LLM-Domain B-GATA Transcription Factors PromoteStomatal Development Downstream of Light SignalingPathways in Arabidopsis thaliana Hypocotyls

Carina Klermund, Quirin L. Ranftl, Julia Diener, Emmanouil Bastakis, René Richter,1 and Claus Schwechheimer2

Plant Systems Biology, Technische Universität München, 85354 Freising, Germany

ORCID IDs: 0000-0002-4438-606X (C.K.); 0000-0003-0269-2330 (C.S.)

Stomata are pores that regulate the gas and water exchange between the environment and aboveground plant tissues,including hypocotyls, leaves, and stems. Here, we show that mutants of Arabidopsis thaliana LLM-domain B-GATA genes aredefective in stomata formation in hypocotyls. Conversely, stomata formation is strongly promoted by overexpression ofvarious LLM-domain B-class GATA genes, most strikingly in hypocotyls but also in cotyledons. Genetic analyses indicate thatthese B-GATAs act upstream of the stomata formation regulators SPEECHLESS (SPCH), MUTE, and SCREAM/SCREAM2and downstream or independent of the patterning regulators TOO MANY MOUTHS and STOMATAL DENSITY ANDDISTRIBUTION1. The effects of the GATAs on stomata formation are light dependent but can be induced in dark-grownseedlings by red, far-red, or blue light treatments. PHYTOCHROME INTERACTING FACTOR (PIF) mutants form stomata in thedark, and in this genetic background, GATA expression is sufficient to induce stomata formation in the dark. Since theexpression of the LLM-domain B-GATAs GNC (GATA, NITRATE-INDUCIBLE, CARBON METABOLISM-INVOLVED) and GNC-LIKE/CYTOKININ-RESPONSIVE GATA FACTOR1 as well as that of SPCH is red light induced but the induction of SPCH iscompromised in a GATA gene mutant background, we hypothesize that PIF- and light-regulated stomata formation inhypocotyls is critically dependent on LLM-domain B-GATA genes.

INTRODUCTION

Plants use stomata to control the exchange of oxygen, carbondioxide, and water with their environment (Casson and Hether-ington, 2010). Therefore, stomata are important for the efficiencyof photosynthesis in plants and their activity contributes to theatmospheric environment (Dow and Bergmann, 2014). Stomatalpores are surrounded by guard cells in the epidermis of above-ground plant organs such as hypocotyls, cotyledons, stems,leaves, and floral organs, but they are absent from roots and onlyrarely found in the hypocotyls of dark-grown seedlings (Pillitteriand Torii, 2012; Wengier and Bergmann, 2012).

For stomatal development, protodermal cells enter the stomatallineage in response to a number of input signals that include lightand phytohormones (Casson et al., 2009; Casson and Hether-ington, 2012, 2014; Kim et al., 2012; Wang et al., 2015). Stomataldevelopment and patterning are initiated by signaling events thatemploy the plasma membrane-resident receptor-like proteinsERECTA (ER), ER-LIKE1 (ERL1), and ERL2 and the coreceptorTOO MANY MOUTHS (TMM), as well as the peptide ligandsEPIDERMALPATTERNINGFACTOR1 (EPF1), EPF2, and theEPF-LIKE proteins (EPFLs), including EPFL9/STOMAGEN, EPFL6/CHALLAH (CHL), EPFL5/CHALLAH-LIKE1 (CLL1), and CLL2

(EPFL4/CLL2) (Yang and Sack, 1995; Shpak et al., 2005; Haraet al., 2007, 2009; Hunt and Gray, 2009; Sugano et al., 2010;Abrash et al., 2011; Lee et al., 2015). These peptide ligands act inpart as activators and in part as repressors of stomatal de-velopment and, ashasbeen recently demonstrated in studieswithEPFL9/STOMAGEN and EPF2, promote and repress stomataformation through antagonistic interactions with the ER-TMMreceptor pair (Lee et al., 2015). Interestingly, mutants of differentmembers of the stomatal patterning genes may have differentialdefects in different tissues. This is most prominent in the tmmmutant, which forms more stomata than the wild type in cotyle-dons but lacks stomata in the hypocotyl and the stem (Yang andSack, 1995; Nadeau and Sack, 2002).Stomatal differentiation requires the consecutive activities of

the basic helix-loop-helix (bHLH) transcription factors SPEECH-LESS (SPCH), MUTE, and FAMA, which all interact with thefunctionally redundant bHLHs SCREAM/ICE1 and SCREAM2(Ohashi-Ito and Bergmann, 2006; MacAlister et al., 2007; Pillitteriet al., 2007;Kanaokaetal., 2008).SPCH is selectivelyexpressed incells recruited to the stomatal lineage where it promotes the firstasymmetric division in stomatal development that gives rise toa meristemoid. This meristemoid undergoes further amplifyingasymmetric divisions, resulting in the formation of an oval-shapedguard mother cell. FAMA then promotes the differentiation of thisoval-shaped cell into the guard cell pair of the fully differentiatedstomatal pore (MacAlister et al., 2007). Following the initial SPCH-dependent divisions, a number of sister cells are produced thatmay also enter the stomatal lineage. It has been suggested thatSPCH participates in the amplifying divisions of meristemoids(MacAlister et al., 2007) and, on the other side, that MUTE isrequired for the termination of the transient stem cell-like

1Current address: Max Planck Institute for Plant Breeding Research,Carl-von-Linné-Weg 10, 50829 Cologne, Germany.2 Address correspondence to claus.schwechheimer@wzw.tum.de.The author responsible for distribution of materials integral to the findingspresented in this article in accordance with the policy described in theInstructions for Authors (www.plantcell.org) is: Claus Schwechheimer(claus.schwechheimer@wzw.tum.de).www.plantcell.org/cgi/doi/10.1105/tpc.15.00783

The Plant Cell, Vol. 28: 646–660, March 2016, www.plantcell.org ã 2016 American Society of Plant Biologists. All rights reserved.

asymmetric divisions (Pillitteri et al., 2007). Furthermore, FAMA isrequired to prevent further symmetric divisions of the guardmother cell (Ohashi-Ito and Bergmann, 2006). The proposed roleof SPCH as a key regulator of stomatal development recentlyfoundstrongsupport after the identificationofSPCHtargetgenes,which includedmanyknown regulators of stomatal patterning anddifferentiation (Lau et al., 2014).

Stomata formation is light regulated. Whereas dark-grownseedlings only occasionally form stomata in hypocotyls andcotyledons, stomata formation in these embryonic tissues isstrongly promoted by light (Casson et al., 2009; Casson andHetherington, 2014). This process is at least in part dependent onthe activities of the red light photoreceptor phytochrome B (phyB)and PHYTOCHROME INTERACTING FACTOR (PIF) transcriptionfactors, which act downstreamof phyB (Casson et al., 2009). PIFsare stabilized in etiolated seedlings in the dark, but they are in-activated through proteasomal degradation following phyB ac-tivation in the light (Martínez-Garciaet al., 2000;Leivaret al., 2008).How PIFs promote stomata formation is presently unclear.

GATA factors are evolutionarily conserved transcription factorsthat recognize DNA sequences with the motif W-G-A-T-A-R(Reyes et al., 2004). In Arabidopsis thaliana, GATAs of the B-subfamily (B-GATAs) are defined by their highly conserved DNAbinding domain and the presence of either an LLM- or a HAN-domain (Behringer and Schwechheimer, 2015). HAN-domain B-GATAs were first described in the Arabidopsis floral developmentregulators HAN (HANABA TARANU) and GATA19 (HAN-LIKE)(Zhaoet al., 2004;Zhanget al., 2013).Bycontrast, LLM-domainB-GATAs contain the strictly conserved leucine-leucine-methionine(LLM) sequence, and they were initially defined following studieswith the paralogous GNC (GATA, NITRATE-INDUCIBLE, CAR-BON METABOLISM-INVOLVED) and GNL (GNC-LIKE/CYTOKI-NIN-RESPONSIVE GATA FACTOR1) (Richter et al., 2010;Behringer and Schwechheimer, 2015).

Besides GNC and GNL, the Arabidopsis LLM-domain B-GATAsubfamily also includes GATA15, GATA16, GATA17, andGATA17L, and these six GATAs have redundant functions in theregulation of multiple developmental processes (Behringer et al.,2014; Behringer and Schwechheimer, 2015; Ranftl et al., 2016).GATA23, a protein described to be important for lateral root ini-tiation in Arabidopsis, is related to LLM-domain B-GATAs (DeRybel et al., 2010). However, this seemingly Brassicaceae-spe-cificGATA has a degenerate LLM-domain and can be functionallydistinguished from the other family members (Behringer et al.,2014;Behringer andSchwechheimer, 2015). The functional role ofthe LLM-domain is at present unclear, but it was shown thatvariantswith LLM-domainmutations are impaired in their ability tocompletely suppress LLM-domain B-GATA mutant phenotypes(Behringer et al., 2014).

Here, we reveal that the LLM-domain B-GATA factors are re-quired and sufficient to promote cell divisions and stomata for-mation. The effects of the B-GATAs are most prominent in thehypocotyls of light-grown seedlings, where these B-GATAs actupstream of the stomatal regulators SPCH, MUTE, and SCRM/SCRM2 and their activities remain specific for the stomata-producing nonprotruding cell file (Berger et al., 1998;Pillitteri et al.,2008). Furthermore, the effects of the B-GATAs are also strictlylight dependent. Our genetic and physiological studies indicate

that LLM-domain B-GATAs act downstream of PIF transcriptionfactors and in concert with other PIF-dependent factors in theregulation of light-dependent stomata formation.

