PDR-type ABC transporter mediates cellular uptake ofthe phytohormone abscisic acidJoohyun Kanga, Jae-Ung Hwanga, Miyoung Leea, Yu-Young Kima, Sarah M. Assmannb, Enrico Martinoiaa,c,1,and Youngsook Leea,d,1,2

aPOSTECH-UZH Global Research Laboratory, Division of Molecular Life Sciences, Pohang University of Science and Technology (POSTECH), Pohang 790-784,Korea; bBiology Department, Penn State University, University Park, PA 16802; cInstitut für Pflanzenbiologie, Universität Zürich (UZH), 8008 Zürich,Switzerland; and dDivision of Integrative Biology and Biotechnology, POSTECH, Pohang 790-784, Korea

Edited by Zhen-Ming Pei, Duke University, Durham, NC, and accepted by the Editorial Board December 18, 2009 (received for review August 14, 2009)

Abscisic acid (ABA) is a ubiquitous phytohormone involved in manydevelopmental processes and stress responses of plants. ABAmoveswithin the plant, and intracellular receptors for ABA have beenrecently identified; however, noABA transporter has been describedtodate.Here,wereport the identificationof theATP-bindingcassette(ABC) transporter Arabidopsis thaliana Pleiotropic drug resistancetransporter PDR12 (AtPDR12)/ABCG40 as a plasma membrane ABAuptake transporter. Uptake of ABA into yeast and BY2 cells express-ing AtABCG40 was increased, whereas ABA uptake into protoplastsof atabcg40 plants was decreased compared with control cells. Inresponse to exogenous ABA, the up-regulation of ABA responsivegenes was strongly delayed in atabcg40 plants, indicating thatABCG40 is necessary for timely responses to ABA. Stomata of loss-of-functionatabcg40mutants closedmore slowly in response toABA,resulting in reduced drought tolerance. Our results integrate ABA-dependent signaling and transport processes and open another ave-nue for the engineering of drought-tolerant plants.

abscisic acid transporter | drought resistance | guard cell

In both animals and plants, hormones play essential roles in theregulation of growth, development, and environmental response,

and they are circulated throughout the organism in part by theextracellular fluid. Two plant hormones known to be transportedover long distances, auxin and abscisic acid (ABA), are weak acids,and thus, they exist in either protonated, uncharged forms (e.g.,ABAH) or in anionic forms (e.g., ABA-) depending on the pre-vailing pH relative to their pKa. Because the uncharged forms ofthesemolecules can permeate the cell membrane, it was historicallyassumed that this diffusive process would obviate a requirement forspecific uptake transporters for these hormones (1). However,recent progress in the fields of auxin transport and signaling hasestablished that auxin is in fact transported into plant cells by mul-tiple influx carriers that are intricately regulated. Auxin carriers,including AUX1, LAX3, and several ATP-binding cassette (ABC)transporters (2), have been shown to be integral to auxin regulationof both developmental processes, such as lateral root emergence,and environmental responses, such as gravitropism (3, 4). However,comparable progress has not been made in the field of ABAtransport, despite the fact that analyses of the kinetics of ABAuptake suggest that such uptake does not occur solely by a diffusiveprocess (5–8).ABA regulates seed germination and seedling growth, and it is

required for plant resistance to drought and other abiotic and bioticstresses such as salinity and pathogen infection. During drought,ABA levels increase dramatically in plants (9). ABA is perceived byguard cells, which respond so as to minimize loss of water throughtranspiration. Each pair of guard cells in the epidermis delineates astomatal pore through which both carbon-dioxide uptake andtranspirational-water loss occur (10, 11). Stomata generally open inresponse to light, low CO2 concentrations, and high atmospherichumidity and close in response to darkness, high CO2 concen-trations, low humidity, and stress-induced ABA. Although secon-darymessengers and effectors of the ABA signal in guard cells have

been studied extensively (11), the initial steps of ABA perceptionwithin plant cells are just beginning to be understood.Recently, soluble intracellular receptors for ABA (12, 13) have

been identified. The intracellular localization of theseABA sensorshighlights the importance of ABA uptake into the cell for cellularsignaling processes to occur, and it suggests the potential impor-tance of an ABA transporter that could deliver ABA in a regulatedfashion to initiate rapid and controlled responses to the variousstress conditions that are perceived by ABA.We postulated that anABA transporter would be particularly relevant during stress con-ditions, because such conditions are known to elevate extracellularpH (8); this pH increase, in turn, causesABAH to dissociate into itscharged form,which cannot passively diffuse across the lipid bilayer,in contradiction to the need to rapidly deliver the stress hormoneinto the cell to elicit a timely response.TheABCprotein family is one of the largest, andmembers of this

family are found in all phyla (14). Most ABC proteins are integralmembrane proteins, and they act as ATP-driven transporters for avery wide range of substrates, including lipids, drugs, heavy metals,and auxin (15). In plants, several reports have shown that mutationof various ABC proteins results in impaired stomatal movement.Deletion of AtMRP5/ABCC5 results in impaired ABA and Ca2+

signaling and reduced anion-channel activity (16). AtABCB14 ishighly expressed in guard cells and functions as a malate importerthat modulates the stomatal response to CO2 (17). Here, we showthat an ABC transporter arabidopsis thaliana pleiotropic drugresistance transporter PDR12 (AtPDR12)/ABCG40 is a plasma-membrane ABA-uptake transporter in guard cells and other typesof plant cells. AtPDR12/ABCG40 is necessary for timely closure ofstomata in response to drought stress as well as for normal seedgerminationand lateral root development.Fromthis knowledge,weidentified a transporter that mediates the uptake of the phyto-hormone ABA into plant cells.

