The Arabidopsis GRF-INTERACTING FACTOR Gene Family ... · expressors contained more cells (Kim and...

The Arabidopsis GRF-INTERACTING FACTOR GeneFamily Performs an Overlapping Function inDetermining Organ Size as Well as MultipleDevelopmental Properties1[C][W][OA]

Byung Ha Lee, Jae-Heung Ko, Sangman Lee, Yi Lee, Jae-Hong Pak, and Jeong Hoe Kim*

Department of Biology (B.H.L., J.-H.P., J.H.K.) and School of Applied Bioscience (S.L.), Kyungpook NationalUniversity, Daegu 702–701, Korea; Department of Forestry, Michigan State University, East Lansing, Michigan48824 (J.-H.K.); and Department of Industrial Plant Science and Technology, Chungbuk National University,Cheongju 361–763, Korea (Y.L.)

Previously, the GRF-INTERACTING FACTOR1 (GIF1)/ANGUSTIFOLIA3 (AN3) transcription coactivator gene, a member of asmall gene family comprising three genes, was characterized as a positive regulator of cell proliferation in lateral organs, suchas leaves and flowers, of Arabidopsis (Arabidopsis thaliana). As yet, it remains unclear how GIF1/AN3 affects the cellproliferation process. In this study, we demonstrate that the other members of the GIF gene family, GIF2 and GIF3, are alsorequired for cell proliferation and lateral organ growth, as gif1, gif2, and gif3 mutations cause a synergistic reduction in cellnumbers, leading to small lateral organs. Furthermore, GIF1, GIF2, and GIF3 overexpression complemented a cell proliferationdefect of the gif1 mutant and significantly increased lateral organ growth of wild-type plants as well, indicating that membersof the GIF gene family are functionally redundant. Kinematic analysis on leaf growth revealed that the gif triple mutant as wellas other strong gif mutants developed leaf primordia with fewer cells, which was due to the low rate of cell proliferation,eventually resulting in earlier exit from the proliferative phase of organ growth. The low proliferative activity of primordialleaves was accompanied by decreased expression of cell cycle-regulating genes, indicating that GIF genes may act upstream ofcell cycle regulators. Analysis of gif double and triple mutants clarified a previously undescribed role of the GIF gene family: gifmutants had small vegetative shoot apical meristems, which was correlated with the development of small leaf primordia. giftriple mutants also displayed defective structures of floral organs. Taken together, our results suggest that the GIF gene familyplays important roles in the control of cell proliferation via cell cycle regulation and in other developmental properties that areassociated with shoot apical meristem function.

The size and shape of lateral organs, such as leavesand flowers, are under the control of developmentalgenetic programs exhibiting species-specific charac-teristics. In general, lateral organs from plant species ofdifferent size differ in cell number rather than cell size,as exemplified in the case where big petals of Brassicanapus have the same size of cells as small petals ofArabidopsis (Arabidopsis thaliana) do (Mizukami andFischer, 2000; Mizukami, 2001). Thus, cell number is aprimary determinant of organ size of a species, al-though concomitant and/or subsequent cell expan-

sion amplifies and modifies the effect of cell number,contributing to the final size and shape of lateralorgans (Tsukaya, 2005; Cho et al., 2007).

Many genetic approaches have been employed tounderstand the control of cell proliferation and its rolein organogenesis, and these have identified a numberof genes that play important roles in the process. Someof these genes, such as AINTEGUMENTA (ANT),KLUH (KLU), and STRUWWELPETER (SWP), playpositive roles (i.e. their loss-of-function mutants de-veloped small lateral organs due to a reduction in cellnumbers; Mizukami and Fischer, 2000; Autran et al.,2002; Anastasiou et al., 2007). On the other hand,overexpressors of those genes have bigger organsresulting from more cells, although the SWP over-expressors resulted in smaller organs despite therebeing more cells, because of smaller cell size. It hasbeen proposed that those genes positively regulate cellproliferation and organ growth by controlling theduration of cell proliferation. ANT, KLU, and SWPencode the APETALA2 domain transcription factor,the P450 cytochrome oxidase CYP78A5, and theMed150/RGR1-like subunits of the Mediator tran-scriptional regulatory complex, respectively. There

1 This work was supported by the Korea Research Foundation(grant nos. KRF–2006–331–C00264 and KRF–2008–314–C00345).

* Corresponding author; e-mail [email protected] author responsible for distribution of materials integral to the

findings presented in this article in accordance with the policydescribed in the Instructions for Authors (www.plantphysiol.org) is:Jeong Hoe Kim ([email protected]).

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

[W] The online version of this article contains Web-only data.[OA] Open Access articles can be viewed online without a sub-

scription.www.plantphysiol.org/cgi/doi/10.1104/pp.109.141838

Plant Physiology�, October 2009, Vol. 151, pp. 655–668, www.plantphysiol.org � 2009 American Society of Plant Biologists 655 www.plantphysiol.orgon February 1, 2020 - Published by Downloaded from

Copyright © 2009 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

are also negative regulators of cell proliferation andorgan growth, such as PEAPOD and BIG BROTHERgenes, which encode putative DNA-binding proteinsand an E3 ubiquitin ligase, respectively (Disch et al.,2006; White, 2006). Those negative regulators affect theduration of cell proliferation in lateral organs as well.

We have previously uncovered a novel classof transcriptional factors, GROWTH-REGULATINGFACTOR (GRF), comprising nine members in Arabi-dopsis (Kim et al., 2003). The grf triple loss-of-functionmutant of GRF1 through GRF3 developed narrow andsmall leaves, whereas overexpression of GRF1 andGRF2 resulted in large leaves. Reexamination of cel-lular parameters revealed that the grf triple mutantproduced fewer palisade cells, whereas the GRF over-expressors contained more cells (Kim and Kende,2004; Kim and Lee, 2006), indicating that GRF proteinsact as positive regulators of cell proliferation. In searchof partner proteins for GRF1, we found another genefamily, GRF-INTERACTING FACTOR (GIF), whichcomprises three members (Kim and Kende, 2004).GIF proteins are thought to be functional homologsof the human SYT transcriptional coactivator. Like thegrf triple mutant, the gif1 single mutant showed nar-row and small leaves as well as flower petals, whichwere caused by a reduction in cell numbers, indicatingthat GRF and GIF1 proteins form a functional complexinvolved in regulating cell proliferation and, thus,growth and shape of lateral organs (Kim and Kende,2004). Horiguchi et al. (2005) also confirmed the resultby demonstrating that angustifolia3 (an3) mutations,which occurred at theGIF1 gene, and the grf5mutationcaused a reduction in cell numbers, leading to smalland narrow lateral organs, while overexpression ofGIF1/AN3 increased leaf size with normal shape.

Meanwhile, it still remains to be elucidated by whatmechanism the GIF1/AN3 gene regulates cell prolif-eration and what biological roles the other two mem-bers, GIF2 and GIF3, play in Arabidopsis growth anddevelopment. In this article, we show that all of theGIF genes act as positive regulators of cell proliferationof lateral organs in a functionally redundant manner.Our results indicate that GIF genes may affect theduration of cell proliferation by modulating the ex-pression level of cell cycle regulators. We also presentexperimental evidence that GIF genes are required forother developmental processes, such as plastochronand flower structure.

RESULTS

In Silico Expression Profiles of GIF Genes and Isolationof T-DNA Insertional Mutants

Using RNA gel-blot analysis, we have previouslyshown that the GIF family genes share an overlappingexpression pattern and that GIF1/AN3 mRNAs aremost abundant compared with GIF2 and GIF3 (Kimand Kende, 2004). To compare the tissue-specific ex-

pression pattern of GIF genes in more detail, wecompiled the microarray data from the AtGenExpressexpression atlas (Schmid et al., 2005), which containedexpression profiles of all threeGIF genes in 45 differentorgans or developmental stages (Supplemental Fig.S1). The overall expression patterns of GIF genes aresimilar, suggesting that those genes may be function-ally redundant. It should be noted, however, that themaximum level of GIF1/AN3 mRNAwas about threetimes higher than that of GIF2 and that of GIF2 wasabout two times higher than that of GIF3. All of theGIF genes show a distinctively high expression inthe shoot apex containing lateral organ primordia andthe apical meristem. The in silico expression patternand strength are consistent with the previous RNAgel-blot results (Kim and Kende, 2004) and suggestthat their biological function may be relevant espe-cially to growth and development of the shoot apicalmeristem (SAM) and/or lateral organs developingfrom the flanking region of the SAM.

