2009) The plant NADPH oxidase RBOHD mediates rapidsystemic signaling in response to diverse stimuli. Sci Signal 2: ra45

Mittler R, Vanderauwera S, Suzuki N, Miller G, Tognetti VB, VandepoeleK, Gollery M, Shulaev V, Van Breusegem F (2011) ROS signaling: the newwave? Trends Plant Sci 16: 300–309

Monshausen GB, Bibikova TN, Messerli MA, Shi C, Gilroy S (2007)Oscillations in extracellular pH and reactive oxygen species modulatetip growth of Arabidopsis root hairs. Proc Natl Acad Sci USA 104:20996–21001

Monshausen GB, Bibikova TN, Weisenseel MH, Gilroy S (2009) Ca2+regulates reactive oxygen species production and pH during mechano-sensing in Arabidopsis roots. Plant Cell 21: 2341–2356

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

Nakayama M, Kaneko Y, Miyazawa Y, Fujii N, Higashitani N, Wada S,Ishida H, Yoshimoto K, Shirasu K, Yamada K, et al (2012) A possibleinvolvement of autophagy in amyloplast degradation in columella cellsduring hydrotropic response of Arabidopsis roots. Planta 236: 999–1012

Peer WA, Cheng Y, Murphy AS (2013) Evidence of oxidative attenuation ofauxin signalling. J Exp Bot 64: 2629–2639

Ponce G, Rasgado F, Cassab GI (2008) How amyloplasts, water deficit androot tropisms interact? Plant Signal Behav 3: 460–462

Rashotte AM, Brady SR, Reed RC, Ante SJ, Muday GK (2000) Basipetalauxin transport is required for gravitropism in roots of Arabidopsis.Plant Physiol 122: 481–490

Royall JA, Ischiropoulos H (1993) Evaluation of 29,79-dichlorofluorescinand dihydrorhodamine 123 as fluorescent probes for intracellular H2O2in cultured endothelial cells. Arch Biochem Biophys 302: 348–355

Von Sachs J (1887) Lecture XXVII. Relations between growth and cell-division in the embryonic tissues. Lect plant Physiol clarendon PressOxford 431–459

Sagi M, Fluhr R (2006) Production of reactive oxygen species by plantNADPH oxidases. Plant Physiol 141: 336–340

Sarath G, Hou G, Baird LM, Mitchell RB (2007) Reactive oxygen species,ABA and nitric oxide interactions on the germination of warm-seasonC4-grasses. Planta 226: 697–708

Schippers JH, Foyer CH, van Dongen JT (2016) Redox regulation in shootgrowth, SAM maintenance and flowering. Curr Opin Plant Biol 29: 121–128

Sharma SS, Dietz K-J (2009) The relationship between metal toxicity andcellular redox imbalance. Trends Plant Sci 14: 43–50

Shkolnik D, Krieger G, Nuriel R, Fromm H (2016) Hydrotropism: rootbending does not require auxin redistribution. Mol Plant 9: 757–759

Suzuki N, Miller G, Morales J, Shulaev V, Torres MA, Mittler R (2011)Respiratory burst oxidases: the engines of ROS signaling. Curr OpinPlant Biol 14: 691–699

Suzuki N, Miller G, Sejima H, Harper J, Mittler R (2013) Enhanced seedproduction under prolonged heat stress conditions in Arabidopsisthaliana plants deficient in cytosolic ascorbate peroxidase 2. J Exp Bot64: 253–263

Takahashi H, Miyazawa Y, Fujii N (2009) Hormonal interactions duringroot tropic growth: hydrotropism versus gravitropism. Plant Mol Biol69: 489–502

Takahashi H, Suge H (1991) Root hydrotropism of an agravitropic peamutant, ageotropum. Physiol Plant 82: 24–31

Takahashi N, Goto N, Okada K, Takahashi H (2002) Hydrotropism inabscisic acid, wavy, and gravitropic mutants of Arabidopsis thaliana.Planta 216: 203–211

Takahashi N, Yamazaki Y, Kobayashi A, Higashitani A, Takahashi H(2003) Hydrotropism interacts with gravitropism by degrading amylo-plasts in seedling roots of Arabidopsis and radish. Plant Physiol 132:805–810

Thimann KV (1935) On the plant growth hormone produced by Rhizopussuinus. J Biol Chem 109: 279–291

Tsukagoshi H, Busch W, Benfey PN (2010) Transcriptional regulation ofROS controls transition from proliferation to differentiation in the root.Cell 143: 606–616

Went FW (1926) On growth-accelerating substances in the coleoptile ofAvena sativa. Proc K Ned Akad Wet 30: 10–19

Winter D, Vinegar B, Nahal H, Ammar R, Wilson GV, Provart NJ (2007)An “Electronic Fluorescent Pictograph” browser for exploring and an-alyzing large-scale biological data sets. PLoS One 2: e718

Xu Y, Xu Q, Huang B (2015) Ascorbic acid mitigation of water stress-inhibition of root growth in association with oxidative defense in tallfescue (Festuca arundinacea Schreb.). Front Plant Sci 6: 807

Young LM, Evans ML (1994) Calcium-dependent asymmetric movement of3H-indole-3-acetic acid across gravistimulated isolated root caps ofmaize. Plant Growth Regul 14: 235–242


Krieger et al.



Documents

Reactive Oxygen Species Tune Root Tropic Responses1[OPEN]...these tropic responses. We note that strong DHR ﬂuo-rescence is detected in the root vasculature above the CEZ at all