ResultsScreening of Potential ABA Transporters. ABA is a sesquiterpenederived from the tetraterpene neoxanthine. The PDR/ABCGsubfamily of plant ABC transporters has been reported to trans-port terpenoids (15, 18). PDR-type ABC transporters have alsobeen reported to be involved in responses to pathogens (19) and abroad range of stresses, including salinity, cold, and heavy metals

Author contributions: J.K., E.M., and Y.L. designed research; J.K. and J.-U.H. performedresearch; M.L. and Y.-Y.K. contributed new reagents/analytic tools; J.K., J.-U.H., E.M., andY.L. analyzed data; and J.K., S.M.A., E.M., and Y.L. wrote the paper.

Conflict of interest statement: Provisional application of USA (Abscisic Acid Transportersand Transformants Expressing the Genes; application date 2009-01-23 3; application num-ber 61/146705.

This article is a PNAS Direct Submission. Z.-M.P. is a guest editor invited by the EditorialBoard.1E.M. and Y.L. contributed equally to this work.2To whom correspondence should be addressed. E-mail: ylee@postech.ac.kr.

This article contains supporting information online at www.pnas.org/cgi/content/full/0909222107/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.0909222107 PNAS | February 2, 2010 | vol. 107 | no. 5 | 2355–2360

PLANTBIOLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 2

8, 2

021

(15, 20, 21). We, therefore, hypothesized that a member of thePDR family would function as anABA transporter. To identify themost promising candidate, we tested the seed germination andstomatal movements of 13 of 15 Arabidopsis PDR homozygousknockout mutants (atabcg29–atabcg41). In our screen, atabcg40exhibited the most pronounced differences from the wild type inseed germination and stomatal movement. Thus, we selectedAtABCG40 as a candidate and examined whether or not it indeedtransportsABAand if its function as anABA transporter is criticalfor plant-stress responses.

AtABCG40 Transports ABA. To assess whether or not AtABCG40 isan ABA transporter, we expressed the AtABCG40 cDNA in a het-erologous system, namely the YMM12 yeast strain, which carriesloss-of-function mutations in eight ABC transporters. Yeast-expressing AtABCG40 took up ABA consistently faster than con-trols containing the empty vector (Fig. 1A). Further evidence forAtABCG40 as an ABA transporter was obtained by expressing theAtABCG40 cDNA in cultured tobacco BY2 cells. ABA uptake wasclearly more efficient in cells expressing AtABCG40 (G1, G2, andG4 in Fig. 1B) than in either control cells or the G3 cell line thatexpresses AtABAG40 at a very low level (Fig. 1 B and C).AtABCG40 is a high-affinity ABA transporter that displays aKM of1 μM (Fig. 1D). The transporter also exhibits a high-substrate spe-cificity, because uptake of 3H-ABA was inhibited only by the phys-iologically active (S)-ABAandnotby (R)-ABA,ABA-glucose ester,indole-3-acetic acid, or benzoic acid at 3-fold excess (Fig. 1E).Inhibitors of ABC transporters such as glibenclamide, verapamil,and vanadate inhibited ABA transport (Fig. 1F).

If AtABCG40 is an ABA transporter, then not only should cellstransgenically expressing the transporter exhibit enhanced rates ofABA uptake, but cells in which the endogenousAtABCG40 gene isdisrupted should show decreased rates of ABA uptake. Accord-ingly, we assessed ABA uptake in mesophyll protoplasts isolatedfrom two independent T-DNA insertional mutants of AtABCG40,abcg40-1 and abcg40-2 (21). Indeed, ABAwas taken upmore slowlyinto protoplasts isolated from atabcg40 leaves compared with thosefromwild type (Fig. 2A), and in the complementedmutant lines, theuptake rate was restored to wild-type level (Fig. 2 B andC). In bothwild-type and atabcg40 plants, ABA uptake was higher at low pH,and it decreased with increasing pH (Fig. 2D), supporting a com-ponent of ABA uptake consisting of pH-dependent diffusion ofABAH.However, the rate ofABA uptake was consistently lower inatabcg40 cells compared with wild type, and the proportion of totalABA taken up by the knockoutmutants decreased in relation to thecorresponding amount taken up by wild type with increasing pH(Fig. 2D). This phenomenon suggests an increasing contribution ofAtABCG40 to total ABA uptake with increasing pH. It should,furthermore, bementioned that, in planta, the ratio of uptake by thetransporter versus uptake by diffusion would be higher than that inour experiments where the concentration of (S)-ABA available fortransport by AtABCG40 is only one-half of the total concentrationindicated. This is because of the high sterospecificity of transport(Fig.1E) versus the lackof sterospecificityofdiffusion. Interestingly,the differences in the absolute amounts of ABA taken up by wild-type and atabcg40 mutant cells, respectively, were independent ofpH (Fig. 2E). This pH-independent component of ABA uptakecannot be explained by diffusion, but it is in agreement with the