A phylogenetic analysis suggested that GIF2 andGIF3 genes might result from a duplication event of acommon ancestral gene after diversification fromGIF1/AN3 (Kim and Kende, 2004). This notion isalso supported by the fact that GIF2 and GIF3 havethe same gene structure, with four introns, whereasGIF1/AN3 lacks the last one (Fig. 1; Kim and Kende,2004). To investigate whether or not GIF2 and GIF3proteins have similar biological function, we isolatedhomozygous T-DNA insertional mutants, gif2 andgif3, from the Syngenta (SAIL_328_A03) and Salk(SALK_072950) lines, respectively (Sessions et al.,2002; Alonso et al., 2003). We found that the gif2 andgif3 mutations resulted from the T-DNA insertions at69 bp from the start of the second exon and 45 bp

Figure 1. Gene structure and representation of T-DNA insertionalmutations. T-DNA integration sites are marked with inverse triangles.Arrows, Orientation of the T-DNA left border (LB); black boxes, exons;white boxes, introns; solid lines, 5# untranslated region; arrowheads,primer sites for RT-PCR analysis at right. The ruler is a scale of bp. Thegel images show RT-PCR determination of GIF mRNA content in wild-type Columbia (Col) and gif mutant plants. Amplification of ACTIN8(ACT8) cDNA was used as a control. The single and double asterisksindicate DNA fragments derived from spliced and unspliced GIF3mRNA species, respectively.

Lee et al.

656 Plant Physiol. Vol. 151, 2009 www.plantphysiol.orgon February 1, 2020 - Published by Downloaded from



upstream of the translational start site, respectively(Fig. 1). Reverse transcription (RT)-PCR analysis usingcDNAs prepared from DNase-treated mRNAs re-vealed that the gif2 mutant completely lacked GIF2mRNA, whereas the gif3 mutant showed a significantbut much lower level of GIF3mRNA than that of wild-type plants (Fig. 1, bottom). The T-DNA insertionalsites and RT-PCR analyses indicate that gif2 and gif3are likely to be a null and a hypomorphic mutant,respectively. Interestingly, wild-type plants exhibitedunspliced species of GIF3 mRNA, whose level wasfurther elevated in the gif3mutant, suggesting that theprimary transcripts of the GIF3 gene were not subjectto a complete splicing and that the T-DNA insertioninto the 5# untranslated region of GIF3 interfered withnormal splicing.

Growth and Developmental Phenotypes ofLateral Organs

As described previously (Kim and Kende, 2004), thegif1 null mutant caused by T-DNA insertion into the

third exon had small and narrow lateral organs, suchas leaves, cotyledons, and flowers (Fig. 2, A–C). Quan-tification of the first pair of primary leaves at theirmaturity revealed that leaf-blade area of gif1 wasdecreased by 52% over the wild type, which was dueto reduction in both blade length and width (Fig. 3A,left). Petiole length was also decreased as much asblade length was. The areas of cotyledons and petalswere reduced as well, by approximately 22% and 40%,respectively (Fig. 3, B and C). Reduction in width wasmore prominent than in length, resulting in narrowmorphology and, thus, increasing the index value (theratio of length to width). To the contrary, dimensionaland morphological phenotypes of gif2 and gif3 singlemutants were not distinctive, although careful obser-vation andmeasurement of leaf area revealed aminutechange in the gif2 single mutant, which was statisti-cally significant by Student’s t test (P , 0.05; Figs. 2and 3). To test synergistic interaction between gifmutations, we constructed double and triple mutantsby genetic crossing and found that the gif1 gif2 gif3(hereafter gif1/2/3) triple mutant developed tiny and

Figure 2. Phenotypes of gifmutants. A, Wild-type Columbia (Col) and gif plants (top) and rosette leaves (bottom) from the first tofourth leaves at day 20. B, Seedlings at day 8. C, Mature flowers. Note that gif1/2/3 has only two petals and split carpels(arrowheads). D, Adaxial palisade cells of mature leaves. E, Adaxial palisade cells of mature cotyledons. F, Subepidermal cells atthe abaxial side of mature petals. Scale bars = 25 mm.

Lateral Organ Growth by GIF Genes

Plant Physiol. Vol. 151, 2009 657 www.plantphysiol.orgon February 1, 2020 - Published by Downloaded from



narrow organs (Fig. 2, A–C, last column). Leaf-bladearea of the triple mutant was only about 10% of thewild type; the areas of cotyledons and petals were 42%and 26%, respectively (Fig. 3, left panels). Phenotypicstrength of the gif1/2 double mutant was intermediatebetween gif1 and the triple mutants. gif1/3 and gif2/3double mutants, however, developed only slightlysmaller organs than their stronger parental lines(Figs. 2 and 3). These results clearly demonstrate thatgif mutations act synergistically to affect the growthand development of lateral organs, and the relativecontribution of each mutation to the phenotype wasgif1 . gif2 . gif3, indicating that GIF genes have anoverlapping function in regulating the size and shapeof lateral organs. The root growth of gif mutants wasnot affected at all (data not shown).

Unexpectedly, we found that the triple mutantdisplayed aberrant structures of floral organs, includ-ing split carpels and reduction in numbers of petalsand anthers, while the wild type, single mutants, andgif2/3 showed normal floral organization and organnumbers; gif1/2 and gif1/3 double mutants displayed

such floral defects but only in the terminal two flowers(Fig. 2C, last column; details will be described else-where). This observation confirmed the contribution ofsingle mutations once more. In addition, gif1/2 andgif1/2/3 mutants were completely sterile, whereasother mutants set seeds, the number of which wasvariable depending on their phenotypic severity (Ta-ble I). Because of their infertility, gif1/2 and gif1/2/3mutant lines were maintained in the heterozygous gif1mutation with other homozygous ones. The sterilityproblem of gif1/2 seemed to stem from an abnormalfunctionality of female reproductive organs ratherthan from pollen. In other words, the gif1/2 doublemutant produced pollen that was fully functional onwild-type carpels, producing as many seeds as wild-type pollen did, whereas wild-type pollen did notproduce even a single seed on gif1/2 carpels (Supple-mental Table S1). The gif1/2/3 triple mutant, besides thestructural abnormality of flower organs, produced nopollen at all. These results indicate that GIF genes arealso required for the maintenance of structural integ-rity and reproductive function of flower organs.

Figure 3. Quantification of mutant phenotypes at the organ and cell level. A, Phenotype of the first two leaves at day 25. Leaf sizeand shape are shown, including index values (left), number and size of adaxial palisade cells (middle), and number of trichomes(right). B, Phenotype of cotyledons at day 15. Cotyledon size and shape are shown, including index values (left) and number andsize of adaxial palisade cells (right). C, Phenotype of mature petals. Petal size and shape are shown, including index values (left)and number and size of subepidermal cells at the abaxial side (right). The numerical marks on the x axis denote gif single andmultiple mutants. Col, Columbia wild type. n = 10 for determination of organ-level parameters; n = 200 from 10 different organsfor cellular parameters. Error bars indicate SE.

Lee et al.




We determined cell numbers and sizes of lateralorgans by observing subepidermal cells under No-marski optics. Numbers of adaxial palisade cells in thefirst two mature leaves of gif1/2/3were only 27% in thelongitudinal axis and 15% in the maximum-width axisover wild-type plants (Fig. 3A, middle). The totalnumber of palisade cells of the triple mutant, calcu-lated by dividing leaf area by cell area, was only 4% ofthe wild type. The number of trichomes on the firsttwo leaves of gif1/2/3 was also reduced to 8% of wild-type plants, which is indicative of a great reduction inthe number of epidermal cells as well (Fig. 3A, right).Reduction in those cell numbers of other single anddouble mutants corresponded to changes in the organ-level dimensional parameters (Fig. 3A, compare leftand middle panels). To the contrary, cell area of thetriple mutant increased by 2.6-fold over the wild type(Figs. 2D and 3A), which probably resulted from a“compensatory” effect due to the drastic reduction incell number (Tsukaya, 2003). It should be noted, how-ever, that the compensatory increase in cell area wasonly partially able to make up for the reduction in leafsize, and there was no change in cell shape (Fig. 2). Thegif1/2 double mutant exerted its compensatory effect toan extent comparable with the triple mutant, and thatof the gif1 single and gif1/3 double mutants was alsoapparent. Other single and double mutants, however,did not show a significant change in cell area. Theseresults indicate that the compensation occurs in directproportion to the change in cell numbers. This is alsotrue for cotyledons and flowers: more gif mutations

resulted in smaller organs with fewer but bigger cells(Fig. 3, B and C). Taken together, our data indicate thatGIF2 and GIF3 as well as GIF1 are involved in thepositive regulation of cell proliferation of lateral or-gans in a functionally redundant manner.