RESULTS

LLM-Domain B-GATAs Promote Stomata Formation

Wepreviously observed that light-grownArabidopsis seedlingsofLLM-domain B-GATA gene overexpressors (GNCox, GNLox,GATA15ox, andGATA17ox) had longer hypocotyls thanwild-typeseedlingsandaccumulatedmorechlorophyll,mostprominently atthe base of the hypocotyl (Figure 1A; Behringer et al., 2014). Inthese overexpression lines, the effects on hypocotyl elongationand, at least in part also on greening, were dependent on thepresence and absence of the LLM-domain and could not beobserved in overexpression lines of the LLM-domain mutatedGNC (GNC_LLM/AAAox or GNCDCTox) of GATA23, which har-bors a degenerate LLM-domain (GATA23ox), or of the HAN-domain B-GATA GATA19 (GATA19ox) (Behringer et al., 2014).To examine the hypocotyl phenotype of these GATA over-

expressors more closely, we analyzed seedling hypocotyls byscanningelectronmicroscopy (Figure1). Interestingly,wenoted inthe LLM-domain B-GATA overexpressors an up to 10-fold in-crease in the number of stomata in the nonprotruding cell files ofhypocotyls that was accompanied by a strong increase in thenumber of cells in this cell file (Figures 1B to 1D).Whereas stomataformation in the wild-type hypocotyl is mainly restricted to theupper part of the hypocotyl (Berger et al., 1998), LLM-domain B-GATA-overexpressing seedlings displayed increased cell divi-sions and stomata numbers in the nonprotruding cell files alongthe entire hypocotyl (Figure 1B). Overexpressors of LLM-domainB-GATAs with mutations of the LLM-domain (GNC_LLM/AAAoxor GNCDCTox) were compromised in their ability to promotestomata formation but maintained the ability to promote cell di-visions (Figures 1B and 1C). Overexpression lines of GATA23(GATA23ox), which has a degenerate LLM-domain, aswell as ofthe HAN-domain B-GATA GATA19 (GATA19ox) did not pro-mote (GATA23ox) or only marginally promoted (GATA19ox)stomata formation and cell divisions in hypocotyls (Figures 1Band1C).Thus,specificallyoverexpressionof the fourLLM-domainB-GATA genes led to a strong increase in stomata formationin hypocotyls. Importantly, there was no apparent correlationbetween the effects of the different GATAs on stomata forma-tion and the expression of the respective transgenic proteins(Supplemental Figure 1).To examine whether LLM-domain B-GATAs are required for

stomata formation,weanalyzedstomataabundance inmutantsofGNC, GNL, GATA17, and GATA17L (Ranftl et al., 2016). We de-tected reductions in stomata numbers in the hypocotyls ofGATAgene single mutants, in the gnc gnl and gata17 gata17l doublemutants, and evenmorepronouncedeffects in thegncgnl gata17ltriple and the gnc gnl gata17 gata17l (gataq) quadruple mutants(Figures 1A to 1C and 1E). Furthermore, we found that the numberof nonstomatal cells in the nonprotruding cell file was not alteredbetween thewild type and the gataqmutant. On the other side, wealso did not find an increase in the number of meristemoids and

LLM-Domain B-GATAs Promote Stomata Formation 647

thus concluded that stomata formation in the hypocotyl was atleast inpart dependenton thepresenceof these fourGATAgenes.

Wenextexaminedstomatal abundanceat theabaxial sideof thecotyledons. There, we again detected significant increases in thenumber of stomata formed in the GATA overexpression lines(Supplemental Figure 2). In this case, the relative increases wereless pronounced than in hypocotyls (1.5- to 2-fold compared withup to 10-fold), and they could be observed in all GATA over-expression lines tested. GNCox variants with LLM-domain

mutations (GNC_LLM/AAAox or GNCDCTox) were again com-promised in their ability to induce stomata formation whencompared with GNCox wild-type overexpressors (SupplementalFigure 2). Conversely, we also detected a weak but significantdecrease in the number of stomata in the cotyledons of the gataqquadruple mutant when compared with the wild type(Supplemental Figure 2). In summary, we concluded that over-expression of the four LLM-domain B-GATAs can promote sto-mata formation in the hypocotyl and in cotyledons, that the four

Figure 1. LLM-Domain B-GATAs Promote Stomata Formation in the Hypocotyl.

(A) Representative photographs of 8-d-old light-grown seedlings with the genotypes as specified in the figure. Bar = 1 mm.(B) and (C)Scanningelectronmicrographs (B) andmagnification of a representative area from thesemicrographs ([C]; framedareas in [B]) of seedlingswiththegenotypes as specified in (A). In (C), all stomataof the selectedarea arecoloredpink, cells of one selectedprotruding cell file are coloreddark-green, andcells of one selected nonprotruding cell file are colored light-green. Bars = 250 µm.(D) and (E) Averages and standard errors of stomata number counts per visible surface in the hypocotyls of 8-d-old light-grown seedlings. The familymembershipof LLM-andHAN-domainB-GATAs is alsospecified.GATA23ox ismarkedwithanasterisk (*GATA23ox) since it representsanLLM-domainB-GATAwithadegenerateLLM-domain thatbehavesdifferently fromcanonical LLM-domainB-GATAs (Behringer et al., 2014).n$8.Student’s t testorMann-Whitney U test if the normality test failed in comparison to the wild-type sample: **P # 0.01 and ***P # 0.001; n.s., not significant.

648 The Plant Cell

LLM-domain B-GATAs participate in stomata formation in wild-type hypocotyls, and that their LLM-domain contributes to theirability to promote stomata formation. Since we noted in somegenotypes (GNC_LLM/AAAox or GNCDCTox) that the increasesin cell numbers in the nonprotruding cell file were not coupled toincreases in stomata numbers, we reasoned that the two pro-cesses may not be linked.

Todeterminewhether the effects ofB-GATAoverexpression onstomata formation in the hypocotyl were cell-autonomous, wemicrografted scions of wild-type and GNCox seedlings ontostocks of GNCox and wild-type seedlings, respectively. In bothcases, we found that the phenotypes of the GNCox seedlings onstomatanumbers, cell divisions, andgreening remained restrictedto the GNCox part of the micrograft, indicating that the effects ofGATA overexpression were not graft-transmittable and thereforelikely cell autonomous (Supplemental Figure 3).

The analysis of a set of transgenic lines expressing the GUS(pGNL:GUS) or a GUS:GFP reporter (pGNC:GUS:GFP andpGATA17:GUS:GFP) under control of ;2 kb GATA17, GNC, orGNL promoter fragments revealed that these LLM-domain B-GATA genes, which we had chosen as representatives for thefamily, were expressed in the hypocotyls and cotyledons of light-grown seedlings (Supplemental Figure 4). Since the strongstaining in the hypocotyls precluded a cell type-specific analysis,we also analyzed the GFP signal of pGNC:GUS:GFP and pGA-TA17:GUS:GFP. In thecaseofpGATA17:GUS:GFP,expressionofthe reporter was observed in stomatal lineage ground cells andstomatal meristemoids of hypocotyls and cotyledons aswell as incells of thehypocotyl cortex and themesophyll (Figure 2A). pGNC:GUS:GFP, in turn,wasmainlydetected in thehypocotyl cortexandin the cotyledon mesophyll but only rarely in isolated cells of theepidermis (Figure 2B). Thus, at least GATA17 but possibly alsoGNL are directly expressed in cells of the stomatal lineage.

TMM and SPCH Expression Are Restricted to StomatalLineage Cells in GNCox Lines

To establish whether the increased number of cells in the non-protruding cell file of LLM-domain B-GATA overexpressors wasa direct consequence of the activation of the stomatal lineagepathway,wenext examined theexpressionof thestomatal lineagemarkerspTMM:GUS:GFP (TMMpromoter fused toGUS:GFP)andpSPCH:nRFP (SPCH promoter fused to nucleus-targeted RFP) inthe LLM-domainB-GATAoverexpression lineGNCoxaswell as inthe gataq mutant (Figure 3) (Nadeau and Sack, 2002; Gudesblatet al., 2012). Becausewehadpreviously established that theLLM-domain B-GATAs had redundant biochemical functions (Beh-ringer et al., 2014) (Figure 1; Supplemental Figure 2), we restrictedthis and further genetic analyses to theGNC overexpression linesGNCox, and in relevant cases also to GNC_LLM/AAAox (Richteret al., 2010; Behringer et al., 2014). In this analysis, wedetected anincreased number of pTMM:GUS:GFP- and pSPCH:nRFP-expressing cells in hypocotyls of GNCox and GNC_LLM/AAAox(pTMM:GUS:GFP tested only) (Figure 3). However, we found thatmost cells originating from cell divisions in the nonprotruding cellfile did not express these markers, indicating that the increasednumber of cells in the nonprotruding cell file was not a conse-quence of a general activation of the stomatal lineage (Figure 3).

Conversely, we also found fewer cells expressing these twomarkers in the gataq mutant than in the wild type (Figure 3).