Fig. 1. AtABCG40 mediates specific uptake of ABA in aheterologous system. (A) Time-dependent uptake of 3H-ABA by YMM12 yeast cells expressing AtABCG40 (ABCG40)or transformed with the empty vector (EV). Yeast wasincubated in SG-URA medium containing 4.5 nM 3H-ABA(7.4 kBq, 1.63 Tba/mmol) at pH 7. Radioactivity taken up wasnormalized to the cell number. Data are mean ± SEM of n =12 from three independent experiments. (B) AtABCG40transcripts in independent BY2 cell lines expressingAtABCG40 (G1-G4) or EV. Actin was amplified as a loadingcontrol. (C) Time-dependent transport of 4.5 nM 3H-ABA(7.4 kBq, 1.63 Tba/mmol) by BY2 cells expressing AtABCG40(G1-G4) or EV at pH 5.7. (D) Concentration-dependentuptake of ABA containing 1 nM 3H-ABA (from 1 nM to 5 μMABA) or 2 nM 3H-ABA (from 10 to 15 μM ABA) in BY2 cells(line G2) for 18 min. Inset shows double reciprocal plotanalysis of the data indicating a KM of 1 μM. Mean is com-piled from three experiments. KMs calculated for the singleexperiments vary between 0.7 and 1.2 μM. Values werecorrected for the effective (S)-ABA concentrations. (E)Uptake of 1 μM (R, S)-ABA containing a trace amount (7.4kBq) of 3H-ABA in BY2 cells (line G2) in the absence (C) orpresence of an additional 3-fold (3 μM) unlabeled (R, S)-ABA, (R)-ABA, ABA-glucose ester (G), indole-3-acetic acid (I),or benzoic acid (B). (F) Inhibition of ABCG40-mediated 4.5nM 3H-ABA (7.4 kBq, 1.63 Tba/mmol) uptake in BY2 cells(line G2) by 3 mM vanadate (VD), 10 μM verapamil (VP), and25 μM glibenclamide (GC). C, control without inhibitorpretreatment. In D–F, ABA uptake mediated by AtABCG40was obtained by subtracting the activity of the empty vectorcontrol incubated under the same conditions from the totalradioactivity to exclude the portion of ABA transported bydiffusion.

2356 | www.pnas.org/cgi/doi/10.1073/pnas.0909222107 Kang et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 2

8, 2

021

notion of an active transport of ABA by AtABCG40. We proposethat AtABCG40 is the major ABA transporter in the leaf-cell pro-toplasts, because, at pH 7, the residual ABA-transport activity inatabcg40mutants isonlyabout 30%of that in thewild type (Fig. 2D).However, it cannot be excluded that close homologs of AtABCG40may also carry out ABA transport functions.

AtABCG40 Is Broadly Expressed and Plasma Membrane Localized. Toaddress the questions of whether or not AtABCG40 functions in abroad range of plant-cell types or if it is specifically associated with aparticular cell type, we assessedAtABCG40 expressionpatterns bothin planta and in silico. According to our promoter-β-glucuronidase(GUS)analysis, theAtABCG40promoter is broadly active, includingactivity in the leaves of young plantlets and in primary and lateralroots (Fig. 3A andB), which is in linewith publishedmicroarray data(www.genevestigator.ethz.ch). In leaves, the expression was by farthe highest in guard cells (Fig. 3C), consistent with microarray datathat reveal that AtABCG40 is expressed 8-fold more in guard cellsthan in mesophyll cells (www.bar.utoronto.ca).

If AtABCG40 functions as a carrier for the initial entry ofABA into plant cells, it would be expected to localize to theplasma membrane. AtABCG40 was previously shown to localizeto the plasma membrane of mesophyll protoplasts when tran-siently expressed under control of the 35S promoter (21). Asshown in Fig. 3 F and G, when AtABCG40-sGFP expression isdriven by the AtABCG40 native promoter in stably transformedplants, GFP fluorescence is also seen at the cell membrane,indicating that AtABCG40 operates at that locale.

atabcg40 Plants Are Strongly Delayed in Expression of ABA-ResponsiveGenes on Treatment with ABA. Given that there is a passive compo-nent of ABA uptake into cells, the question arises as to the impor-tance of AtABCG40 in ABA-regulated cellular functions.Accordingly, we investigated the expression of ABA responsivegenes in 4-week-old whole-rosette tissue of wild-type and atabcg40plants.Up-regulation of transcription factorsAtABR1 andAtRD29Band an ABA biosynthesis enzyme AtNCED3 in response to exoge-nous ABA application was considerably delayed and reduced in themutant plants compared with wild type (Fig. 4 A–C). Evidently,AtABCG40 is required for a fast response to ABA by allowing itsefficient uptake into the ABA-responsive cell. In contrast to thealteredkineticsof inductionofABA-responsivegenes, thekineticsofinduction of auxin-responsive genes of the Small Auxin-Up RNAs(SAUR) family did not differ in themutant from thewild-type plants(Fig. S1). Transcript levels ofAtABCG40 itself increased in responseto ABA treatment (Fig. 4D), indicating a positive feedback loop.