Kinematic Analysis of Leaf Area and Cellular Parametersduring Leaf Growth

To obtain more information about the effect of gifmutations on cell proliferation, we performed kine-matic analysis on leaf growth. The first pairs of leaveswere harvested at the indicated days from 4-d-oldseedlings. Because gif1/2 double and gif1/2/3 triplehomozygous mutants showed distinctively small andnarrow cotyledons (Fig. 2B), we were able to differen-tiate homozygous individuals in the segregating pop-ulations of gif1/+ gif2/gif2 and gif1/+ gif2/gif2 gif3/gif3genotypes.

Leaf area continued to increase over time: slowduring the early stage, then at an accelerated rate, andlastly, leaf expansion slowed toward the final size ofmature leaves (Fig. 4A, left). Leaf expansion of wild-type plants was almost completed at day 20, whereasthat of the gif1 single mutant and the gif1-harboringdouble mutants ceased at 17 and 13 d, respectively.The gif1/2/3 triple mutant showed only a minute leafexpansion during the whole growth period. Plotted ona natural log scale in a magnified version, the leaf areadata from days 4 through 7 revealed that gif1/2/3 leaveswere markedly smaller, by 5.5-fold, than wild-type

Table I. Plastochronic and other developmental phenotypes of gif mutants

Values shown are means 6 SE; n = 10.

Leaf OrderGenotype

Columbia gif1 gif2 gif3 gif1/2 gif2/3 gif1/3 gif1/2/3

Day when leaf attains 1 mm in lengtha

1, 2 8.6 6 0.1 7.8 6 0.1 [0.8]b 8.3 6 0.1 [0.3] 8.5 6 0.1 [0.1] 8.2 6 0.1 [0.4] 8.4 6 0.1 [0.2] 8.1 6 0.1 [0.5] 10.4 6 0.2 [21.9]

(3.9)c (3.0) (3.6) (3.8) (2.7) (3.4) (2.8) (0.9)

3 12.5 6 0.1 10.8 6 0.1 [1.7] 11.9 6 0.1 [0.6] 12.3 6 0.1 [0.1] 10.9 6 0.1 [1.5] 11.8 6 0.1 [0.6] 10.9 6 0.2 [1.6] 11.3 6 0.3 [1.2]

(1.3) (0.8) (1.3) (1.2) (0.8) (1.4) (1.0) (1.1)

4 13.8 6 0.1 11.6 6 0.2 [2.2] 13.2 6 0.1 [0.6] 13.5 6 0.1 [0.3] 11.7 6 0.1 [2.0] 13.2 6 0.2 [0.7] 11.9 6 0.1 [1.9] 12.4 6 0.2 [1.5]

Day when leaf attains 1 mm in lengthd

1, 2 15.8 6 0.1 12.7 6 0.0 [3.2] 13.5 6 0.0 [2.3] 13.9 6 0.3 [2.0] 12.7 6 0.0 [3.2] 13.4 6 0.2 [2.4] 12.8 6 0.1 [3.0] 16.1 6 0.1 [20.2]

(4.6) (3.9) (4.4) (4.6) (4.0) (4.7) (4.1) (1.5)

3 20.5 6 0.2 16.6 6 0.1 [3.9] 17.9 6 0.1 [2.5] 18.5 6 0.1 [2.0] 16.6 6 0.1 [3.8] 18.1 6 0.1 [2.3] 16.9 6 0.1 [3.6] 17.6 6 0.1 [2.9]

(1.4) (1.4) (1.5) (1.7) (1.6) (1.5) (1.5) (0.9)

4 21.8 6 0.1 18.0 6 0.1 [3.8] 9.5 6 0.0 [2.4] 20.2 6 0.2 [1.6] 18.2 6 0.2 [3.6] 19.7 6 0.2 [2.2] 18.4 6 0.3 [3.5] 18.5 6 0.2 [3.4]

No. of rosette leaves when plants have the first open flowersa

10.7 6 0.1 11.3 6 0.4 10.3 6 0.5 11.1 6 0.5 10.3 6 0.3 10.0 6 0.3 11.0 6 0.5 10.0 6 0.3

Bolting time: day when the first flower bud develops to stage 12a

29.5 6 1.6 27.3 6 1.3 28.7 6 1.0 29.5 6 1.2 26.0 6 0.9 24.0 6 1.5 24.8 6 1.5 21.4 6 1.0

No. of seeds per siliquea

42.7 6 1.4 19.6 6 1.1 35.1 6 1.0 40.5 6 2.6 0.0 6 0.0 28.8 6 1.7 8.0 6 1.4 0.0 6 0.0

Primary stem length (cm)a

24.3 6 2.5 28.9 6 3.0 24.3 6 3.2 22.9 6 1.5 30.2 6 2.3 24.1 6 2.9 33.4 6 2.5 14.5 6 2.4

aPlants were grown under long days (18 h of light/6 h of darkness). bNumbers in brackets are differences in days between the wild type and the mu-

tants. cNumbers in parentheses are plastochron lengths between successive leaves. dPlants were grown under short days (8 h of light/16 h of darkness).





leaves already at day 6 and afterward (Fig. 4A, right).A clear distinction between the gif1/2 double mutantand the wild type was also recognized at day 7.Intriguingly, however, leaf area of the gif1 single mu-tant was not smaller than that of the wild type in thoseearly days; only later did it become smaller, eventuallyreaching half the size of wild-type leaves. This isconsistent with what we have observed during plantgrowth: note the similar size of the first two leaves ofthe wild type and the gif1 single mutant even at day 8in Figure 2B. We attribute this apparent discrepancy insize of young and mature gif1 leaves to the strongplastochronic properties of the gif1mutant (see below).

To compare cell proliferation patterns of the wildtype and mutants during leaf growth, we determinedthe number of adaxial palisade cells in a line alongwith a maximum-width axis of the first two leaves,and the resulting kinematic graph clearly showed thatcell proliferation ceased at day 11 in the wild type butat day 7 in the strong mutants harboring gif1 (i.e. gif1,gif1/2, gif1/3, and gif1/2/3; Fig. 4B, left). Previously,Ferjani et al. (2007) also observed that an3-4, anothermutant allele of GIF1, exhibited early cessation of thecell proliferation phase in the leaf-length axis. More-over, leaves of gif1/2 double and gif1/2/3 triple mutantshad only about half the cells of wild-type leavesalready at days 6 and 4, respectively (Fig. 4B, right).Intriguingly again, however, gif1 leaves at the earlystage between days 4 and 6 had more cells than wild-type leaves, at least partially explaining the reasonwhy the leaf size of the gif1 mutant was similar to thatof the wild type in those early days (Fig. 4B). Thistendency in gif1 was consistently observed in an3-4(Ferjani et al., 2007).

Calculation of cell proliferation rate (changes in cellnumbers between consecutive days divided by the cellnumber of the previous day) showed that cell prolif-eration in leaf was most active during the intervalbetween days 5 and 7, but maximum proliferationactivity of most gifmutants was remarkably or slightlylower than that of the wild type; especially, that oftriple mutants was close to zero (Fig. 4C). Further-more, almost no cell proliferation activity of the strongmutant lines was detected around day 8 and thereaf-ter, whereas wild-type activity remained at a substan-tial level until day 10. These data indicate that the gifmutations exert synergistic effects on both the rate andduration of cell proliferation.

Leaf cell expansion of the wild type and all of the gifmutants increased in a sigmoidal pattern throughoutgrowth (Fig. 4D, left). The compensatory enhancementof cell expansion in the gif triple mutant becameprominent from day 6, although it was not possibleto make a clear distinction between the wild type andmutants before that day (Fig. 4D, right). It should benoted, however, that the enhanced cell expansion wasnot sufficient to mask the effect of the reduced cellproliferation on determining leaf size: already at day 6and thereafter, leaf size of the triple mutant was at leastfive times smaller than that of the wild type (Fig. 4A).