Distinct Suppression of the tmm Phenotype byB-GATA Overexpression

tmmmutants bear their name from the increased stomata formedin tmm mutant cotyledons; however, tmm mutants do not formstomata in hypocotyls and stems (Yang and Sack, 1995). Wetherefore wanted to examine to what extent the effects of GATAgene expression on stomata formation were dependent on TMM.Interestingly, when we introduced GNCox and GNC_LLM/AAAoxinto the tmm background, we found that both GNC transgenespromoted stomata formation in tmm hypocotyls (Figures 4A and4B). In addition, GNCox as well as GNC_LLM/AAAox activatedstomata formation in tmmmutant cotyledonsbeyond theextentofthe increased stomata formation observed in tmm (Figure 4C).Thus, overexpressionof theLLM-domainB-GATAGNCpromotesstomata formation downstream of or independently of TMM.Furthermore, overexpression of the LLM-domain mutated GNC,which has only aweak effect in thewild type, had a strong effect inthe stomata-deficient hypocotyls of the tmm mutant.The stomata formation defect in tmm mutant hypocotyls is

suppressed by loss-of-function mutants of CHL/EPFL6, CLL1/EPFL5, andCLL2/EPFL4, for instance, in tmmchalcll1or tmmchalcll1 cll2 mutants (Abrash and Bergmann, 2010). Interestingly,

Figure 2. Expression Analysis of pGATA17:GUS:GFP and pGNC:GUS:GFP.

Representative confocal microscopy images of the epidermis, cortex, andmesophyll of hypocotyls and cotyledons from 5-d-old light-grown wild-typeseedlingsexpressingpGATA17:GUS:GFP (A)orpGNC:GUS:GFP (B).The arrowheads point to one meristemoid as an example. In (B), the as-terisks indicate pGNC:GUS:GFP signals in the underlying cortex of hy-pocotyl cells as a nonepidermal fluorescence signal and in a guard cellwhere pGNC:GUS:GFP expression was occasionally observed. Bars = 25µm.

LLM-Domain B-GATAs Promote Stomata Formation 649

whenwegenerated tmmGNCoxand tmmGNC_LLM/AAAox linesand compared their phenotypes with tmm chal cll1, we noteddifferences in the distribution of the stomata on the hypocotyl.Whereas the distribution of stomata along the tmm chal cll1 hy-pocotyl resembled that of the wild type, where the majority ofstomata are formed in the upper part of the hypocotyl, stomatadistribution in tmm GNCox or tmm GNC_LLM/AAAox resembledthat of GNCox and GNC_LLM/AAAox in the wild-type back-ground, where stomata formed along the entire hypocotyl (Figure4B). Furthermore, while tmm chal cll1 formed stomatal clusters,clusters were only occasionally observed in tmm GNCox ortmm GNC_LLM/AAAox hypocotyls (Figure 4B). By contrast, theabundance of such clusters was increased in cotyledons of tmmGNCox or tmm GNC_LLM/AAAox when compared with tmm,while noclusterswhereobserved inGNCoxorGNC_LLM/AAAoxin thewild-type background (Figure 4C).We thus concluded thatoverexpression of the LLM-domain B-GATA GNC promotedstomata formation in hypocotyls downstream of TMM and af-fected stomatal patterning in cotyledons.

Since CHAL/CLL mutations suppress the tmm phenotype, wewere interested in examiningwhetherCHAL/CLLmutations couldsuppress the stomata formation defect of the gataq (gnc gnl

gata17 gata17l) quadruple mutant. To this end, we generatedgataq chal cll1. However, gataq chal cll1 failed to suppress thegataq phenotype, suggesting that the LLM-domain B-GATAspromoted stomata formation downstream of or independently ofTMM and CHAL/CLL (Figure 4D).

Interaction between LLM-Domain B-GATAs and SDD1

sdd1mutants producemore stomata in the cotyledons than doesthe wild type and these frequently occur in stomatal clusters

Figure 3. Expression Analysis of pTMM:GUS:GFP and pSPCH:nRFP.

Representative confocal microscopy images of hypocotyl epidermis cellsof 5-d-old light-grown seedlings expressing pTMM:GUS:GFP (A) orpSPCH:nRFP ([B] and [C]). Arrowheads mark several meristemoids ex-pressing thepTMM:GUS:GFPorpSPCH:nRFPreporters. In (A)and (B), cellwalls were stained with propidium iodide; in (C), the propidium iodidestaining was omitted to show that the nuclear signal originates from thepSPCH:nRFP reporter rather than from the propidium iodide cell wall stain.Bars = 50 µm.

Figure 4. GNC Overexpression Suppresses the Stomata Formation De-fects in tmm Mutant Hypocotyls.

(A) and (D) Averages and standard errors of stomata number counts pervisible surface in the hypocotyls of 8-d-old light-grown seedlings. n $ 8.Student’s t test or Mann-Whitney U test if normality test failed: **P# 0.01and ***P # 0.001; n.s., not significant.(B) and (C) Scanning electron micrographs of seedling hypocotyls (B) andabaxial cotyledonsurfaces (C). In (B), cells colored inpink areall stomataofthe selected area. Note the differential distribution of stomata in the hy-pocotyls between tmm GNCox (along the entire hypocotyl) and tmm chalcll1 (upper part of the hypocotyl only). Also note the increased formation ofstomatal clusters in cotyledons of tmm GNCox, tmm GNC_LLM/AAAox,and tmm chal cll1. Bars = 100 µm (B) and 50 µm (C).

650 The Plant Cell

(Berger and Altmann, 2000). We were interested in examiningwhether the increased stomata numbers in sdd1were dependenton the LLM-domain B-GATAs andwe thus generated sdd1 gataq.When comparing sdd1 gataq with sdd1, we detected a strongreduction in the number of stomata in sdd1 gataq hypocotyls, butonly a minor decrease in stomata numbers in cotyledons (Figure5). Since the number of stomata in sdd1 gataq hypocotyls washigher than that observed in gataq, we hypothesized that SDD1may repress stomata formation through the GATAs, but alsothrough one or several other mechanisms that act in parallel.

B-GATAs Act Upstream of SPCH, MUTE, and SCRM SCRM2

Stomata formation requires the consecutive activities of the threebHLH transcription factors SPCH, MUTE, and FAMA, whichfunction together with the interacting bHLH proteins SCRM andSCRM2 (MacAlister et al., 2007; Pillitteri et al., 2007; Kanaokaet al., 2008). To examine whether stomatal differentiation down-streamof theB-GATAswasdependent on thesebHLH factors,weintroduced GNCox into spch, mute, and scrm scrm2 loss-of-functionmutants. Inall threeof the resultinggeneticcombinations,namely, GNCox spch, GNCoxmute, and GNCox scrm scrm2, weobserved that stomata formation was blocked in the hypocotyl aswell as in the cotyledons (Figure 6; Supplemental Figure 5).Whereas differentiated stomata were absent in GNCoxmute, thisbackground accumulated an increased number of meristemoids,which were recognizable as small cells originating from asym-metric cell divisions (Figures 6D to 6G). This indicated that sto-matal differentiation had been initiated but could not proceed inthe mute loss-of-function mutant. Importantly, the general in-creases in cell numbers in the nonprotruding cell file that weretypical of GNCox lines could also be observed in GNCox spch,GNCox mute, and GNCox scrm scrm2 (Figures 6C and 6G;Supplemental Figure 5B). Additionally, the strong effects ofGNCox on greening, that are particularly prominent at the hy-pocotyl base, were maintained in the different GNCox genotypes(Figures 6B and 6F; Supplemental Figure 5A). This suggestedagain that the effects of B-GATA overexpression in promotinggreening and cell divisions in the nonprotruding cell files ofhypocotyls were not linked to their role in promoting stomataformation.

In a complementary experiment, we evaluated the importanceof the LLM-domain B-GATAs in the scrm-D gain-of-functionmutant,whichproducesanexcessivenumberofstomata (Kanaokaet al., 2008). Importantly, a scrm-D gataq mutant displayeda significant reduction of stomata formed in the hypocotyl whencompared with scrm-D, whereas the number of stomata formedin the cotyledons were comparable between these two geno-types (Supplemental Figures 5C to 5E). This showed that theactivityof thegain-of-functionSCRM-Dwascompromised in theabsence of the four GATA genes mutated in gataq.

The Effects of B-GATAs on Stomatal Development Are LightDependent and PIF Dependent

Stomata formation in wild-type seedlings is at least partially de-pendent on the activation of the phyB phytochrome receptor byred light and the consequent inactivation of PIF transcription

factors, which are targeted for protein degradation in the light(Casson et al., 2009; Casson andHetherington, 2014).We notedthat the effects of B-GATAoverexpression on stomata formationand cell divisions in the hypocotyl were strictly light dependent(Figure 7). While dark-grown wild-type seedlings as well asGNCox overexpressors did not form stomata in the hypocotyl,stomata formed in seedlings grown in white light as well as red,far-red, and blue light (Figure 7). Interestingly, light-dependentstomata formation was compromised in the gataqmutants and,conversely, enhanced in GNCox (Figure 7). This indicated thatthe four GATA genes mutated in gataq were required and rate-limiting for stomata formation in response to light and thatGATAdosage was also limiting for stomata formation in the wild type(Figure 7).The phyB and PIF-dependent repression of stomata formation

in the dark is suppressed in pif1 pif3 pif4 pif5 (pifq) quadruplemutant seedlings where the requirement for the light inactivationof the respective PIF proteins no longer exists (Martínez-Garciaet al., 2000; Leivar et al., 2008). Since dark-grown pifq mutantsare partially etiolated, we were interested in examining whether

Figure 5. Genetic Interaction between SDD1 and LLM-Domain B-ClassGATAs.