atabcg40 Plants Are Impaired in Stress Tolerance. Drought-stressexperiments provided further evidence that AtABCG40 is integralto stress tolerance. Plants were grown for 2 weeks under standardconditions, and water is subsequently withheld. Leaves of the twomutant lines wilted faster than those of the wild-type plants (Fig.5A). This result suggests that effective guard cell response toABA isimpaired in plants lacking expression of the AtABCG40 ABAtransporter, but this reaction could also be caused in part by alter-ations in ABA responsiveness of other physiological aspects. This isbecause AtABCG40 is widely expressed and is required for rapidtranscriptional responses in whole leaves. Additionally, atabcg40mutants also show impairment in ABA regulation of seed germi-nation and root development (Fig. S2). Because of the high levels ofAtABCG40 expression that we observed in guard cells and becauseof the known importance of ABA in stomatal regulation, we par-ticularly wanted to assess the impact of AtABCG40 knockout onguard-cell function. We pursued two approaches to address thisissue. Because transpiration lowers leaf temperature throughevaporative cooling, we employed thermal-imagingmethods (22) toassess transpirational water loss, which correlates with stomatalapertures. When hydroponically grown plants were treated with 1μM ABA or 0.5 M mannitol to induce stomatal closure and thus,reduce transpiration, leaf temperature increased more slowly inatabcg40 plants compared with wild-type plants (Fig. 5 B andC andFig. S3); this indicates that stomatal apertures were not as efficientlyreduced in response to ABA or osmotic stress as in the wild type.To confirm the results from thermal-imaging analysis implicat-

ing ABCG40 in guard-cell function, we also directly examined theeffect of ABA on stomatal movements in wild-type and atabcg40mutant plants. Wild-type and atabcg40 plants indeed differed intheir responses to ABA during light-induced stomatal opening.Stomata of the two mutant lines atabcg40-1 and atabcg40-2 plantsopened faster, attaining apertures of 2.64 ± 0.07 μm and 2.57 ±0.05 μm, respectively, whereas wild-type plants attained 2.25 ±0.06 μm under the same conditions (Fig. 5D). Furthermore, incomparison with the wild type, the two atabcg40 lines exhibiteddelayed and reduced stomatal closure in response to ABA (Fig.5E). Complementation of the atabcg40-1 mutant line with theAtABCG40 cDNA driven by its own promoter confirmed that theobserved mutant phenotype was caused by AtABCG40 loss of

Fig. 2. AtABCG40 mediates specific uptake of ABA in Arabidopsis. (A) Theisolatedmesophyll cells were incubated in a bathing solution containing 4.5 nM3H-ABA at pH 5.7. The chlorophyll content of the protoplasts was used to nor-malize 3H-ABA uptake. A representative of four independent experiments withsimilar results is shown (mean ± SEM; n = 4). (B) Presence of the AtABCG40transcript in atabcg40-1 mutant plants complemented with AtABCG40pro::AtABCG40::sGFP. Ubiquitin was amplified as a loading control. Total RNA wasisolated from shoots of wild-type atabcg40-1 mutant (g40-1) and AtABCG40complemented atabcg40-1 mutant plants (C1, 2, 3, and 4 ), respectively. (C)Uptake of 3H-ABA into mesophyll protoplasts of wild-type and AtABCG40-complemented atabcg40-1 lines C1 and C3 are shown as a function of time. Thechlorophyll content of the protoplasts was used to normalize the 3H-ABAuptake.Dataaremean±SEMofn=12fromthree independentexperiments. (D)pH-dependentuptakeof 3H-ABA intomesophyll protoplasts fromwild-typeandatabcg40 lines. The chlorophyll contentof theprotoplastswasused tonormalize3H-ABA values. Radioactivity of samples incubated for 2 min was subtractedfrom thatof samples incubated for 3 h.Dataaremean± SEMofn=16 from fourindependent experiments. (E) To obtain AtABCG40-mediated ABA transport,3H-ABA taken upby abcg40-1or abcg40-2protoplasts was subtracted from thatofwild-typeprotoplasts. Thedata are from the sameexperiments as shown inD.

Kang et al. PNAS | February 2, 2010 | vol. 107 | no. 5 | 2357

PLANTBIOLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 2

8, 2

021

function (Fig. 5F). In contrast to ABA, no difference was observedfor stomatal closure in response to Ca2+ (Fig. S4), suggesting thatthe mutation affected the level of ABA or its perception ratherthan the downstream ABA signaling. The results obtained usingepidermal strips and thermal imaging of intact plants collectivelyshow that the guard cells of atabcg40 plants exhibit a stronglydelayed response toABA.Wepropose thatAtABCG40 is anABAimporter required for efficient response to ABA.

DiscussionRapid adjustment to a stress, such as drought stress, is a pre-requisite for plant survival. Despite the diffusion of ABAHthrough the lipid bilayer, our data indicate that a transporter isrequired for optimal ABA uptake during stress. Neither ABAuptake nor the response to ABA is totally abolished whenAtABCG40 is nonfunctional, indicating that a contribution ofdiffusion to ABA uptake at an apoplastic pH of 5.5–6.0, typical ofthe nonstressed plant, does occur.However, for rapid and efficientsignaling under stress conditions, AtABCG40 activity is required.As mentioned above, the pH of the xylem and apoplast increasesduring drought stress (8), and the advantage is that uncontrolleddiffusion of ABAH into nontarget cells and hence, loss of ABA onits way to its target cells (i.e., the guard cells) is avoided. Passivediffusion of ABA into the target cells would, however, be reducedas well, and thus, a transporter-mediated uptake process is nec-essary to overcome this disadvantage (Fig. S5).After it is inside thecell, ABA binds to ABA receptors localized in the cytosol (12, 13)or at the plasma membrane (23), thereby initiating ABA signaltransduction. The very recent discovery of a class of soluble ABAreceptors is in perfect agreement with our results (12, 13). How-ever, ABA receptors localized at the plasma membrane (23) mayalso recognize ABA by domains exposed to the cytosol. An ABAtransporter would allow fast and controlled delivery ofABA to thereceptor and thus, ensure a rapid response to the stress hormone,which is not achievable by diffusive influx of ABA alone. Theprevious observation that stomatal apertures andactivity of inwardK+ channels that mediate K+ uptake during stomatal opening areboth responsive to intracellular application of ABA is also con-sistent with the importance of an ABA uptake carrier (24). The