Figure 4. Kinematic analysis of leaf growth. Measurements were per-formed on the first two leaves of the wild-type Columbia (Col) and gifplants grown on soil for the indicated times. Adaxial palisade cells wereanalyzed. A, Leaf blade area. B, Numbers of cells in a line along themaximum width region. C, Relative cell proliferation rate derived fromB. The rate was calculated by dividing the difference in cell numbers oftwo successive measurements by the cell numbers of the previous one.D, Area of adaxial palisade cells. The graphs at right show magnifiedversions to indicate differences in the early stages of leaf growth. Notethat the right graph in A was plotted on the natural log scale. The samenumbers of samples were analyzed as in Figure 3. Error bars indicate SE.

Lee et al.




Expression of Cell Cycle-Regulating Genes

The low rate of cell proliferation of gif mutantsprompted us to examine the expression of cell cycle-regulating genes. Total RNAwas isolated from the firsttwo leaves of 6-d-old wild-type and gif1/2/3 triplemutant plants, in which their cell proliferation oc-curred at a maximum rate, and subjected to RT-PCR.The transcript levels of two cyclin genes, CycB1;1 andCycD3;1, as well as a cyclin-dependent kinase gene,Cdc2b, were significantly reduced in the triple mutant(Fig. 5). Both CycB1;1 and Cdc2b are involved in theG2/M transition (Mironov et al., 1999; Vandepoeleet al., 2002), and CycD3;1 has been known to domi-nantly drive the G1/S transition (Dewitte et al., 2003;Menges et al., 2006). The S-phase-specific proliferatingcell nuclear antigen PCNA1 (Kosugi and Ohashi,2002; Anderson et al., 2008) was remarkably down-regulated as well. These data indicate that GIF genesmay control cell proliferation by positively regulatingcell-cycling gene expression in the primordial leaves.

Mutant Phenotypes in the SAM Tissues

Since cells in the SAM continue to be recruited intodeveloping leaf primordia (Telfer and Poethig, 1994)and since the gif mutations affect the cell number ofprimordial leaves, we investigated the effect of gifmutations in controlling the size of the SAM cell pool.Nuclei of mature embryos were stained with propi-dium iodide, and the SAM size and its cell number in alongitudinal median section were determined by con-focal microscopy. We found that the SAM size of the

gif1 mutant but not of gif2 and gif3 was smaller thanthat of the wild type (Fig. 6A). Counting the number ofnuclei revealed that the gif1 SAM was made up offewer cells in total and in its L1 layer and that asubstantial portion of gif2 mutant individuals had theSAM with fewer cells (Fig. 6, B and C). We alsodetermined the number of SAM cells in segregatingindividual embryos from the selfed siliques of gif1/+gif2/gif2 and gif1/+ gif2/gif2 gif3/gif3 genotypes. Amajority of the progeny of gif1/+ gif2/gif2 had theSAM comprising fewer cells than their parental mu-tant, resulting in a smaller SAM; most individuals ofthe F1 progeny of gif1/+ gif2/gif2 gif3/gif3 had evensmaller SAM with fewer cells than the segregatingpopulation of gif1/+ gif2/gif2. These results indicatethat the gif triple mutations synergistically reduced thesize of the vegetative SAM cell pool.

Plastochronic Properties of gif Mutants

Apart from the cell proliferation phenotype, wenoticed that gifmutations seemed to accelerate variousdevelopmental processes. To estimate the develop-mental acceleration, we measured the time whenleaves reach 1 mm in length with a stereoscope inlong-day conditions. The first two leaves of the gif1mutant attained that length at day 7.8, which was 0.8 dearlier than the wild type (Table I). The third andfourth gif1 leaves weremuch earlier than the wild-typeleaves, by 1.7 and 2.2 d, respectively. The differencebetween the wild type and gif1 was profound inthe short-day conditions (i.e. gif1 leaves attained thelength more than 3 d earlier compared with thecorresponding wild-type leaves, clearly indicatingthat the gif1 mutation accelerated the early develop-mental pace). The time intervals at which successiveleaves reached that length (plastochron) were alsosignificantly shortened in the gif1 mutant comparedwith the wild type: for instance, it took 3.9 and 4.6 d inthe respective photoperiods for the wild-type thirdleaf to attain 1 mm of length after its first two leavesdid so, but it took 3.0 and 3.9 d for the gif1 third leafunder the corresponding conditions. However, thefinal numbers of wild-type and gif1 leaves were notsignificantly different from each other, and neither wasgermination time (data not shown). Taken together,these results clearly indicate that the gif1 mutantdisplays short plastochronic length at the early devel-opmental stage as well as the hasty emergence of thefirst two leaves; furthermore, they explain why the gif1single mutant has more cells in the primordial leavesat the early stage than wild-type primordium does,despite its lower cell proliferation activity, as men-tioned above.

In contrast, the time when the double mutantsharboring the gif1 mutation, gif1/2 and gif1/3, attained1-mm-long leaves was slightly delayed or similar,compared with their parental gif1 single mutant inboth photoperiodic regimens, although they did soearlier than the wild type. Especially, the first two

Figure 5. Determination of mRNA levels of various cell-cycling genesby RT-PCR. Total RNAs were isolated from the first two leaves of wild-type Columbia (Col) and the gif1/2/3 triple mutant (gif) at day 6 andused for RT-PCR analysis. The numbers at the top indicate dilutionfactors of cDNAs derived from the RT reaction. ACTIN8 (ACT8) wasused as an amplification control.





leaves of the gif1/2/3 triple mutant were more delayedthan even the wild-type plant, by 1.9 and 0.2 d (com-pare the numbers in brackets in Table I for bothphotoperiodic regimens). It is important to realize,however, that this does not necessarily mean that no orantagonistic plastochronic synergism exists betweenthose mutations. Rather, this observation means thatleaves of gif double and triple mutants, unlike gif1leaves, were so short that the time for the attainment ofa 1-mm-long leaf could not be an appropriate param-

eter for plastochronic comparison with the wild typeor gif1. In fact, the gif1/2/3 plastochron between the firsttwo and third leaves was exceptionally shorter thanthat of gif1 (compare the numbers in parentheses inTable I), and synergistic acceleration of other develop-mental processes was evident: the gif1mutant bolted 2d earlier than the wild type did, and addition of thegif2 and gif3mutations gradually shortened the boltingtime, resulting in the gif triple mutant bolting 8 dearlier than the wild type (Table I). It is interesting that,in spite of their minute effect on leaf size, gif2/3 andgif1/3 double mutations shortened the bolting time alittle more than the gif1/2 double mutation did.

We also found that mutants containing the gif1mutation developed more axillary branches, as shownin Figure 7A, in which all of the primary shoots andleaves, except two cotyledons, were removed to ex-pose axillary branches at day 30. The triple mutantshowed a long primary branch (Fig. 7A, far right).During the vegetative stage or just before bolting, onlythe gif triple mutant was able to develop primarybranches from cotyledons and rosette leaves (i.e. nineout of 10 gif1/2/3 mutants showed a pair of rapidlygrowing axillary leaves from both cotyledons and thefirst two rosette leaves at day 17.06 0.7 and 19.16 0.7,respectively, whereas none of the other mutants aswell as the wild type did so during the same period,although gif1/2 double mutants occasionally did). Thetriple mutant produced about five more primary ax-illary branches from rosette leaves and even cotyle-dons than the wild type, and other mutants harboringgif1 mutations showed intermediate numbers (Fig.7C). To the contrary, there was no significant differencein numbers of primary branches derived from caulineleaf axils (Fig. 7D). These data indicate that the loss ofGIF genes accelerates various developmental pro-cesses, including leaf plastochron.

Finally, the gif triple mutant developed short inflo-rescence stems, although other mutant plants contain-ing the gif1 mutation tend to have a little longer stemthan the wild type (Table I), suggesting that all threeGIF genes are necessary for normal growth of theinflorescence stem.