(A) and (C) Averages and standard errors of stomata number counts percotyledon area (600 µm2) (A) and hypocotyl (C) of 8-d-old light-grownseedlings. n$ 8. Student’s t test or Mann-Whitney U test if normality testfailed: **P # 0.01 and ***P # 0.001; n.s., not significant.(B) and (D) Scanning electron micrographs of abaxial cotyledon surfaces(B) and seedling hypocotyls (D). All stomata of the selected area arecolored in pink. Note the reduced number of stomata formed in hypocotylsof sdd1 gataq compared with sdd1 (D) but the presence of stomatalclusters in sdd1 as well as in sdd1 gataq (B). Bars = 100 µm.

LLM-Domain B-GATAs Promote Stomata Formation 651

LLM-domain B-GATA overexpression was sufficient to inducestomata formation in this background in the dark. To this end, weanalyzed pifq GNLox, an overexpression line of the GNC paralogGNL that we had generated in the context of earlier work (Richteret al., 2010). Indeed, expression of the GNLox transgene in pifqresulted in increases in stomata formation in dark-grown seedlinghypocotyls that exceeded the number of stomata formed in pifq(Figure 8D). Importantly, GNLox and pifqGNLox seedlings, whengrown in the light, produced comparable numbers of stomata inthe hypocotyl (Figure 8E). Taken together, these findings sug-gested, on the one hand, that LLM-domain B-GATAs are rate-limiting factors for stomata formation in hypocotyls downstreamof PIFs and, on the other, that the GATAs are not sufficient toinduce stomata formation in hypocotyls, but require other factorsdownstream of PIFs (or the PIFs themselves) in this process.

The Expression of LLM-Domain B-GATAs Is Light Regulated

GNL hadoriginally been identified as a red light-induced gene, butthe functional role of this regulation had remained unexplored(Naito et al., 2007). We examined the regulation of the four LLM-domain B-GATA genes, GNC, GNL, GATA17, and GATA17L, inresponse to red, far-red, and blue light and found that, mostprominently, the expression of GNL but to some extent also ofGNCwas induced in all three light conditions (Figures 9A and 9B;Supplemental Figure 6). At the same time, the expression ofGATA17 and GATA17L was slightly but significantly repressedafter any of the respective light treatments in the wild type(Supplemental Figure 6). These transcriptional light responseswere compromised in the respective phyA phyB or cry1 cry2

receptor mutants and, in the case of the red light-responsivetranscription, also in the pifq mutant (Figures 9A and 9B;Supplemental Figure 6).Wehadpreviously shown thatGNCandGNLwere repressedby

PIFs and that the expression particularly of GNL but also of GNCwas increased in light-grown pif mutant seedlings (Richter et al.,2010). Since increased expression of the GATA genes may ac-count for stomata formation in dark-grown pifq seedlings, weexamined the expression of the fourGATAs (GNC,GNL,GATA17,and GATA17L) by qRT-PCR (Figure 9C). This analysis revealeda strong upregulation particularly ofGNL, but also ofGNC, in pifq,suggesting that the upregulation of these GATAs may be caus-ative for the induction of stomata formation in dark-grown pifqseedlings. Since our analyses of GATA mutant phenotypes hadbeen concentrated on the gataq quadruple mutant, but GNLwas particularly strongly upregulated after red light treatment, wehypothesized that gnl single mutants may already be compro-mised in the light-dependent induction of stomata formation.However, gnl single mutants were indistinguishable from the wildtype with regard to stomata formation in the hypocotyls of redlight-shifted dark-grown seedlings (Figure 9D). The impairment ofstomata formation observed in gataqmutants grown in the sameconditions further substantiated our conclusions phrased above(Figure 9D) and from our earlier studies showing that the genesencoding LLM-domain B-GATAs act in a functionally redundantmanner (Bi et al., 2005; Richter et al., 2010; Behringer et al., 2014;Ranftl et al., 2016).Since the light-dependent effects of LLM-domain B-GATA

overexpression inGNCox (or GNLox)may potentially be the resultof the differential degradation or accumulation of the transgenic

Figure 6. SPCH and MUTE Act Downstream of LLM-Domain B-GATAs in Stomata Formation.

(A) and (E)Averages and standard errors of stomata number counts per visible surface in the hypocotyls of 8-d-old light-grown seedlingswith the specifiedgenotypes. n = 8. Student’s t-test or Mann-Whitney U test if normality test failed: *P # 0.05, **P # 0.01, and ***P # 0.001.(B) and (F) Representative photographs of 8-d-old light-grown seedlings. Bars = 1 mm.(C), (D), and (G)Scanningelectronmicrographsof representative areasof seedlinghypocotyls ([C]and [G]) and theabaxial sideof cotyledons (D). In (C)and(G), cells of one selected protruding and nonprotruding cell file are colored in dark-green and green, respectively. Bars = 100 µm.

652 The Plant Cell

proteins, we examined the abundance of the HA-tagged GNCprotein inGNCoxby immunoblot analysis in dark- and light-grownseedlings aswell as in dark-grown seedlings that hadbeen shiftedto the light for one day. Sincewe detected similar levels of GNCoxinall lightconditions,weexcluded thepossibility that theobserved

effects were the result of differences in protein abundance(Supplemental Figure7A).Wealso found that theexpressionof thepTMM:GUS:GFP marker in the hypocotyls and cotyledons ofdark-grown seedlings was restricted to a comparable number ofmeristemoids in thewild-type,gataq, andGNCoxseedlings,whiletheir number varied in light-grown seedlings in correlationwith thenumbers of stomata formed in these genotypes (SupplementalFigures 7B and 7C).

Figure 7. Red, Far-Red, and Blue Light Promote Stomata Formationthrough LLM-Domain B-GATAs in Hypocotyls.

Representative areas from scanning electron micrographs of hypocotylsfrom8-d-oldseedlingsgrowninthedark (A), in far-redlight (0.35µmolm22s21)(B), red light (7.5 µmol m22 s21) (C), blue light (4.25 µmol m22 s21) (D), orwhite light (150 µmolm22 s21) (E) of the genotypes specified in the figure.Bar = 100 µm.

Figure 8. LLM-Domain B-GATAs Promote Stomata Formation Down-stream of PIFs.

(A) Representative areas from scanning electron micrographs of hypo-cotyls from 8-d-old dark-grown seedlings. All stomata and all mer-istemoids are colored in pink and blue, respectively; meristemoids are alsohighlighted by asterisks. Cells of one selected protruding and non-protruding cell file are colored in dark-green and green. Note the relativeincrease in meristemoids in pifq GNLox (asterisks). Bar = 50 µm.(B) and (C) Representative photographs of 8-d-old dark-grown (B) andlight-grown seedlings (C). Bar = 1 mm.(D) and (E) Averages and standard errors of counts of stomata andmeristemoids per hypocotyl area (800 µm2) of 8-d-old dark-grown seed-lings (D) and per hypocotyl of light-grown seedlings (E). n$ 8. Student’s ttest or Mann-Whitney U test if normality test failed: **P# 0.01 and ***P #

0.001; n.s., not significant.

LLM-Domain B-GATAs Promote Stomata Formation 653

SPCH Is a Strongly Light-Regulated Gene

Todeterminewhether theexpressionofanyof thecritical regulatorsof stomata development was light regulated, we examined theexpression of key regulators of stomata formation, namely, SPCH,MUTE, FAMA, SCRM, SCRM2, STOMAGEN, and TMM, in dark-grown seedlings and following short-term red light treatments byqRT-PCR (Figure 10A). These analyses identifiedSPCH as the onlygene that was already strongly induced after a 30-min red lighttreatment (Figure 10A). Only after 120min did we detect significantincreases in the expression of other regulatory genes, namely,MUTE, SCRM2, and TMM (Figure 10B). Since SPCH expression inthe dark could potentially be repressed by PIFs and relieved fromrepression in red light, we examined the levels of SPCH in the dark-grown pifq mutants as well as the kinetics of SPCH expression inpifq mutants. There, we detected increases in the expression ofSPCH indark-grownseedlingsandacomparatively faster inductionofSPCH expression in comparison to thewild typewhen seedlingswere shifted to red light (Figures 10C and10D). This suggested thatPIFsmay indeedsuppressSPCHexpression inthedarkandthat thepifqmutant, where four of the at least seven red light-sensitive PIFproteins from Arabidopsis are mutated (Leivar and Monte, 2014),was sensitized to the red light treatment.Inversely, we observed thatSPCH expression in dark-grown gataq

seedlingswas comparable to that of wild-type seedlings and that thered light-dependent induction of SPCH was delayed in the gataqmutant (Figures 10C and 10D). This suggested that the GATAsmaypositively regulateSPCHexpression in thered lightsignalingcascade.Since the experiments described above indicated that the

LLM-domain B-GATAs may directly regulate SPCH expression,wewere interested in identifyingGNL targetgenes. To this end,wemade use of a transgenic line expressing the LLM-domainB-GATA GNL fused to YFP, an HA-tag, and the hormone bindingdomain of the rat glucocorticoid receptor (GR) in a gnc gnlbackground. Such a GR fusion allows the cytoplasm-to-nucleusshuttling of the GNL:YFP:HA-GR protein after treatment withthe glucocorticoid dexamethasone (Lloyd et al., 1994). We thenperformed chromatin immunoprecipitation of GNL:YFP:HA-GRfollowed by next-generation sequencing (ChIP-seq) using 10-d-old light-grown seedlings that had been treated with 10 µMdexamethasoneor amock control for 4 h. Aftermapping the readsto theArabidopsis genome,we inspectedall known lociwith a roleinstomataldevelopmentand identified, asasinglegenehit, apeaklocated 1479 bp upstream of the SPCH start codon that wasabsent in the control sample (Figure 10D). This peak aligned witha TGATAA promoter element, which was thus a predicted pro-moter element for GATA transcription factors (Figure 10D). Wethen confirmed the binding to an amplicon comprising this pro-moter element (set 1) by qPCR after an independent ChIP witha transgenic lineexpressingHA-taggedGNLfromaGNLpromoterin the gnc gnlmutant background (ProGNL:GNL:HA; Figure 10E).When we tested the neighboring region (set 2) that had not beenidentified in the ChIP-seq experiment, we did not find such anenrichment (Figure 10F). When taken together with the findingsthat the red light-dependent induction of SPCH was impaired ingataq mutants, this finding suggested that GNL was a goodcandidate for a protein that promotes SPCH expression andconsequently stomata formation.