observation that ABA-responsive transcription factors, presentnot only in guard cells but in many other cell types, respond muchfaster in the presence of AtABCG40 (Fig. 4) indicates thattransporter-catalyzed ABA transport is important for ABA sig-naling in many cell types. Supporting this, lateral root formationwas delayed in atabcg40 as well (Fig. S2C). Decreased sensitivity ofatabcg40 seeds to exogenously addedABA(Fig. S2B) suggests thatthis transporter is active in seeds as well, although it is expressed ata very low level in this organ (www.bar.utoronto.ca). An intriguingpossibility is that the seeds may take up ABA through AtABCG40during development and store the hormone to prevent precociousgermination. Supporting this possibility, atabcg40 seeds exhibitedslightly accelerated germination on normal growth medium with-out added ABA (Fig. S2A).We previously reported that the atabcg40 plants are compromised

in lead resistance (21). Because ABC transporters are well-known astransporters of structurally unrelated compounds (15), we testedwhether or not AtABCG40 transports both lead and ABA using acompetition assay; however, lead did not compete with ABA foruptake intoplantcells (Fig.S6).Thus,wesuggest that thereduced leadtolerance of the mutant is likely caused by compromised ABAtransport, which indirectly affects heavy metal tolerance of the plant.Several reports postulate a role ofABA in heavymetal tolerance. Forexample, ABA content in leaves increases when plants are exposed tothe heavy metals lead, cadmium, or aluminum (25–27). ABA regu-lates the expression level of Type 4metallothionein (28), amember ofan important metal chelator family. Application of exogenous ABAreduced the root-to-shoot translocation of Cd (29) by inducing sto-matal closure. In short, ABA, which is increased in heavy metal-treated plants, plays an important role in plant tolerance to heavymetals through at least two different pathways. First, it promotesstomatal closure,which reducesheavymetal translocation to the shootthrough the transpirational stream. Second, it modifies expression ofgenes that can contribute to heavy metal tolerance (28–30).ABCG40hasbeen suggested to transport diterpenoids associated

with plant defense based on three findings (18): (i) it is up-regulated

A B

C

E

D

G

F

Fig. 3. Tissue-specific expression and subcellular localization of AtABCG40.(A–C) AtABCG40 promoter-GUS (AtABCG40pro::uidA) reporter gene expres-sion in 2-week-old plants (A) and in primary and lateral roots (B). C shows aclose-up image of a leaf showing expression in guard cells. (D–G) Plasma-membrane localization of ABCG40::sGFP expression that is driven by thenative promoter inArabidopsisguard cells (F andG).Wild-type cells are shownin D and E. Images show the same two cells under bright field illumination (Dand F) and with GFP fluorescence (E and G).

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

0 15 30 45 60

Time [min]

AtA

BC

G40

(fo

ld in

du

ctio

n)

WTD

WTatabcg40-1

0

1

2

3

4

5

6

7

0 15 30 45 60

Time [min]

AtR

D29

B(f

old

ind

uct

ion

)

WTatabcg40-1

A

C

0

1

2

3

4

5

6

7

8

0 15 30 45 60

Time [min]

AtA

BR

1(f

old

ind

uct

ion

)

WTatabcg40-1

B

0

2

4

6

8

10

12

14

0 15 30 45 60

Time [min]

AtN

CE

D3

(fo

ld in

du

ctio

n)

Fig. 4. AtABCG40 modulates the expression of ABA-responsive genes. (A–D) Q-PCR analyses of transcripts of ABA-responsive genes in total RNA iso-lated from shoots of wild-type and atabcg40-1 plants, respectively, that aretreated with ABA. Data were normalized using Tubulin 8 and are presentedas “fold induction” relative to time = 0. Data are mean ± SEM from fourindependent experiments. For AtABR1 and AtRD29B, the initial values weresimilar in the wild type (WT) and atabcg40-1, whereas for AtNCED3, theinitial value in atabcg40-1 was 2.4 -fold higher than that of the WT.

2358 | www.pnas.org/cgi/doi/10.1073/pnas.0909222107 Kang et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 2