Overexpression of GIF1, GIF2, and GIF3 in gif1 Mutantsand Wild-Type Plants

To test whether or not GIF2 and GIF3 genes will beable to complement the gif1 phenotype, gif1 mutantplants were transformed with GIF1, GIF2, and GIF3cDNAs under the control of the cauliflower mosaicvirus 35S promoter. Dozens of T1 generation plantswere obtained for each construct, and most of themshowed a wild-type-like phenotype (Fig. 8A; data notshown). Quantification data revealed that leaf dimen-sion parameters of gif1 plants overexpressing eachconstruct were very close to those of the wild type (Fig.8B). Leaf indices were also recovered to the wild-typevalue, displaying leaves with round shape rather thanthe narrow one of the gif1 mutant. All of these dimen-

Figure 6. Phenotype of the SAM tissue. A, Size of the SAM. B, Totalnumber of SAM cells. C, Number of L1 layer cells. Nuclei of matureembryos were stained with propidium iodide and visualized by con-focal microscopy. Each rank at the x axis in B means a three-cell range(i.e. cell numbers in each rank are equal to the numerical marks 6 1).Col, Columbia wild type.

Lee et al.




sional changes corresponded exactly to those in cellnumbers (Fig. 8C). We also established overexpressorlines whose gif1 mutation was segregated out aftercrossing the complementation lines to wild-type plantsand found that overexpression of each gene in thewild-type genetic background increased both leaf areaand cell numbers up to 14% (Fig. 8, D and E). Theseresults indicate that the gain-of-function effects of GIFson lateral organ growth are highly similar to eachother, stimulating cell proliferation and organ growth.

DISCUSSION

In this study, we presented, to our knowledge, thefirst loss- and gain-of-function evidence that GIF2 andGIF3 acted as positive regulators for lateral organgrowth, as did GIF1. We were also able to determinecontributions of individual GIF genes to the biologicalfunction through phenotypic analyses of a series ofmultiple gif mutants. gif triple mutants also revealednovel phenotypes that are relevant to functional ac-tivities of the shoot apical meristem, such as plasto-chronic and axillary branching properties, as well asreproductive organ function and structure.

GIF Genes Have Functional Redundancy in DeterminingOrgan Size as Well as in Regulating MultipleDevelopmental Processes

We demonstrated that GIF genes have an overlap-ping expression pattern in many different tissues andthat gif mutations act synergistically to induce dra-matic reductions in cell numbers of lateral organs,such as leaves and flowers as well as cotyledons (Figs.

1–3), resulting in tiny plants. The mutational effect oncell proliferation was less severe in cotyledons than inother lateral organs, probably because the postembry-onic cell division of cotyledons occurred for a narrowtime window (Masubelele et al., 2005). The reductionin cell numbers was accompanied by a remarkableenlargement of cell size, which only partially compen-sated for the reduced organ size. The gifmutations alsosynergistically reduced the vegetative SAM size bydecreasing cell numbers (Fig. 6). Furthermore, wehave demonstrated that overexpression of GIF1,GIF2, and GIF3 promotes cell proliferation and leafsize and that GIF2 and GIF3 proteins are functionalequivalents of GIF1 (Fig. 8). Horiguchi et al. (2005)have previously shown that overexpression of theGIF1/AN3 gene stimulates cell proliferation as well,leading to enlarged leaves by about 20%. These resultsled us to the conclusion that all of the GIF genesfunction redundantly as positive regulators of cellproliferation, thereby determining plant organ size.The overlapping expression pattern and functionalredundancy are in line with our previous biochemicalstudies showing that GIF2 proteins interact with GRF1proteins as does GIF1 (Kim and Kende, 2004), raisingthe possibility that all three GIF proteins may partic-ipate in the formation of the functional transcriptionalcomplex with GRF proteins in a combinatorial manner.However, not all GRFs may be in the complex to thesame extent, because GIF1 interacted strongly withGRF5 and GRF9 but weakly with GRF4 (Horiguchiet al., 2005).

Overexpression of GIF and GRF genes stimulatedorgan growth but to a limited extent (Fig. 8; Kim et al.,2003; Horiguchi et al., 2005). It may be due to other

Figure 7. Shoot branching phenotype. A, Branching phenotype of mature plants at 35 d. The main shoot and all of the rosetteleaves were removed to expose rosette leaf branches and cotyledons. Arrowheads indicate rosette leaf branches with flowerbuds. B, Schematic representation of Arabidopsis branching structure. C, Number of rosette leaf primary branches. D, Number ofcauline leaf primary branches. n = 10. Error bars indicate SE. Col, Columbia wild type.





factors limited in the mechanisms of organ size deter-mination, and GIFs and GRFs are very likely to belimiting factors to each other. To test this interestingpossibility, we set out to construct double overexpres-sors of GIFs and GRFs. The resultant transgenic plantsmight not only give a detailed understanding of mo-lecular mechanisms of organ growth with regard tothe transcriptional complex but also might provide abasic knowledge for the development of agriculturallyuseful traits.

Judged from the abnormal structure of flower or-gans, as mentioned briefly, and the plastochronicphenotype (Figs. 2 and 7; Table I), it seems that GIFgenes play an important role in regulating other as-pects of shoot development. It remains unknownwhether those phenotypes are in any causal relation-ship to the cell proliferation process. With respect tothe pleiotropic phenotype of gif mutants, it is interest-ing that the human homolog of the GIF proteins, SYTtranscription coactivators, also show diverse biologi-cal roles in human cells via interaction with a plethoraof proteins (Eid et al., 2000; de Bruijn et al., 2001; Katoet al., 2002; Perani et al., 2003, 2005). For example,the human SYT proteins interact with the chromatin-remodeling proteins BRM and BRG1, which are knownto act as positive or negative regulators of globalexpression (Thaete et al., 1999; Kato et al., 2002; Peraniet al., 2003). SYT also interacts with the histone acetyl-transferase p300/CBP to control cell adhesion (Eidet al., 2000). The Arabidopsis BRM gene has beenreported to be involved in numerous aspects of plantgrowth and development (Hurtado et al., 2006; Kwonet al., 2006). Han et al. (2007) demonstrated that, inArabidopsis, several genes encoding histone acetyl-transferase play a role in regulating plant growth andflowering time. We have previously shown that themolecular behavior of GIF1 proteins is similar to that ofSYT, including their existence as nuclear speckles andtranscriptional activation activities (Kim and Kende,2004). Therefore, apart from its known interactingpartner GRF, GIF proteins may have other interactingproteins to affect multiple developmental processes.

Mechanisms by Which GIF Genes Regulate CellProliferation during Organogenesis

It is conceivable that cell numbers in a plant organcould be determined in several different ways: (1) thefrequency of cell proliferation; (2) the duration of cellproliferation; and (3) the number of cells in the pri-mordial cell pool (Autran et al., 2002). Studies onmanygenes that are involved in cell proliferation duringorganogenesis suggest that most of those genes seemto control cell numbers by changing the duration of

Figure 8. Overexpression phenotype of GIF genes in gif1 (A–C) andwild-type (D and E) plants. The first pairs of 23-d-old plants were usedfor the analysis of leaf dimensional and cellular parameters (n = 10).Error bars indicate SE. Col, Columbia wild type. [See online article forcolor version of this figure.]