Figure 9. GNL and GNC Are Light-Regulated Genes and DifferentiallyExpressed in pifq.

(A) and (B) Averages and standard errors of normalized fold changes fromqRT-PCR analyses of GNL (A) and GNC (B) obtained from 5-d-old dark-grownseedlingsof thespecifiedgenotypesafter 60min red, far-red, orbluelight exposure. For each genotype, fold changes were calculated relativeto the values obtained from dark-grown seedlings of the respectivegenotypes. GATA17 and GATA17L expression was not light activated(Supplemental Figure 6).(C) Averages and standard deviations of normalized fold changes fromqRT-PCR analyses of GNC, GNL, GATA17, and GATA17L expression indark-grownwild-typeandpifqseedlings.Foreachgene, foldchangeswerecalculated relative to the values obtained with the respective dark-grownwild-type seedlings.(D)Averages and standard errors of stomata numbers in the upper 500-µmsection of the imaged side of hypocotyls grown for 3 d in the dark and anadditional 4 d in red light. n$ 8. Student’s t test or Mann-Whitney U test ifnormality test failed: *P # 0.05, **P # 0.01, and ***P # 0.001; in all othercases, differences are not significant.

654 The Plant Cell

DISCUSSION

We identified the redundantly acting LLM-domain B-GATAtranscription factors GNC, GNL, GATA17, and GATA17L ascritical regulators of light-dependent stomata formation in thehypocotyl of Arabidopsis (Figures 1 and 7). We show that theseLLM-domain B-GATA genes act upstream of SPCH and otherbHLH transcription factors required for stomatal differentiation(Figure6) anddownstreamof far-red, red, andblue light signalingpathways (Figure7).Furthermore,wefindthatGATAoverexpression

can suppress the loss of stomata formation phenotype in tmmhypocotyls and that the loss of the GATAs can partially impair in-creased stomata formation as observed in the sdd1mutant (Figures4 and 5).While the overexpression of these GATA factors strongly

promotes stomata formation in light-grown hypocotyls, theiroverexpression does not show an effect in hypocotyls ofdark-grown wild-type seedlings (Figure 7). In pifqmutants, four ofthe at least seven light-sensitive PIFs from Arabidopsis are mu-tated (Leivar and Monte, 2014). Our observation that GATA

Figure 10. SPCH Is a Red Light-Regulated Gene.

(A)Averages and standard errors of normalized fold changes fromqRT-PCRanalyses of genes involved in stomata formation after red light exposure for theduration specified in the figure. For each gene, fold changes were calculated relative to the values obtained with the respective dark-grown seedlings.(B)Averagesandstandarderrorsofstomatanumbers in theupper500-µmsectionof the imagedsideofhypocotylsgrown for3d in thedarkandanadditional4 d in red light. n $ 8. P values in parentheses refer to tests of the same time points between different genotypes.(C) Averages and standard errors of normalized fold changes from qRT-PCR analyses of SPCH expression in dark-grown wild-type, pifq, and gataqseedlings. Fold changes were calculated relative to the values obtained with the wild-type control.(D) Graphic displays of the CSAR enrichment scores after mapping of sequencing reads from two ChIP replicate samples of GNL:YFP:HA:GR afterdexamethasone induction (sample 1 and sample 2) and two mock controls (mock 1 and mock 2) over the genomic region around SPCH. Set 1 representsa region identifiedbyChIP-seqofGNL:YFP:HA:GRthatcontainsaGATAmotif and thatwassubsequently tested forbindingbyChIP-qPCR; set2 representsa neighboring control region. TheSPCHgenemodel is displayed,where boxes are exonic regions and lines represent introns.Colored triangles identify fourdifferent types of GATA sequence motifs, putative GATA transcription factor binding sites that can be found in the represented genomic sequence.(E) and (F)Results from independent ChIP-qPCR analyses after immunoprecipitation of GNL:HA from aProGNL:GNL:HA gnc gnl transgenic line and a gncgnlnegative control over an amplicon from theSPCHpromoter containing aTGATAAsequence (set 1; [E]) and aneighboringamplicon asanegative control(set 2; [E]). Student’s t test or Mann-Whitney U test if normality test failed: *P # 0.05, **P # 0.01, and ***P # 0.001; n.s., not significant.

LLM-Domain B-GATAs Promote Stomata Formation 655

overexpression can promote stomata formation in the hypocotylsofdark-grownpifqmutantssuggests thatGATAsact togetherwithother factors downstream fromPIFs, or with the PIFs themselves,to promote stomata formation (Figure 8). Conversely, stomataformation in the hypocotyl is impaired in red light and, thus, in lightconditions where PIFs are inactivated by proteolytic degradation.This indicates that the GATAs are critical regulators of light-dependent stomata formation in the hypocotyl. GATA over-expression results in stomata formation along theentire hypocotyl(Figures 1 and 4), whereas stomata formation in the wild type isgenerally restricted to theupperpart of thehypocotyl (Berger et al.,1998). This may be taken as an indication that GATA expressionlevels are limiting in this process in wild-type hypocotyls.

We previously provided evidence for a direct regulation andtranscriptional repression of the B-GATAsGNC andGNL by PIFs,and we also showed that GNC and GNL transcript abundance isupregulated in light-grown pifq quadruple mutants (Richter et al.,2010). We now show that the LLM-domain B-GATA factor GNL,and to some extent alsoGNC, are transcriptionally activated afterred light treatment of dark-grown seedlings and that their ex-pression is increased in dark-grown pifq mutants (Figure 9). Thisderepression of GATA gene expression in the pifq mutants andtheir light-dependent expression is in line with a model where theGATAs are downstream from PIF transcription factors in theregulation of light-regulated stomata formation.

Interestingly,ourexaminationofSPCHexpression following redlight treatment of dark-grown seedlings indicated thatSPCHwas,among several stomatal development regulators tested, the onlygene rapidly induced in response to red light (Figure 10). SPCHmay therefore represent the critical light-regulated target in thecontext of light-dependent stomata formation. The recently re-ported findings that SPCH directly regulates a broad range ofgeneswith a role in stomata formation and patterning strengthensthe predicted critical role of SPCH in regulating stomatal de-velopment, in general and here specifically downstream from thelight signaling pathway (Lau et al., 2014).

The strong red light-induced expression ofSPCH suggests thatSPCH functions downstream of phytochrome and PIFs. In linewith the notion that the red-labile PIFs act as repressors of SPCHexpression, red light-inducedSPCH expression is stronger in pifqmutants than in the wild type, suggesting that the pifq mutants,in which four but not all red light-regulated PIFs are mutated,are sensitized to red light (Figures 10B and 10C). Inversely,we observed a delayed induction of SPCH expression in thegataqmutants, suggesting that theLLM-domainB-GATAsmaybeactivators of SPCH expression (Figures 10B and 10C). Sucha model finds further support in our observation from a ChIPanalysiswith theLLM-domainB-GATAGNLthat identifiedaGATAmotif-containing element in the SPCH promoter (Figures 10Dto 10F). These ChIP-seq data were obtained from light-grownseedlings, and future experiments will have to explore thebiological relevance of this binding for SPCH expression andlight-dependent stomata formation.

Here, we show that the effects of the gain- and loss-of-LLM-domain B-GATA function on stomata formation are most prom-inent in the hypocotyl. While the positive effects of LLM-domainB-GATA overexpression on stomata formation can still beobserved in cotyledons, the negative effects of the loss of

LLM-domain B-GATA expression in the gataq quadruplemutant in cotyledons are significant but comparatively subtle.This may be explained through the activities of the other twoLLM-domain B-GATAs, GATA15 and GATA16, that are stillactive in the gataq mutant or possibly also the compensatoryupregulation of LLM-domain B-GATA gene expression in thegataq mutant background, as suggested already by otherstudies of this gene family (Richter et al., 2010, 2013a, 2013b;Behringer et al., 2014; Ranftl et al., 2016). Alternatively, andspecifically with regard to the weak effects of the loss ofGATAs on stomata formation in cotyledons (SupplementalFigure 2), it is likely that other input pathways exist thatregulate stomata formation in addition to the LLM-domainB-GATAs. For example, brassinosteroids were shown to regulateSPCH by posttranslational modification of the SPCH proteinand brassinosteroid signaling could thus act in parallel anddownstream of the B-GATAs in the regulation of stomatal de-velopment (Casson and Hetherington, 2012; Gudesblat et al.,2012; Kim et al., 2012). Furthermore, it may be possible thatthe other B-GATA genes also participate in the regulation ofstomata formation in this tissue, since we observed note-worthy effects of their overexpression in cotyledons, but notin hypocotyls (Supplemental Figure 2).The presence of the C-terminal LLM-domain defines