8, 2

021

by pathogen infection, (ii) its amino acid sequence is similar to thoseof other diterpenoid transporters such as NpABC1 and SpTUR2,and (iii) the atabcg40 mutant plant is altered in tolerance to thetoxicity of diterpenoid sclareol.ABA is a sesquiterpenoid, and thereis increasing evidence that ABA plays an important role in plant-pathogen responses (31–33). Therefore, either ABCG40 transportsvarious terpenoids or many of the previous findings on pathogen-related phenomena are indirect effects of the ABA-transportfunction of ABCG40. We favor the latter possibility, becauseABCG40 showed very narrow substrate specificity (Fig. 1E). In linewith this possibility, we couldnot observe anydifference ingrowthofthe atabcg40 mutants versus the wild type in medium containingsclareol (21). Moreover, although sclareol is structurally similar toABA, it is not likely to be a physiological substrate of ABCG40,because this compound is not produced in Arabidopsis. Finally, lowmicromolar concentrations of sclareol did not inhibit (S)-ABAuptake into AtABCG40-expressing BY2 cells, indicating that atleast the uptake activity ofAtABCG40 is specific forABA (Fig. S6).In one publication, the authors suggested that AtABCG40 canexport sclareol (18); however, they did not perform transport

studies.Whether or not AtABCG40 can transport in two directionsmust be addressed in future work.In conclusion, wehave presented evidence forABC transporter-

mediated ABA uptake and its importance for rapid responses toenvironmental stress. This work underlines the importance oftransporters for hormones that also diffuse into the cell across thelipid bilayer. Our results may promote the production of plantswith increased drought tolerance (e.g., through guard cell-specificexpression of AtABCG40), which may lead to faster stomatalclosure. Manipulation of a very specific pathway for uptake ofABA, a step localized at the apex of the ABA-signaling pathway,may allow development of plants permanently primed to respondquickly to stress. Transporters related toAtABCG40 alsomay playan important role in other organisms, because ABA has recentlybeen reported to induceCa2+ release duringToxoplasma infectionand in human granulocytes (34, 35).

Materials and MethodsPlant Material and Growth Conditions. Wild-type (ecotype Columbia-0) andAtABCG40 transgenic plants (21) were grown on soil (22/18 °C; 16/8 h day/night; 40 μmol m−2s−1 light).

1.0

2.0

3.0

4.0

5.0

6.0

0 30 60 90Time [min]

WTabcg40-1

C2

C4

C1

C3

FED

Stom

atal

aper

ture

[µm

]

0.8

1.2

1.6

2.0

2.4

2.8

0 1 2 3Time [h]

WTabcg40-1abcg40-2

1.0

2.0

3.0

4.0

5.0

6.0

0 30 60 90Time [min]

WTabcg40-1abcg40-2

B

4 34 64 94 [min]

C

WT

abcg40-1

[ C]

0.0

0.2

0.4

0.6

0.8

1.0

1.2

34 64 94 124Time [min]

Tem

pera

ture

[C

]

WTabcg40-1

*

** **

Control Drought

WT WTabcg40-1 abcg40-2 WT abcg40-2WT abcg40-1A

Control Drought

20

16

Fig. 5. Stomataofatabcg40plantsareless sensitivetoABA. (A)Two-week-oldsoil-grownplants (24°C;16-h lightand8-hdarkconditions)wereexposedtodroughtstressbywithholdingwater for8days. (B)Delayedelevationof leaf temperatureafterABAtreatmentofatabcg40plants comparedwithwild type. Leaf temperaturewasmonitoredusing an Infrared Thermal Imaging Camera (FLIR systems; P25) after the addition of ABA into the hydroponic culture medium to a final concentration of 1 μM. Resultsrepresentative of three experiments with similar results are shown. (C) Increase in leaf temperature (Δ Temperature) after ABA treatment of plants is shown. The leaftemperaturewasquantifiedfromthree independentexperiments, includingtheoneshowninB.Datarepresentchangeintemperaturesofthewhole-leafareaoftheplants.ΔTemperature = (Temperature at indicated time point)− (Temperature at 4min). The initial temperatureswere similar inWT and atabcg40-1.Data aremean± SEM (*P<0.05; ** P< 0.01 comparedwithWT under the same treatment conditions by Student's t test). (D) Opening of stomata in the presence of 1 μMABA. Epidermal strips werepeeled from leaves thathadbeenfloatedon10mMKClbuffer in thepresenceof1μMABAunder170 μmolm−2sec−1white light. (Eand F)ABA-induced stomatal closure inWT and atabcg40 plants (E) and in four independent lines complemented withAtABCG40 (F). Data aremean ± SEM of n = 83∼108 (D), n = 80∼112 (E), and n = 75∼95 (F)stomata (obtained from three independent experiments).

Kang et al. PNAS | February 2, 2010 | vol. 107 | no. 5 | 2359

PLANTBIOLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 2

8, 2

021

DNA Constructs. The AtABCG40 promoter::uidA reporter gene construct wasmade by PCR amplification of a 1.8-kb promoter region of AtABCG40 fromgenomic DNA using primers containing HindIII and BamHI restriction sites (5′-AAGCTTACGCCGGCCGCCGCCGCGGCAG-3′ and 5′-GGATCCTTTGTATCCAA-GAAATCAAAGT-3′) and ligation of the PCR product into pBI101.2. The ORF ofAtABCG40 was amplified by PCR from cDNA generated from total RNAextracted from wild-type seedlings using primers containing BamHI restrictionsites (5′-CCCGGGGGGGATCCATGGAGGGAACTAGTTTTCACCAAGCGAGTA-3′and 5′-GGATCCGCGGCCGCCTATCGTTTTTGGAAATTGAAACTCTTGATTC-3′). Togenerate complemented and tagged lines of atabcg40-1 mutants, theAtABCG40promoterwas inserted intopBI101.2 atHindIII andBamHI restrictionsites; sGFP was amplified from the 326-sGFP vector and fused to the 3′-end oftheAtABCG40promoter. Finally, the full-length cDNAofAtABCG40was ligatedinto the BamHI site of the construct. All constructswere verified by sequencing.