Lee et al.




cell proliferation rather than its rate, regardless of theirrole as positive or negative regulators. ANT and KLUaffect cell proliferation by regulating the duration,rather than the rate, of cell proliferation (Mizukamiand Fischer, 2000; Anastasiou et al., 2007). As positiveregulators, GIF genes apparently influence the dura-tion of cell proliferation as well, as the strong gifmutants ceased cell proliferation several days earlierthan the wild type (Fig. 4). However, the mechanismby which GIF genes affect the duration seems to differfrom that of ANT and KLU. First, ANT and KLU genesare involved in maintaining the meristematic compe-tence of cells; thus, their mutational effect was man-ifested primarily in the later stage of the proliferationphase rather than in young organs (Mizukami andFischer, 2000; Anastasiou et al., 2007). To the contrary,the effect of gif mutations was obvious in the primor-dial leaves, where most of the cells are actively divid-ing. This was clearly in line with the fact that GIF1/AN3 expression, assayed by GUS histochemical stain-ing in promoter-GUS fusion transgenic lines, washighly active only in the very early stage of primordialleaves (Horiguchi et al., 2005). Second, while thefrequency of cell proliferation in young organs of antand klumutants is very similar to that in the wild type(Mizukami and Fischer, 2000; Anastasiou et al., 2007),that of gif double and triple mutants is low, especiallyat the very early stage showing the maximum prolif-erating activity (Fig. 4). The low frequency of cellproliferation in gifmutants is supported by the findingthat the transcript levels of cell cycle-regulating genesare reduced in the gif triple mutant at day 6 (Fig. 5).Like GIF genes that are highly expressed in the shootapex (Fig. 1), those cell cycle genes have been shown tobe actively expressed in both the SAM and developingprimordia in Arabidopsis and tobacco (Nicotiana taba-cum; Segers et al., 1996; Donnelly et al., 1999; Kosugiand Ohashi, 2002; Dewitte et al., 2003). This raises thepossibility that the function of GIF genes may bemediated through the control of expression of cell cyclegenes. For instance, the expression level of CycB1;1was significantly reduced in the gif triple mutant(Fig. 5), and the GUS activity under the promoter ofCycB1;1 was especially high at marginal positions ofthe wild-type leaf primordia that were rapidly ex-panding along the mediolateral axis (Donnelly et al.,1999). In addition, employing promoter::GUS reporterlines and RT-PCR analysis, Horiguchi et al. (2005) alsoshowed that the expression gradient of GIF1/ANparalleled that of the G2/M transition-specific cyclingene along the longitudinal axis of developing leaves.Taken together, these results suggest that a decrease inthe marginal proliferation activity would lead to anarrow leaf, which is in a good agreement with the gifphenotype.The fact that the gif triple mutant produces leaf

primordia with fewer cells raises the following possi-bility: besides low frequency of cell proliferation (Fig.4), the leaf primordium of gif mutants may serve as apoor reservoir of progenitor cells available for further

cell division. That, in turn, would lead to earlierexhaustion of meristematic cells, resulting in the ear-lier cessation and, thus, the short duration of cellproliferation. Autran et al. (2002) demonstrated thatthe swp mutant produced small leaf primordia withreduced cell numbers, suggesting that the primordialcell pool might be an important determinant of thefinal leaf size.

Relationship between the Sizes of the SAM Cell Pool,Leaf Primordium, and Plastochron

One of the key functions of the SAM is to formlateral organs, such as leaves and flowers, from itsperipheral zone. It has been proposed that the growthof young developing primordium is preceded byactive mitosis in a large part of the SAM L1 layer(Laufs et al., 1998b). Clonal and anatomical analysessuggested that 12 to 30 founder cells of the peripheralzone in a mature embryo of Arabidopsis contribute toeach of the first two leaf primordia and that cells in theSAM continue to be recruited into the developingprimordium even after imbibition of seeds (Furnerand Pumfrey, 1992; Irish and Sussex, 1992). In the SAMof narrow sheath mutants of maize (Zea mays), fewercells are recruited into leaf founder cells, producingextremely narrow leaves (Scanlon et al., 1996; Scanlon,2000). Another maize mutant, rough sheath2, developedreduced numbers of leaf founder cells in the SAM,resulting in narrow or bladeless leaves (Schneebergeret al., 1998; Timmermans et al., 1999). Therefore, it isconceivable that the small size of leaf primordia in thegif mutants may be, in part, attributable to the smallsize of the SAM cell pool, probably because the SAMtissues do not have sufficient numbers of founder cells.This notion is in line with the fact that all of thecytokinin-signaling mutants and transgenic plantswith reduced cytokinin activities have both the smallerSAM cell pool and leaf organs with fewer cells com-pared with the wild type (Werner et al., 2001, 2003;Higuchi et al., 2004; Nishimura et al., 2004; Miyawakiet al., 2006). However, precise data on the causalrelationship between altered numbers of the SAMcells and leaf cells are scarce in Arabidopsis. A betterunderstanding of the gif mutants may help clarify therelationship between the sizes of the SAM and the leafprimordium in the future.

Meanwhile, it has been well documented that, ingeneral, plastochron length and SAM size are in-versely correlated in Arabidopsis. The cytokinin mu-tants and transgenic plants mentioned above havemuch smaller SAM and fewer leaves compared withthe wild type, which is indicative of a longer plasto-chron (Werner et al., 2001, 2003; Higuchi et al., 2004;Nishimura et al., 2004; Miyawaki et al., 2006). Up-regulation of SQUAMOSA PROMOTER BINDINGPROTEIN-LIKE9 expression results in both a smallersize of the meristem and a longer plastochron (Wanget al., 2008). On the other hand, bigger SAM size ofclavata1 and altered meristem program1 mutants was





accompaniedbyshorterplastochron length (Chaudhuryet al., 1993; Kwon et al., 2005; Vidaurre et al., 2007). Theinverse correlation, however, does not appear to berobust, as gif mutants display a significant shorteningof plastochron length, despite their reduced SAM size(Fig. 6; Table I). In addition, recessive mutations inMGOUN1 and MGOUN2 cause a reduction in leafnumber and larger meristem (Laufs et al., 1998a).Therefore, it appears that alteration in the plasto-chronic property of mutant plants may depend notonly on the size of the SAM cell pool but also on theregulatory nature of each mutation in the SAM tissue.

MATERIALS AND METHODS

Plant Material

The Arabidopsis (Arabidopsis thaliana) seeds were sown on wet soil (Mix5;

Sunshine), stratified at 4�C for 3 d, and transferred to a growth room at 23�Cunder a photoperiod of 16 h of light/8 h of darkness, which was marked as

day 0 throughout the experiments. For measurement of leaf plastochron,

plants were grown under the short-day conditions of 8 h of light/16 h of

darkness. Wild-type plants and all of the T-DNA insertional mutants were in

the Columbia ecotype. gif1 (SALK_150407; Kim and Kende, 2004), gif2

(SAIL_328_A03; Sessions et al., 2002), and gif3 (SALK_072950; Alonso et al.,

2003) seeds were obtained from the Arabidopsis Biological Resource Center.

Identification of T-DNA Insertional Mutants andConstruction of Multiple Mutants

Homozygous mutant plants were selected by PCR-assisted genotyping.

Primers for amplification of wild-type GIF genes and T-DNAs are described in

Supplemental Table S2. DNA fragments amplified with the gene-specific and

left-border primers were sequenced to confirm the T-DNA insertion site. The

double homozygous lines, gif1 gif2, gif2 gif3, and gif1 gif3, were established

through crosses between each homozygous line. The gif1 gif2 double mutant

was maintained in the gif1/+ gif2/gif2 genotype, as it was female sterile,

although it produced fertile pollen. The gif1 gif2 gif3 triple mutant was

obtained as the F2 progeny of the cross between gif1 gif2 (male) and gif2 gif3

(female) and maintained in gif1/+ gif2/gif2 gif3/gif3.

Construction of Transgenic Plants OverexpressingGIF Genes

GIF1, GIF2, and GIF3 cDNAs were amplified by PCR using primer pairs

containing the Gateway partial recombination site at the 5# end, attB1 or attB2(for primer sequence information, see Supplemental Table S2). The amplified

DNA fragments were used as template for a second round of PCR with

adaptor primers. The final cDNA fragments were inserted into the entry

vector pDONR221 by the BP recombinant reaction, according to the manu-

facturer’s protocol (Invitrogen). The resulting plasmids were used in the LR

reaction with the destination vector pB2GW7,0 (http://www.psb.ugent.be/

gateway) to produce GIF overexpression constructs driven by the cauliflower

mosaic virus 35S promoter. The recombinant binary plasmids were intro-

duced into gif1 mutant plants by Agrobacterium tumefaciens-mediated trans-

formation (Clough and Bent, 1998). Dozens of independent T1 plants

overexpressing GIF1, GIF2, and GIF3 were selected on soil (Sunshine Mix1)

soaked with the herbicide ammonium glufosinate (Bayer Cropscience).

Single-insertion homozygous T3 lines of the GIF1, GIF2, and GIF3 over-

expressors in the gif1 mutant background were selected and established. For

the construction of GIF1, GIF2, and GIF3 overexpressors in the wild-type

background, the established T3 lines were crossed to wild-type plants, after

which plants homozygous for transgenes were established in the F3 progeny

via PCR genotyping and herbicide resistance.

RT-PCR Analyses

For determination of GIF mRNA content in T-DNA insertional mutants,

total RNAs were extracted with TRIzol reagent (Invitrogen) from 15-d-old

wild-type and mutant plants, treated with DNase I (DNA-free; Ambion), and

subjected to RT (SuperScript II; Invitrogen). The resulting cDNAs were used

for PCR amplification, the conditions for which were as follows: denaturation

at 95�C for 2 min, followed by 34 cycles of 95�C for 15 s, 52�C for 30 s, and 72�Cfor 1 min (for primer sequence information, see Supplemental Table S2).

For determination of transcript levels of cell-cycling regulators, cDNAs

were prepared from the 6-d-old first two leaves as mentioned above, serially

diluted to the indicated factors, and used for amplification. PCR conditions

were as follows: denaturation at 95�C for 2 min, followed by 29 cycles of 95�Cfor 15 s, 52�C for 30 s, and 72�C for 1 min (for primer sequence information, see

Supplemental Table S2).

Microarray Data Profiling

The microarray data of the AtGenExpress expression atlas were retrieved

from The Arabidopsis Information Resource (Schmid et al., 2005; ftp://ftp.

arabidopsis.org/home/tair/Microarrays/Datasets /AtGenExpress).

Measurement of Dimensional Parameters of Leaves,Cotyledons, and Petals

Digital images of detached leaves, cotyledons, and petals were acquired

using a scanner. Area, length, and width of leaves, cotyledons, and mature

petals as well as petioles were determined with the image-analyzing program

SCIONIMAGE (Scion). For kinematic analysis of leaf area, young leaves up to

day 9 were detached and mounted on slide glass for light-microscopic image

analysis.

Number and Size of Leaf-Blade, Cotyledon, andPetal Cells

Leaf, cotyledon, and petal tissues were fixed with ethanol:acetic acid (6:1)

for 4 h and were washed with 100% ethanol three times and then 70% ethanol

once. Finally, all tissues were cleared in a chloral hydrate solution (8 g of

chloral hydrate, 1 mL of glycerol, and 2mL of distilled water) andmounted on

slide glass. The microscopic images were obtained using a differential

interference contrast microscope (Zeiss Axioplan). Subepidermal cells aligned

along a longitudinal axis just beside the midvein or along a transverse axis in

the maximum width region were counted. To determine cell area, 20 cells

grouped halfway from the midvein to the leaf margin at the widest point were

analyzed with SCIONIMAGE software.

Confocal Laser Scanning Microscopy for the SAM Cells

Mature embryos were prepared and stained with propidium iodide

according to Running et al. (1995), after which the shoot apical region of

mature embryos was analyzed and photographed by confocal laser scanning

microscopy (Bio-Rad MRC-1024). Nuclei in the SAM region were counted to

determine cell numbers.

Sequence data from this article can be found in the GenBank/EMBL data

libraries under accession numbers AY102639 through AY102641 for GIF1

through GIF3, respectively.

Supplemental Data

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

Supplemental Figure S1. Expression profile of GIF genes.

Supplemental Table S1. Numbers of seeds produced from crosses be-

tween Columbia and the gif1 gif2 double mutant.

Supplemental Table S2. Primer sequences for PCR amplification.

ACKNOWLEDGMENTS

We thank Dr. Woo Taek Kim for technical support, critical reading of the

manuscript, and helpful comments; Dr. Myeong Min Lee for technical

support; and Drs. Soon Ki Park and Jong Tae Song for sharing their growth

Lee et al.




room. We also thank the Arabidopsis Biological Resource Center for the

mutant seeds.

Received May 21, 2009; accepted July 27, 2009; published July 31, 2009.

LITERATURE CITED

Alonso JM, Stepanova AN, Leisse TJ, KimCJ, ChenH, Shinn P, Stevenson

DK, Zimmerman J, Barajas P, Cheuk R, et al (2003) Genome-wide

insertional mutagenesis of Arabidopsis thaliana. Science 301: 653–657

Anastasiou E, Kenz S, Gerstung M, MacLean D, Timmer J, Fleck C,

Lenhard M (2007) Control of plant organ size by KLUH/CYP78A5-

dependent intercellular signaling. Dev Cell 13: 843–856

Anderson HJ, Vonarx EJ, Pastushok L, NakagawaM, Katafuchi A, Gruz P,

Di Rubbo A, Grice DM, Osmond MJ, Sakamoto AN, et al (2008)

Arabidopsis thaliana Y-family DNA polymerase h catalyses translesion

synthesis and interacts functionally with PCNA2. Plant J 55: 895–908

Autran D, Jonak C, Belcram K, Beemster GT, Kronenberger J, Grandjean

O, Inze D, Traas J (2002) Cell numbers and leaf development in

Arabidopsis: a functional analysis of the STRUWWELPETER gene.

EMBO J 21: 6036–6049

Chaudhury AM, Letham S, Craig S, Dennis ES (1993) amp1: a mutant with

high cytokinin levels and altered embryonic pattern, faster vegetative

growth, constitutive photomorphogenesis and precocious flowering.

Plant J 4: 907–916

Cho KH, Jun SE, Lee YK, Jeong SJ, Kim GT (2007) Developmental

processes of leaf morphogenesis in Arabidopsis. J Plant Biol 50: 282–290

Clough SJ, Bent AF (1998) Floral dip: a simplified method for Agrobacterium-

mediated transformation of Arabidopsis thaliana. Plant J 16: 735–743

de Bruijn DR, dos Santos NR, Thijssen J, Balemans M, Debernardi S,

Linder B, Young BD, Geurts van Kessel A (2001) The synovial sarcoma

associated protein SYT interacts with the acute leukemia associated

protein AF10. Oncogene 20: 3281–3289

Dewitte W, Riou-Khamlichi C, Scofield S, Healy JM, Jacqmard A, Kilby

NJ, Murray JA (2003) Altered cell cycle distribution, hyperplasia, and

inhibited differentiation in Arabidopsis caused by the D-type cyclin

CYCD3. Plant Cell 15: 79–92

Disch S, Anastasiou E, Sharma VK, Laux T, Fletcher JC, Lenhard M (2006)

The E3 ubiquitin ligase BIG BROTHER controls Arabidopsis organ size in

a dosage-dependent manner. Curr Biol 16: 272–279

Donnelly PM, Bonetta D, Tsukaya H, Dengler RE, Dengler NG (1999) Cell

cycling and cell enlargement in developing leaves of Arabidopsis. Dev

Biol 215: 407–419

Eid JE, Kung AL, Scully R, Livingston DM (2000) p300 interacts with the

nuclear proto-oncoprotein SYT as part of the active control of cell

adhesion. Cell 102: 839–848

Ferjani A, Horiguchi G, Yano S, Tsukaya H (2007) Analysis of leaf

development in fugu mutants of Arabidopsis reveals three compensa-

tion modes that modulate cell expansion in determinate organs. Plant

Physiol 144: 988–999

Furner IJ, Pumfrey JE (1992) Cell fate in the shoot apical meristem of

Arabidopsis thaliana. Development 115: 755–764

Han SK, Song JD, Noh YS, Noh B (2007) Role of plant CBP/p300-like

genes in the regulation of flowering time. Plant J 49: 103–114

Higuchi M, Pischke MS, Mahonen AP, Miyawaki K, Hashimoto Y, Seki

M, Kobayashi M, Shinozaki K, Kato T, Tabata S, et al (2004) In planta

functions of the Arabidopsis cytokinin receptor family. Proc Natl Acad Sci

USA 101: 8821–8826

Horiguchi G, Kim GT, Tsukaya H (2005) The transcription factor AtGRF5

and the transcription coactivator AN3 regulate cell proliferation in leaf

primordia of Arabidopsis thaliana. Plant J 43: 68–78

Hurtado L, Farrona S, Reyes JC (2006) The putative SWI/SNF complex

subunit BRAHMA activates flower homeotic genes in Arabidopsis

thaliana. Plant Mol Biol 62: 291–304

Irish VF, Sussex IM (1992) A fate map of the Arabidopsis embryonic shoot

apical meristem. Development 115: 745–753

Kato H, Tjernberg A, Zhang W, Krutchinsky AN, An W, Takeuchi T,

Ohtsuki Y, Sugano S, de Bruijn DR, Chait BT, et al (2002) SYT

associates with human SNF/SWI complexes and the C-terminal region

of its fusion partner SSX1 targets histones. J Biol Chem 277: 5498–5505

Kim JH, Choi D, Kende H (2003) The AtGRF family of putative transcrip-

tion factors is involved in leaf and cotyledon growth in Arabidopsis. Plant

J 36: 94–104

Kim JH, Kende H (2004) A transcriptional coactivator, AtGIF1, is involved

in regulating leaf growth and morphology in Arabidopsis. Proc Natl

Acad Sci USA 101: 13374–13379

Kim JH, Lee BH (2006) GROWTH-REGULATING FACTOR4 of Arabidopsis

thaliana is required for development of leaves, cotyledons, and shoot

apical meristem. J Plant Biol 49: 463–468

Kosugi S, Ohashi Y (2002) E2F sites that can interact with E2F proteins

cloned from rice are required for meristematic tissue-specific expression

of rice and tobacco proliferating cell nuclear antigen promoters. Plant J

29: 45–59

Kwon CS, Chen C, Wagner D (2005) WUSCHEL is a primary target for

transcriptional regulation by SPLAYED in dynamic control of stem cell

fate in Arabidopsis. Genes Dev 19: 992–1003

Kwon CS, Hibara K, Pfluger J, Bezhani S, Metha H, Aida M, Tasaka M,

Wagner D (2006) A role for chromatin remodeling in regulation of CUC

gene expression in the Arabidopsis cotyledon boundary. Development

133: 3223–3230

Laufs P, Dockx J, Kronenberger J, Traas J (1998a)MGOUN1 andMGOUN2:

two genes required for primordium initiation at the shoot apical and

floral meristems in Arabidopsis thaliana. Development 125: 1253–1260

Laufs P, Grandjean O, Jonak C, Kieu K, Traas J (1998b) Cellular param-

eters of the shoot apical meristem in Arabidopsis. Plant Cell 10: 1375–1390

Masubelele NH, Dewitte W, Menges M, Maughan S, Collins C, Huntley

R, Nieuwland J, Scofield S, Murray JA (2005) D-type cyclins activate

division in the root apex to promote seed germination in Arabidopsis.

Proc Natl Acad Sci USA 102: 15694–15699

Menges M, Samland AK, Planchais S, Murray JA (2006) The D-type cyclin

CYCD3;1 is limiting for the G1-to-S-phase transition in Arabidopsis. Plant

Cell 18: 893–906

Mironov VV, De Veylder L, Van Montagu M, Inze D (1999) Cyclin-

dependent kinases and cell division in plants: the nexus. Plant Cell 11:

509–522

Miyawaki K, Tarkowski P, Matsumoto-Kitano M, Kato T, Sato S,

Tarkowska D, Tabata S, Sandberg G, Kakimoto T (2006) Roles of

Arabidopsis ATP/ADP isopentenyltransferases and tRNA isopentenyl-

transferases in cytokinin biosynthesis. Proc Natl Acad Sci USA 103:

16598–16603

Mizukami Y (2001) A matter of size: developmental control of organ size in

plants. Curr Opin Plant Biol 4: 533–539

Mizukami Y, Fischer RL (2000) Plant organ size control: AINTEGUMENTA

regulates growth and cell numbers during organogenesis. Proc Natl

Acad Sci USA 97: 942–947

Nishimura C, Ohashi Y, Sato S, Kato T, Tabata S, Ueguchi C (2004)

Histidine kinase homologs that act as cytokinin receptors possess

overlapping functions in the regulation of shoot and root growth in

Arabidopsis. Plant Cell 16: 1365–1377

Perani M, Antonson P, Hamoudi R, Ingram CJ, Cooper CS, Garrett MD,

Goodwin GH (2005) The proto-oncoprotein SYT interacts with SYT-

interacting protein/co-activator activator (SIP/CoAA), a human nu-

clear receptor co-activator with similarity to EWS and TLS/FUS family

of proteins. J Biol Chem 280: 42863–42876

Perani M, Ingram CJ, Cooper CS, Garrett MD, Goodwin GH (2003)

Conserved SNH domain of the proto-oncoprotein SYT interacts with

components of the human chromatin remodelling complexes, while the

QPGY repeat domain forms homo-oligomers. Oncogene 22: 8156–8167

Running MP, Clark SE, Meyerowitz EM (1995) Confocal microscopy of the

shoot apex. Methods Cell Biol 49: 217–229

Scanlon MJ (2000) NARROW SHEATH1 functions from two meristematic

foci during founder-cell recruitment in maize leaf development. Devel-

opment 127: 4573–4585

Scanlon MJ, Schneeberger RG, Freeling M (1996) The maize mutant

narrow sheath fails to establish leaf margin identity in a meristematic

domain. Development 122: 1683–1691

Schmid M, Davison TS, Henz SR, Pape UJ, Demar M, Vingron M,

Scholkopf B, Weigel D, Lohmann JU (2005) A gene expression map of

Arabidopsis thaliana development. Nat Genet 37: 501–506

Schneeberger R, Tsiantis M, Freeling M, Langdale JA (1998) The ROUGH

SHEATH2 gene negatively regulates homeobox gene expression during

maize leaf development. Development 125: 2857–2865

Segers G, Gadisseur I, Bergounioux C, de Almeida Engler J, Jacqmard A,

Van Montagu M, Inze D (1996) The Arabidopsis cyclin-dependent kinase

gene cdc2bAt is preferentially expressed during S and G2 phases of the

cell cycle. Plant J 10: 601–612





Sessions A, Burke E, Presting G, Aux G, McElver J, Patton D, Dietrich B,

Ho P, Bacwaden J, Ko C, et al (2002) A high-throughput Arabidopsis

reverse genetics system. Plant Cell 14: 2985–2994

Telfer A, Poethig RS (1994) Leaf development in Arabidopsis. In EM

Meyerowitz, CR Somerville, eds, Arabidopsis. Cold Spring Harbor

Laboratory Press, Cold Spring Harbor, NY, pp 379–401

Thaete C, Brett D, Monaghan P, Whitehouse S, Rennie G, Rayner

E, Cooper CS, Goodwin G (1999) Functional domains of the SYT and

SYT-SSX synovial sarcoma translocation proteins and co-localization

with the SNF protein BRM in the nucleus. Hum Mol Genet 8: 585–591

Timmermans MC, Hudson A, Becraft PW, Nelson T (1999) ROUGH

SHEATH2: a Myb protein that represses knox homeobox genes in maize

lateral organ primordia. Science 284: 151–153

Tsukaya H (2003) Organ shape and size: a lesson from studies of leaf

morphogenesis. Curr Opin Plant Biol 6: 57–62

Tsukaya H (2005) Leaf shape: genetic controls and environmental factors.

Int J Dev Biol 49: 547–555

Vandepoele K, Raes J, De Veylder L, Rouze P, Rombauts S, Inze D (2002)

Genome-wide analysis of core cell cycle genes in Arabidopsis. Plant Cell

14: 903–916

Vidaurre DP, Ploense S, Krogan NT, Berleth T (2007) AMP1 and MP

antagonistically regulate embryo and meristem development in Arabi-

dopsis. Development 134: 2561–2567

Wang JW, Schwab R, Czech B, Mica E, Weigel D (2008) Dual effects of

miR156-targeted SPL genes and CYP78A5/KLUH on plastochron length

and organ size in Arabidopsis thaliana. Plant Cell 20: 1231–1243

Werner T, Motyka V, Laucou V, Smets R, Van Onckelen H, Schmulling T

(2003) Cytokinin-deficient transgenic Arabidopsis plants show multiple

developmental alterations indicating opposite functions of cytokinins in

the regulation of shoot and root meristem activity. Plant Cell 15: 2532–2550

Werner T, Motyka V, Strnad M, Schmulling T (2001) Regulation of plant

growth by cytokinin. Proc Natl Acad Sci USA 98: 10487–10492

White DW (2006) PEAPOD regulates lamina size and curvature in

Arabidopsis. Proc Natl Acad Sci USA 103: 13238–13243

Lee et al.




The Arabidopsis GRF-INTERACTING FACTOR Gene Family ... · expressors contained more cells (Kim and...

Documents

Transcript of The Arabidopsis GRF-INTERACTING FACTOR Gene Family ... · expressors contained more cells (Kim and...