LLM-domain B-GATA factors, but the biochemical function ofthe LLM-domain is unclear. Here, we show that the overexpressionof LLM-domain B-GATAs with a mutation in the LLM-domaincompromises the ability of the B-GATAs to promote stomataformation (Figure1;SupplementalFigure2).However,LLM-domainmutated B-GATAs are not fully impaired in this activity, as mostobviously shown in the tmm mutant, where overexpression ofLLM-domain mutated GATAs efficiently promoted stomata for-mation (Figure 4). Furthermore, our study shows that the promotionof stomata formation is not restricted to LLM-domain-containingB-GATAs, since the other B-GATAs also promote stomata for-mation, at least to some extent, in hypocotyls and cotyledons(Figure 1; Supplemental Figure 2). A better understanding of thebiochemical function of the LLM-domain and of the target genespecificity of the different GATAswill help unravel how this domaincontributes to the activation of the stomata formation pathway.Besides stomata formation, the overexpression of the

LLM-domain B-GATAs also promoted greening and cell divisionsin the nonprotruding cell file of seedling hypocotyls (Figure 1).Here, we did not investigate the molecular basis of this GATA-dependent activity butmade several observations that allowed usto rule out the possibility that the effects of GATA overexpressionon greening, cell division, and stomata formation are interrelated.First, we observed that not all cells in the nonprotruding cellfile express the stomatal lineage reporters for TMM and SPCH,indicating that not all cells in this cell file have acquired stomatallineage fate (Figure 3). Thus, the supernumerary cells in thenonprotruding cell file are not undifferentiated stomatal lineagecells. Second, we observed increased cell numbers in the non-protruding cell file in the GNCox spch and GNCox mute back-grounds as well as in the GNCox scrm scrm2 background, whichdo not form stomata (Figure 6; Supplemental Figure 5). Thus, theincreased cell divisions as observed in the GATA overexpressorsare not a consequence of increased stomata formation.

656 The Plant Cell

Loss-of-function mutants of the LLM-domain B-GATA genesare impaired in greening and GATA overexpressors accumulatemore chlorophyll than the wild type (Bi et al., 2005; Richter et al.,2010; Hudson et al., 2011; Chiang et al., 2012; Behringer et al.,2014; Ranftl et al., 2016). Although it is intriguing that bothgreening and stomata formation are related to the efficiency ofphotosynthesis, we exclude, based on our data, a causal linkbetween the stomata formation and greening phenotypes of theloss-of-function and overexpression backgrounds. One obser-vation that supports this conclusion is the fact that defectivestomata formation in the hypocotyl as observed in GNCox spch,GNCox mute, as well as in GNCox scrm scrm2 does not impairhypocotyl greening (Figure 6; Supplemental Figure 5). Thisobservation is most prominent in the GNCox scrm scrm2 back-ground (Supplemental Figure 5) and less prominent in the GNCoxspch and GNCoxmute backgrounds because the spch andmutemutants themselves have greening defects (Figure 6). Thus, ourgenetic data suggest that the role of the LLM-domain B-GATAs inpromoting stomatal development can be uncoupled from theirapparent role in promoting greening.

The molecular mechanisms that control greening in anLLM-domain B-GATA-dependent manner have, at least tosome extent, already been elucidated (Bi et al., 2005; Hudsonet al., 2011; Chiang et al., 2012). However, how the LLM-domainB-GATAspromote cell divisions and how this is achieved in a celltype-specificmanner, remains unclear (Figure1).Wenoted in theliterature two previously published reports that may be relevantin this context. First, it was shown that the overexpression ofthe E2F-DP cell cycle repressors induces, similarly to the GATAoverexpressors described here, cell divisions specifically inthe nonprotruding cell file (De Veylder et al., 2002). Second, itwas reported that mutants of the D-type cyclin CYCD4 havea reduced number of cells and stomata in the nonprotruding cellfile of hypocotyls and, conversely, that CYCD4 overexpressiongives rise to an increased number of cells and stomata in this cellfile (Kono et al., 2007). Thus, theGATA factorsmay act upstreamor downstream of E2F-DP and CYCD4. Our existing ChIP-seqdata set does not give any indications that the GATAs examinedhere actively regulate cell cycle genes. Howcell cycle regulation,stomatal differentiation, and cell file specificity of these regu-latory mechanisms function together will be an interesting topicfor future research.

METHODS

Biological Material

All experiments were performed in the Arabidopsis thaliana ecotype Co-lumbia (Col). The following B-GATA gene-related mutants and transgeniclines were used in this study: gnc (SALK_001778), gnl (SALK_003995),gata17 (SALK_101994),gata17l (SALK_026798),gataq, GNCox,GNC_LLM/AAAox, GNC_DCTox, GATA17ox, GATA19ox, GATA23ox, pGNL:GUS, andpGNL:GNL:HA gnc gnl (Richter et al., 2010, 2013a; Behringer et al., 2014;Ranftl et al., 2016). The experiment with pifq GNLox and GNLox was per-formedwith the overexpression construct described by Richter et al. (2010),and all other experiments with GNL overexpression were performed withGNLox as described by Behringer et al. (2014).

The following mutants were obtained from NASC (Nottingham Arabi-dopsisStockCenter):phyAphyB (phytochromeA-211phyB-9) (Reedet al.,

1993), tmm (too many mouths-1) (Nadeau and Sack, 2002), tmm chal cll1(tmm-1 challah-2 challah-like1-1), and chal cll1 (Abrash et al., 2011). Thefollowing previously publishedmutants and reporter lineswere used in thisstudy and generously provided by Dominique Bergmann (Stanford, CA),Keiko Torii (Seattle, WA), or Eugenia Russinova (Ghent, Belgium): pTMM:GUS:GFP (NadeauandSack, 2002), pSPCH:nRFP (Gudesblat et al., 2012),spch (speechless-3) (MacAlister et al., 2007),mute (mute) (Pillitteri et al., 2007),scrmscrm2 (ice1-2scream2-1),scrm-D (scream-D) (Kanaokaetal.,2008),andsdd1 (stomata density and distribution1) (Berger andAltmann, 2000). pif1 pif3pif4 pif5 (pifq [phytochrome interacting factor quadruple]; pif1-1 pif3-3 pif4-2pif5-3) was generously provided by Peter Quail (Berkeley, CA) (Leivar et al.,2008) and the cry1 cry2mutant by Chentao Lin (Los Angeles, CA) (Mockleret al., 1999). Primers for genotyping are listed in Supplemental Data Set 1.

Cloning Procedures

The GATA17Lox transgenic line was generated as previously describedusing the primers listed in Supplemental Data Set 1 (Behringer et al., 2014).The pGNC:GUS:GFP and pGATA17:GUS:GFP constructs where gener-ated using Gateway technology (Life Technologies) with pDONR201 andthe destination vector pFAST-G04 (Shimada et al., 2010).

To generate 35S:GNL:YFP:HA:GR, an AscI restriction site was in-troduced after the HA-tag in GNLox as described (Behringer et al., 2014).The GR domain was amplified from the vector pTA7002 (Aoyama andChua, 1997) as a fragment flanked by AscI sites and inserted into themodified GNLox. The resulting construct was transformed into the gnc gnldouble mutant using the floral dip method (Clough and Bent, 1998). Thefunctionality of the transgene in the transgenic lines was verified bycomplementation and confocal analysis of GNL:YFP:HA distribution.Primers used for cloning are listed in Supplemental Data Set 1.

Microscopy

Fluorescence microscopy was performed with an Olympus FV1000confocal laser scanning microscope. Scanning electron microscopy wasperformed with a TM-3000 table-top scanning electron microscope(Hitachi).

Physiological Experiments

All plants were cultivated on sterile 0.53 Murashige and Skoog mediumwithout sugar under continuous white light (150 µmol m22 s21). The ex-periments were repeated three times with comparable results and theresult of one representative experiment is shown. For experiments withdifferent light qualities, seedlingswere grown in CLF FloraLED chambersusing red (7.2 µmol m22 s21), far-red (0.35 µmol m22 s21), or blue light(4.25 µmol m22 s21).

Immunoblotting

Toassayforproteinexpression intransgenicplants,totalproteinextractswereprepared from8-d-old light-grownseedlings inanextractionbuffercontaining50 mM Tris-HCl, pH 7.0, 150 mM NaCl, 0.5% (v/v) Triton X-100, 1 mMphenylmethylsulfonyl fluoride, and Protease Inhibitor Cocktail (Roche). Totalprotein(45µg)wasseparatedona10%(v/v)SDS-PAGEgelanddetectedwithanti-HA-horseradish peroxidase (Roche) or anti-HSC70 (HEAT SHOCKCOGNATE PROTEIN 70; Agrisera) and anti-rabbit-peroxidase (Sigma-Aldrich). A part of a Coomassie Brilliant Blue-stained gel correspondingto the 50-kD molecular mass proteins was used as a loading control.

qRT-PCR

To measure the light-dependent induction of gene expression byqRT-PCR,5-d-olddark-grownseedlingswere transferred toCLFFloraLED

LLM-Domain B-GATAs Promote Stomata Formation 657

chamberswith constant red (7.2µmolm22 s21), far-red (0.35 µmolm22 s21),or blue (4.25 µmol m22 s21) light. Total RNA was extracted with an RNeasyPlantMini Kit (Qiagen). Then, 2 µg total RNAwas reverse transcribedwith anoligo(dT) primer and M-MuLV reverse transcriptase (Fermentas). Transcriptlevelsweredeterminedusing iQSYBRGreenSupermix (Bio-Rad) inaCFX96real-time system cycler. The results were normalized to PP2A (PROTEINPHOSPHATASESUBUNIT2A; AT1G13320) in red and far-red light and toACT2 (ACTIN2; AT3G18780) in blue light. Expression in the dark sampleswas set to one. Primers for qRT-PCR analyses are listed in SupplementalData Set 1.