Verification of AtABCG40-Complemented Arabidopsis Plants. The abundanceofAtABCG40 transcript in the homozygous complemented lines C1, C2, C3, and C4andwild-type plantswas assayed by RT-PCR. Total RNAwas extracted fromwholeseedlings, and RT-PCR was performed after DNaseI (Roche) treatment using pri-mers specific for AtABCG40 (5′-CTGCTTTTGGGTCCTCCAAGTTCT-3′, and 5′-GAGATTGAATGTCTCTGGCGCAG-3′). As a loading control, the Tubulin8 transcriptwas amplified using the primers (5′-CTCACAGTCCCGGAGCTGACAC-3′ and 5′-GCTTC AGTGAACTCCATCTCGT-3′).

Assays of Stomatal Movements. Stomatal apertures were determined in 5-week-old leaves. For assays of stomatal opening, detached whole leaves werefloated on 10 mM KCl buffer, with or without 1 μM ABA, under 170 μmolm−2sec−1 white light at 23 °C. For assays of stomatal closure, leaves werepreincubated on 10 mM KCl buffer for 3 h under 170 μmol m−2sec−1 whitelight at 25 °C to open stomata, and then, they were transferred to 10 mM KClbuffer containing 1 μMABA (time = 0). Stomatal apertures were measured asdescribed previously (36).

Infrared Thermography. Arabidopsis plants were grown under hydroponicconditions (22 °C; 8 h of light photoperiod; 40 μmol m−2s−1 light) for 6 weeks(37). After adding ABA or mannitol to the hydroponic medium to a finalconcentration of 1 μM or 0.5 M, respectively, the leaf temperature of intactArabidopsis plants was measured as described previously (22, 36).

3H-ABA Uptake into Arabidopsis Protoplasts. Transportactivitywasdeterminedas described previously (38). Protoplasts were incubated in 1 mL of a bathingsolution containing 4.5 nM (R, S)-3H-ABA (Amersham Biosciences; 7.4 kBq and1.63 TBq/mmol).

Heterologous Expression of AtABCG40 in Yeast and Transport Assays. The yeastYMM12strainwasakindgift fromKarlKuchler (Vienna,Austria).AtABCG40wascloned into the BamHI and NotI sites of pYES2NT/C. 3H-ABA uptake was moni-tored as described previously withminormodifications (17). Cells were culturedin minimum salt-galactose minus uracil (SG-URA) medium, supplemented with0.5%raffinose,atpH7.0,andharvestedbycentrifugationatmidlogphase. Theywere washed two times using SG-URA medium, and they were resuspended inthe samemediumat anOD600 = 6. 3H-ABA (4.5 nM, 7.4 kBq, 1.63 TBq/mmol)wasthen added to the cell suspension and gently mixed. At the times indicated, thecell suspension was filtered through nitrocellulose membranes, and the cellsremaining on the filter were washed with 500 μL of ice-cold SG-URA medium.The radioactivity on the filter was determined by liquid-scintillation counting.

Transport Assay Using BY2 Cells. See SI Materials and Methods.

Quantitative Real-Time RT-PCR. See SI Materials and Methods.

ACKNOWLEDGMENTS. We thank N. Amrhein and Shaun Peters for readingthe manuscript. This work was supported by the Global Research Laboratoryprogram of the Ministry of Science and Technology of Korea and the CropFunctional Genomics Center of KoreaGrant CG1-1-23 (to Y.L.). Thework in thelaboratory of E.M. was partially supported by the Swiss National Foundation.S.M.A. was partially supported by the U.S. National Science Foundation.

1. Salisbury FB, Ross CW (1985) Plant Physiology (Wadsworth Publishing Company,Belmont, California) 3rd Ed, pp 312.

2. Vieten A, Sauer M, Brewer PB, Friml J (2007) Molecular and cellular aspects of auxin-transport-mediated development. Trends Plant Sci 12:160–168.

3. Marchant A, et al. (1999) AUX1 regulates root gravitropism in Arabidopsis byfacilitating auxin uptake within root apical tissues. EMBO J 18:2066–2073.

4. Swarup K, et al. (2008) The auxin influx carrier LAX3 promotes lateral root emergence.Nat Cell Biol 10:946–954.

5. Jiang F, Hartung W (2008) Long-distance signaling of abscisic acid (ABA): The factorsregulating the intensity of the ABA signal. J Exp Bot 59:37–43.

6. Daeter W, Hartung W (1993) The permeability of the epidermal cell plasmamembrane of barley leaves to abscisic acid. Planta 191:41–47.

7. Windsor ML, Milborrow BV, McFarlane IJ (1992) The uptake of (+)-S- and (-)-R-abscisicacidby suspensionculture cells ofHopbush (Dodonaeaviscosa).PlantPhysiol100:54–62.

8. Wilkinson S, Davies WJ (1997) Xylem sap pH Increase: A drought signal received at theapoplastic face of the guard cell that involves the suppression of saturable abscisicacid uptake by the epidermal symplast. Plant Physiol 113:559–573.

9. Zeevaart JAD (1980) Changes in the levels of abscisic acid and its metabolites inexcised leaf blades of Xanthium strumarium during and after water stress. PlantPhysiol 66:672–678.

10. Hetherington AM (2001) Guard cell signaling. Cell 107:711–714.11. Israelsson M, et al. (2006) Guard cell ABA and CO2 signaling network updates and Ca2+

sensor priming hypothesis. Curr Opin Plant Biol 9:654–663.12. Park SY, et al. (2009) Abscisic acid inhibits type 2C protein phosphatases via the PYR/

PYL family of START proteins. Science 324:1068–1071.13. Ma Y, et al. (2009) Regulators of PP2C phosphatase activity function as abscisic acid

sensors. Science 324:1064–1068.14. Verrier PJ, et al. (2008) Plant ABC proteins—a unified nomenclature and updated

inventory. Trends Plant Sci 13:151–159.15. Rea PA (2007) Plant ATP-binding cassette transporters. Annu Rev Plant Biol 58:

347–375.16. Suh SJ, et al. (2007) TheATP binding cassette transporter AtMRP5modulates anion and

calcium channel activities in Arabidopsis guard cells. J Biol Chem 282:1916–1924.17. Lee M, et al. (2008) The ABC transporter AtABCB14 is a malate importer and

modulates stomatal response to CO2. Nat Cell Biol 10:1217–1223.18. Campbell EJ, et al. (2003) Pathogen-responsive expression of a putative ATP-binding

cassette transporter gene conferring resistance to the diterpenoid sclareol is regulatedby multiple defense signaling pathways in Arabidopsis. Plant Physiol 133:1272–1284.

19. Stein M, et al. (2006) Arabidopsis PEN3/PDR8, an ATP binding cassette transporter,contributes to nonhost resistance to inappropriate pathogens that enter by directpenetration. Plant Cell 18:731–746.

20. Moons A (2008) Transcriptional profiling of the PDR gene family in rice roots inresponse to plant growth regulators, redox perturbations and weak organic acidstresses. Planta 229:53–71.

21. Lee M, et al. (2005) AtPDR12 contributes to lead resistance in Arabidopsis.. Plant

Physiol 138:827–836.22. Merlot S, et al. (2002) Use of infrared thermal imaging to isolate Arabidopsis mutants

defective in stomatal regulation. Plant J 30:601–609.23. Pandey S, Nelson DC, Assmann SM (2009) Two novel GPCR-type G proteins are abscisic

acid receptors in Arabidopsis. Cell 136:136–148.24. Schwartz A, Wu WH, Tucker EB, Assmann SM (1994) Inhibition of inward K+ channels

and stomatal response by abscisic acid: An intracellular locus of phytohormone action.

Proc Natl Acad Sci USA 91:4019–4023.25. Barcelo J, Cabot C, Poschenrieder CH (1986) Cadmium-induced decrease of water

stress resistance in bush bean plants (Phaseolus vulgaris L. cv. Contender). II. Effects of

Cd on endogenous abscisic acid levels. J Plant Physiol 125:27–34.26. Atici O, Agar G, Battal P (2005) Changes in phytohormone contents in chickpea seeds

germinating under lead or zinc stress. Biol Plant 49:215–222.27. Kasai M, et al. (1993) Possible involvement of abscisic acid in increases in activities of two

vacuolarH+-pumps inbarley roots under aluminumstress.PlantCell Physiol34:1335–1338.28. Chatthai M, et al. (1997) The isolation of a novel metallothionein-related cDNA

expressed in somatic and zygotic embryos of Douglas-fir: Regulation by ABA,

osmoticum, and metal ions. Plant Mol Biol 34:243–254.29. Perfus-Barbeoch L, Leonhardt N, Vavasseur A, Forestier C (2002) Heavy metal toxicity:

Cadmium permeates through calcium channels and disturbs the plant water status.

Plant J 32:539–548.30. Talanova VV, Titov AF, Boeva NP (2000) Effect of increasing concentrations of lead

and cadmium on cucumber seedlings. Biol Plant 43:441–444.31. Zhao Z, ZhangW, Stanley BA, Assmann SM (2008) Functional proteomics ofArabidopsis

thalianaguardcells uncovers newstomatal signalingpathways.PlantCell20:3210–3226.32. Ton J, Flors V, Mauch-Mani B (2009) The multifaceted role of ABA in disease

resistance. Trends Plant Sci 14:310–317.33. Zhang W, He SY, Assmann SM (2008) The plant innate immunity response in stomatal

guard cells invokes G-protein-dependent ion channel regulation. Plant J 56:984–996.34. Nagamune K, et al. (2008) Abscisic acid controls calcium-dependent egress and

development in Toxoplasma gondii. Nature 451:207–210.35. Bruzzone S, et al. (2007) Abscisic acid is an endogenous cytokine in humangranulocytes

with cyclic ADP-ribose as second messenger. Proc Natl Acad Sci USA 104:5759–5764.36. Jeon BW, et al. (2008) The Arabidopsis small G protein ROP2 is activated by light in

guard cells and inhibits light-induced stomatal opening. Plant Cell 20:75–87.37. Chen A, Komives EA, Schroeder JI (2006) An improved grafting technique for mature

Arabidopsis plants demonstrates long-distance shoot-to-root transport of phytochelatins

in Arabidopsis. Plant Physiol 141:108–120.38. Kim D-Y, et al. (2007) The ABC transporter AtPDR8 is a cadmium extrusion pump

conferring heavy metal resistance. Plant J 50:207–218.

2360 | www.pnas.org/cgi/doi/10.1073/pnas.0909222107 Kang et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 2

8, 2

021