ChIP

For ChIP-seq analyses, 10-d-old 35S:GNL:YFP:HA:GR gnc gnl seedlingsweregrownonGMmediumunder constantwhite light and then induced for4 h with 10 µM dexamethasone or with a corresponding mock solution.Three biological replicate samples (2 g tissue) were fixed for 20 min in 1%formaldehyde, processed using GFP-TRAP_A (Chromotek) and preparedfor DNA sequencing using Illumina MiSeq sequencing as previously de-scribed (Kaufmann et al., 2010). Illumina sequencing reads were thenmapped to theArabidopsis genome (TAIR10) usingSOAPv1 (Li et al., 2008)with the following settings: threemismatches,mapping touniquepositionsonly, no gaps allowed, iteratively trimming set to 41 to 50. Subsequentanalysis for peak identification was performed with CSAR, thereby re-taining only peaks with an false discovery rate < 0.05 as statistically sig-nificant peaks (Muiño et al., 2011). The data from this analysis are availableas PRJNA291959 at NCBI-SRA (http://www.ncbi.nlm.nih.gov/sra/). Toverify thebinding ofGNL to theSPCHpromoter, ChIP-PCRwasperformedwith pGNL:GNL:HAgnc gnl aswell asgncgnl seedlings grown onGMagarplates for 10 d under long-day conditions (16 h light/8 h dark). Plant tissuewas subsequently fixed for 20min in a solution containing1%formaldehydeand the remainder of the ChIP experiment was performed as previouslydescribed using ChIP-grade anti-HA tag antibodies (Abcam) (Kaufmannet al., 2010). Primer sequences are listed in Supplemental Data Set 1.

Accession Numbers

Sequence data from this article can be found in the GenBank/EMBL datalibraries under the following accession numbers:ACT2 (AT3G18780);CHL(AT2G30370); CLL1 (AT3G22820); CLL2 (AT4G14723); cryptochrome1(cry1; AT4G08920); cry2 (AT1G04400); FAMA (AT3G24140); GATA17(AT3G16870); GATA17L (AT4G16141); GATA19 (AT4G36620); GATA23(AT5G26930); GNC (AT5G56860); GNL/CGA1 (AT4G26150); MUTE(AT3G06120); phyA (AT1G09570); phyB (AT2G18790); PIF1 (AT2G20180);PIF3 (AT1G09530); PIF4 (AT2G43010); PIF5 (AT3G59060); PROTEINPHOSPHATASE2 (PP2; AT1G13320); STOMATA DENSITY ANDDIFFERENTIATION1 (SDD1; AT1G04110); SCREAM/ICE1 (AT3G26744);SCRM2 (AT1G12860);SPCH (AT5G53210);STOMAGEN (AT4G12970); andTMM (AT1G80080).

Supplemental Data

Supplemental Figure 1. Immunoblot analysis confirms transgene ex-pression in B-GATA overexpression lines.

Supplemental Figure 2. B-GATAs promote stomata formation incotyledons.

Supplemental Figure 3. Micrografting experiments are indicative fora cell-autonomous effect of LLM-domain B-GATA overexpression onstomata formation and greening.

Supplemental Figure 4. GUS reporter gene analysis reveals broadexpression patterns of LLM-domain B-GATAs in light-grown seedlings.

Supplemental Figure 5. LLM-domain B-class GATAs function down-stream from SCREAM and SCREAM2.

Supplemental Figure 6. Expression of LLM-domain B-GATAs is lightregulated.

Supplemental Figure 7. The promoting effects of B-GATAs on stomataformation are light dependent.

Supplemental Data Set 1. Primers used in this study.

ACKNOWLEDGMENTS

We thank Keiko Torii (Seattle, WA) and Lieven De Veylder (Ghent, Belgium)for helpful comments on the work and the manuscript. This work wassupported by a grant-in-aid from the Deutsche ForschungsgemeinschafttoC.S. (SCHW751/9-1). C.S. thanks TheNirit andMichael Shaoul Fund forVisiting Scholars and Fellows for financial support and Eilon Shani andDaniel Chamovitz (Department ofMolecularBiologyandEcologyofPlants)for hosting C.S. at Tel Aviv University during manuscript preparation. C.K.and C.S. thank Dominik Hege (Technische Universität München) for pre-paring relevant DNA constructs and transgenic lines.

AUTHOR CONTRIBUTIONS

C.K., Q.L.R., and C.S. designed and performed research. J.D. performedresearch. R.R., Q.L.R., and E.B. provided important research materialor data. C.K., E.B., and C.S. analyzed data. C.K. and C.S. wrote themanuscript.

ReceivedSeptember8,2015; revisedFebruary5, 2016;acceptedFebruary22, 2016; published February 25, 2016.

REFERENCES

Abrash, E.B., and Bergmann, D.C. (2010). Regional specification ofstomatal production by the putative ligand CHALLAH. Development137: 447–455.

Abrash, E.B., Davies, K.A., and Bergmann, D.C. (2011). Generationof signaling specificity in Arabidopsis by spatially restricted buff-ering of ligand-receptor interactions. Plant Cell 23: 2864–2879.

Aoyama, T., and Chua, N.H. (1997). A glucocorticoid-mediatedtranscriptional induction system in transgenic plants. Plant J. 11:605–612.

Behringer, C., Bastakis, E., Ranftl, Q.L., Mayer, K.F., andSchwechheimer, C. (2014). Functional diversification within thefamily of B-GATA transcription factors through the leucine-leucine-methionine domain. Plant Physiol. 166: 293–305.

Behringer, C., and Schwechheimer, C. (2015). B-GATA transcriptionfactors - insights into their structure, regulation, and role in plantdevelopment. Front. Plant Sci. 6: 90.

Berger, D., and Altmann, T. (2000). A subtilisin-like serine proteaseinvolved in the regulation of stomatal density and distribution inArabidopsis thaliana. Genes Dev. 14: 1119–1131.

Berger, F., Linstead, P., Dolan, L., and Haseloff, J. (1998). Stomatapatterning on the hypocotyl of Arabidopsis thaliana is controlled bygenes involved in the control of root epidermis patterning. Dev. Biol.194: 226–234.

Bi, Y.M., Zhang, Y., Signorelli, T., Zhao, R., Zhu, T., and Rothstein,S. (2005). Genetic analysis of Arabidopsis GATA transcriptionfactor gene family reveals a nitrate-inducible member importantfor chlorophyll synthesis and glucose sensitivity. Plant J. 44: 680–692.

658 The Plant Cell

Casson, S.A., Franklin, K.A., Gray, J.E., Grierson, C.S., Whitelam,G.C., and Hetherington, A.M. (2009). phytochrome B and PIF4regulate stomatal development in response to light quantity. Curr.Biol. 19: 229–234.

Casson, S.A., and Hetherington, A.M. (2010). Environmental regula-tion of stomatal development. Curr. Opin. Plant Biol. 13: 90–95.

Casson, S.A., and Hetherington, A.M. (2012). GSK3-like kinasesintegrate brassinosteroid signaling and stomatal development. Sci.Signal. 5: pe30.

Casson, S.A., and Hetherington, A.M. (2014). phytochrome B Is re-quired for light-mediated systemic control of stomatal development.Curr. Biol. 24: 1216–1221.

Chiang, Y.H., Zubo, Y.O., Tapken, W., Kim, H.J., Lavanway, A.M.,Howard, L., Pilon, M., Kieber, J.J., and Schaller, G.E. (2012).Functional characterization of the GATA transcription factors GNCand CGA1 reveals their key role in chloroplast development, growth,and division in Arabidopsis. Plant Physiol. 160: 332–348.

Clough, S.J., and Bent, A.F. (1998). Floral dip: a simplified method forAgrobacterium-mediated transformation of Arabidopsis thaliana.Plant J. 16: 735–743.

De Rybel, B., et al. (2010). A novel aux/IAA28 signaling cascade ac-tivates GATA23-dependent specification of lateral root founder cellidentity. Curr. Biol. 20: 1697–1706.

De Veylder, L., Beeckman, T., Beemster, G.T., de Almeida Engler,J., Ormenese, S., Maes, S., Naudts, M., Van Der Schueren, E.,Jacqmard, A., Engler, G., and Inzé, D. (2002). Control of pro-liferation, endoreduplication and differentiation by the ArabidopsisE2Fa-DPa transcription factor. EMBO J. 21: 1360–1368.

Dow, G.J., and Bergmann, D.C. (2014). Patterning and processes:how stomatal development defines physiological potential. Curr.Opin. Plant Biol. 21: 67–74.

Gudesblat, G.E., Schneider-Pizon, J., Betti, C., Mayerhofer, J.,Vanhoutte, I., van Dongen, W., Boeren, S., Zhiponova, M., deVries, S., Jonak, C., and Russinova, E. (2012). SPEECHLESS in-tegrates brassinosteroid and stomata signalling pathways. Nat. CellBiol. 14: 548–554.

Hara, K., Kajita, R., Torii, K.U., Bergmann, D.C., and Kakimoto, T.(2007). The secretory peptide gene EPF1 enforces the stomatalone-cell-spacing rule. Genes Dev. 21: 1720–1725.

Hara, K., Yokoo, T., Kajita, R., Onishi, T., Yahata, S., Peterson,K.M., Torii, K.U., and Kakimoto, T. (2009). Epidermal celldensity is autoregulated via a secretory peptide, EPIDERMALPATTERNING FACTOR 2 in Arabidopsis leaves. Plant Cell Phys-iol. 50: 1019–1031.

Hudson, D., Guevara, D., Yaish, M.W., Hannam, C., Long, N.,Clarke, J.D., Bi, Y.M., and Rothstein, S.J. (2011). GNC and CGA1modulate chlorophyll biosynthesis and glutamate synthase (GLU1/Fd-GOGAT) expression in Arabidopsis. PLoS One 6: e26765.

Hunt, L., and Gray, J.E. (2009). The signaling peptide EPF2 controlsasymmetric cell divisions during stomatal development. Curr. Biol.19: 864–869.

Kanaoka, M.M., Pillitteri, L.J., Fujii, H., Yoshida, Y., Bogenschutz,N.L., Takabayashi, J., Zhu, J.K., and Torii, K.U. (2008). SCREAM/ICE1 and SCREAM2 specify three cell-state transitional stepsleading to Arabidopsis stomatal differentiation. Plant Cell 20: 1775–1785.

Kaufmann, K., Wellmer, F., Muiño, J.M., Ferrier, T., Wuest, S.E.,Kumar, V., Serrano-Mislata, A., Madueño, F., Krajewski, P.,Meyerowitz, E.M., Angenent, G.C., and Riechmann, J.L. (2010).Orchestration of floral initiation by APETALA1. Science 328: 85–89.

Kim, T.W., Michniewicz, M., Bergmann, D.C., and Wang, Z.Y. (2012).Brassinosteroid regulates stomatal development by GSK3-mediatedinhibition of a MAPK pathway. Nature 482: 419–422.

Kono, A., Umeda-Hara, C., Adachi, S., Nagata, N., Konomi, M.,Nakagawa, T., Uchimiya, H., and Umeda, M. (2007). The Arabi-dopsis D-type cyclin CYCD4 controls cell division in the stomatallineage of the hypocotyl epidermis. Plant Cell 19: 1265–1277.

Lau, O.S., Davies, K.A., Chang, J., Adrian, J., Rowe, M.H.,Ballenger, C.E., and Bergmann, D.C. (2014). Direct roles ofSPEECHLESS in the specification of stomatal self-renewing cells.Science 345: 1605–1609.

Lee, J.S., Hnilova, M., Maes, M., Lin, Y.C., Putarjunan, A., Han,S.K., Avila, J., and Torii, K.U. (2015). Competitive binding of antag-onistic peptides fine-tunes stomatal patterning. Nature 522: 439–443.

Leivar, P., and Monte, E. (2014). PIFs: systems integrators in plantdevelopment. Plant Cell 26: 56–78.

Leivar, P., Monte, E., Oka, Y., Liu, T., Carle, C., Castillon, A., Huq,E., and Quail, P.H. (2008). Multiple phytochrome-interacting bHLHtranscription factors repress premature seedling photomorpho-genesis in darkness. Curr. Biol. 18: 1815–1823.

Li, R., Li, Y., Kristiansen, K., and Wang, J. (2008). SOAP: shortoligonucleotide alignment program. Bioinformatics 24: 713–714.

Lloyd, A.M., Schena, M., Walbot, V., and Davis, R.W. (1994).Epidermal cell fate determination in Arabidopsis: patterns definedby a steroid-inducible regulator. Science 266: 436–439.

MacAlister, C.A., Ohashi-Ito, K., and Bergmann, D.C. (2007).Transcription factor control of asymmetric cell divisions that es-tablish the stomatal lineage. Nature 445: 537–540.

Martínez-García, J.F., Huq, E., and Quail, P.H. (2000). Direct tar-geting of light signals to a promoter element-bound transcriptionfactor. Science 288: 859–863.

Mockler, T.C., Guo, H., Yang, H., Duong, H., and Lin, C. (1999).Antagonistic actions of Arabidopsis cryptochromes and phytochromeB in the regulation of floral induction. Development 126: 2073–2082.

Muiño, J.M., Kaufmann, K., van Ham, R.C., Angenent, G.C., andKrajewski, P. (2011). ChIP-seq analysis in R (CSAR): An R packagefor the statistical detection of protein-bound genomic regions. PlantMethods 7: 11.

Nadeau, J.A., and Sack, F.D. (2002). Control of stomatal distributionon the Arabidopsis leaf surface. Science 296: 1697–1700.

Naito, T., Kiba, T., Koizumi, N., Yamashino, T., and Mizuno, T.(2007). Characterization of a unique GATA family gene that re-sponds to both light and cytokinin in Arabidopsis thaliana. Biosci.Biotechnol. Biochem. 71: 1557–1560.

Ohashi-Ito, K., and Bergmann, D.C. (2006). Arabidopsis FAMAcontrols the final proliferation/differentiation switch during stomataldevelopment. Plant Cell 18: 2493–2505.

Pillitteri, L.J., Bogenschutz, N.L., and Torii, K.U. (2008). The bHLHprotein, MUTE, controls differentiation of stomata and the hyda-thode pore in Arabidopsis. Plant Cell Physiol. 49: 934–943.

Pillitteri, L.J., Sloan, D.B., Bogenschutz, N.L., and Torii, K.U.(2007). Termination of asymmetric cell division and differentiationof stomata. Nature 445: 501–505.

Pillitteri, L.J., and Torii, K.U. (2012). Mechanisms of stomatal de-velopment. Annu. Rev. Plant Biol. 63: 591–614.

Ranftl, Q.L., Bastakis, E., Klermund, C., and Schwechheimer, C.(2016). LLM-domain containing B-GATA factors control differentaspects of cytokinin-regulated development in Arabidopsis thaliana.Plant Physiol. http://dx.doi.org/10.1104/pp.15.01556.

Reed, J.W., Nagpal, P., Poole, D.S., Furuya, M., and Chory, J.(1993). Mutations in the gene for the red/far-red light receptorphytochrome B alter cell elongation and physiological responsesthroughout Arabidopsis development. Plant Cell 5: 147–157.

Reyes, J.C., Muro-Pastor, M.I., and Florencio, F.J. (2004). TheGATA family of transcription factors in Arabidopsis and rice. PlantPhysiol. 134: 1718–1732.

LLM-Domain B-GATAs Promote Stomata Formation 659

Richter, R., Bastakis, E., and Schwechheimer, C. (2013a). Cross-repressive interactions between SOC1 and the GATAs GNC andGNL/CGA1 in the control of greening, cold tolerance, and floweringtime in Arabidopsis. Plant Physiol. 162: 1992–2004.

Richter, R., Behringer, C., Müller, I.K., and Schwechheimer, C.(2010). The GATA-type transcription factors GNC and GNL/CGA1repress gibberellin signaling downstream from DELLA proteins andPHYTOCHROME-INTERACTING FACTORS. Genes Dev. 24: 2093–2104.

Richter, R., Behringer, C., Zourelidou, M., and Schwechheimer,C. (2013b). Convergence of auxin and gibberellin signaling onthe regulation of the GATA transcription factors GNC and GNL inArabidopsis thaliana. Proc. Natl. Acad. Sci. USA 110: 13192–13197.

Shimada, T.L., Shimada, T., and Hara-Nishimura, I. (2010). A rapidand non-destructive screenable marker, FAST, for identifyingtransformed seeds of Arabidopsis thaliana. Plant J. 61: 519–528.

Shpak, E.D., McAbee, J.M., Pillitteri, L.J., and Torii, K.U. (2005).Stomatal patterning and differentiation by synergistic interactions ofreceptor kinases. Science 309: 290–293.

Sugano, S.S., Shimada, T., Imai, Y., Okawa, K., Tamai, A., Mori, M.,and Hara-Nishimura, I. (2010). Stomagen positively regulatesstomatal density in Arabidopsis. Nature 463: 241–244.

Wang, M., Yang, K., and Le, J. (2015). Organ-specific effects ofbrassinosteroids on stomatal production coordinate with the actionof Too Many Mouths. J. Integr. Plant Biol. 57: 247–255.

Wengier, D.L., and Bergmann, D.C. (2012). On fate and flexibility instomatal development. Cold Spring Harb. Symp. Quant. Biol. 77: 53–62.

Yang, M., and Sack, F.D. (1995). The too many mouths and four lipsmutations affect stomatal production in Arabidopsis. Plant Cell 7:2227–2239.

Zhang, X., Zhou, Y., Ding, L., Wu, Z., Liu, R., and Meyerowitz, E.M.(2013). Transcription repressor HANABA TARANU controls flowerdevelopment by integrating the actions of multiple hormones, floralorgan specification genes, and GATA3 family genes in Arabidopsis.Plant Cell 25: 83–101.

Zhao, Y., Medrano, L., Ohashi, K., Fletcher, J.C., Yu, H., Sakai, H.,and Meyerowitz, E.M. (2004). HANABA TARANU is a GATA tran-scription factor that regulates shoot apical meristem and flowerdevelopment in Arabidopsis. Plant Cell 16: 2586–2600.

660 The Plant Cell

DOI 10.1105/tpc.15.00783; originally published online February 25, 2016; 2016;28;646-660Plant Cell

SchwechheimerCarina Klermund, Quirin L. Ranftl, Julia Diener, Emmanouil Bastakis, René Richter and Claus

HypocotylsArabidopsis thalianaLight Signaling Pathways in LLM-Domain B-GATA Transcription Factors Promote Stomatal Development Downstream of

 This information is current as of January 12, 2020

 

Supplemental Data /content/suppl/2016/02/23/tpc.15.00783.DC1.html

References /content/28/3/646.full.html#ref-list-1

This article cites 55 articles, 28 of which can be accessed free at:

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists