Pii: s1369-5266(99)00036-9

Three ways to learn the ABCs
Medard Ng* and Martin F Yanofsky†
The ABC model of flower development represents a milestone carpels. An example of the A gene is APETALA1 (AP1), the in explaining how the fate of emerging floral organ primordia is B genes are APETALA3 (AP3) and PISTILLATA (PI) and specified. This model states that organ identity is specified by the C gene is AGAMOUS (AG) [3–8]. Subsequent cloning of different combinations of the activities of the A, B and C class AP1, AP3, PI and AG show that they are all members of the homeotic genes. In spite of the remarkable simplicity of this MADS-box gene family and are expressed only in regions model, the complex regulatory interactions that establish the of the developing flowers that require their activities initial pattern of A, B and C gene activity have yet to be fully (Figure 1, reviewed in [9]). Although these results explain explained. It has been shown that the LEAFY gene functions their region-specific requirement in specifying floral organ early to promote flower meristem identity, and that it is identities, they also raise the issue of how the floral subsequently required for the normal expression of the ABC homeotic genes are activated in floral meristems, and how genes. Recently, LEAFY has been identified as an immediate they come to be expressed in spatially restricted domains.
upstream regulator of the floral homeotic genes, thus openingup an avenue to examine the transcriptional interactions that In order to address these issues, genes acting upstream of the floral homeotic genes must be identified. The meris-tem identity genes, which include LEAFY (LFY), Addresses
UNUSUAL FLORAL ORGANS (UFO) and AP1, are excel- Department of Biology, University of California at San Diego, La Jolla, lent candidates for such upstream acting genes [3,7,10–14].
Plants carrying mutations in any of the meristem identity genes produce flowers with shoot-like characters, which suggests these genes normally instruct the meristems to Current Opinion in Plant Biology 2000, 3:47–52
adopt a floral fate. As the determination of floral meristemidentity precedes that of floral organ identity, the meristem 1369-5266/00/$ — see front matter 2000 Elsevier Science Ltd. All rights reserved.
identity genes must act upstream of the floral homeoticgenes. LFY encodes a novel transcription factor, whereas Abbreviations
UFO encodes an F-box-containing protein [12,14,15,16••].
AG
AP
LFY, UFO and AP1 are expressed very early during flower FIM
development, consistent with the idea that they can acti- LFY
vate the expression of the floral homeotic genes (Figure 1).
PI
Three recent papers now convincingly show that AP1 and UFO
AG are direct targets of LFY, and that LFY activates AP1,AP3 and AG using different mechanisms [16••–18••].
Introduction
AG is a direct target of LFY
The Arabidopsis flower is arguably the best understood Mutations in AG result in flowers with third-whorl petals, plant model system pertaining to pattern formation.
and the fourth whorl develops as a new ag mutant flower Flowers originate from small groups of undifferentiated [4]. Transcripts of the C class gene AG can be detected in cells called floral meristems, which, in turn, are derived the center of a wild-type flower from floral stage 3, corre- from the shoot apical meristem (SAM). Similar to most sponding to whorls three and four (Figure 1, [4,19]). The other dicotyledonous plants, an Arabidopsis flower consists first indication that LFY may be critical for AG expression of four types of floral organs arranged in four concentric came from analyzing the expression pattern of AG in strong whorls. Four sepals, four petals, six stamens and two fused lfy mutants, in which the early arising flowers are trans- carpels can be found from the periphery to the center of formed into leaves with associated shoots, whereas the later arising flowers develop bracts in whorl one, lackpetals and stamens, and develop irregularly fused carpels In order to understand how flowers develop into their final in the center [12,20,21]. In the later arising flowers the shape and form, a genetic approach has been used to iden- onset of AG expression is delayed, although AG eventually tify genes that when inactivated will perturb the flower morphology. In this review, we focus on the meristem iden-tity genes and the floral homeotic genes. Analyses of To begin probing into the mechanism by which LFY acti- mutations in the latter class of genes led to the formulation vates AG expression, Parcy et al. [16••] generated a of the ‘ABC’ model of flower development (reviewed in gain-of-function LFY allele, called LFY:VP16, that is consti- [1,2]). According to this model, the A genes specify sepals, tutively active by inserting the transcriptional activation the A and B genes together specify petals, the B and C domain of VP16 into LFY. This experiment aimed to address genes together specify stamens, and the C gene specifies whether the role of LFY in activating floral homeotic genes Growth and development
Expression patterns of some meristem identitygenes and floral organ identity genes thatfunction early in flower development. Theexpressions of the meristem identity genes overlap spatially and temporally with the organidentity genes, consistent with the idea thatthe former acts upstream of the latter.
Numbers indicate floral stages.
could be separated from its earlier role in specifying the iden- This observation is consistent with the fact that ubiquitous tity of floral meristems. It turned out that LFY:VP16 expression of LFY:VP16 in the vegetative tissue also leads transgenic plants have unaltered ability to initiate flowers, to parallel ubiquitous expression of AG.
whereas the flower morphology is clearly affected, implicat-ing LFY in regulating floral homeotic genes.
If LFY:VP16 can activate AG in all tissue types, why doesLFY, which is present throughout the floral meristem from As with all gain-of-function alleles, it is important to show floral stages 1 to 3, normally induce AG only in the center that the LFY:VP16 phenotypes reveal the normal functions of the flower [12,13,16••]? Parcy et al. [16••] have proposed of the endogenous LFY gene. Therefore, Parcy et al. [16••] two models to explain this conundrum. First, a repressor performed two elegant control experiments. First, the activity, such as APETALA2, is present in the periphery of LFY:VP16 phenotypes are enhanced by decreasing the the floral meristem and prevents LFY from activating AG gene dosage of the endogenous LFY, which strongly sug- expression [24]. Second, the repressor activity is selective- gests that LFY:VP16 competes for the same target genes as ly overcome in the center of the stage 3 floral meristem. In the endogenous LFY. Furthermore, LFY:VP16 rescues the either model, regional expression of AG requires LFY, flower initiation defects of lfy mutants. Second, a mutant which provides floral meristem-specific cue, and an unde- LFY:VP16 version (called LFY:mVP16), in which a mutant fined molecule ‘X’, which provides the C-region cue. Parcy VP16 domain that is inactive in transcriptional activation et al. [16••] have suggested further that the VP16 transcrip- was inserted into LFY, fully rescues lfy mutants but does tional activation domain renders the LFY protein not affect floral morphology. These results show that inser- independent of other regulators of AG, leading to the acti- tion of a foreign polypeptide per se does not activate or vation of AG expression throughout the flower.
In a follow-on study, LFY has been shown to activate AG A detailed examination of LFY:VP16 plants shows that expression directly by binding to an enhancer in the first sepals are converted to carpels, and petals to stamens [16••].
intron of AG [17••]. It was previously demonstrated that the In short, they resemble plants constitutively expressing AG first intron of AG is crucial for its expression [25]; Bush et al.
[23]. RNA in situ hybridizations confirm that AG expression [17••] have gone on to show that transcriptional enhancers is ectopic, precocious and at high levels in LFY:VP16 plants.
present in the first intron of AG are sufficient to confer a Three ways to learn the ABCs Ng and Yanofsky 49
wild-type AG expression pattern. They embarked on a ‘tour Wagner et al. [18••] fused the hormone-binding domain of de force’ approach to define the smallest piece of DNA that the rat glucocorticoid receptor to LFY [32]. The chimeric retains full activities of the AG enhancer, and found that protein is expressed constitutively under the control of the two non-overlapping fragments from the intron can confer 35S promoter (35S::LFY-GR). In the absence of glucocor- AG-specific expression. Busch et al. [17••] decided to focus ticoid, LFY-GR should be tethered in the cytoplasm by on the smaller 3′ enhancer because its expression seems to interaction with the chaperon proteins, rendering the tran- be less complex. Further deletions of the 3′ enhancer led to scription factor LFY inactive [18••,32,33•]. Upon progressive reduction of its activity; however, LFY respon- treatment with glucocorticoid, LFY-GR dissociates from siveness could be followed using LFY:VP16. This approach the chaperons, translocates to the nucleus, and regulates defines a LFY responsive element to a 230 bp region. Using target gene expression. As activation of LFY-GR is post- an in vitro DNA-binding assay, two closely spaced LFY translational, the immediate effect of LFY activation on binding sites were identified in the 230 bp fragment of the transcription of any LFY target genes can be monitored in AG intron. To assess the functional significance of these the presence of a protein synthesis inhibitor.
sites in vivo, a small deletion including the LFY bindingsites, and point mutations abolishing LFY binding in vitro The 35S::LFY-GR construct was introduced into strong lfy were introduced into the AG 3′ enhancer. Mutating both mutants. To show that the translational fusion does not binding sites inactivates the enhancer, whereas mutating compromise LFY activity, Wagner et al. [18••] examined one site significantly attenuates the enhancer. Taken the phenotype of these plants upon dexamethasone (a together, these results show that LFY is a direct upstream strong glucocorticoid) treatment. They found that, after such treatment, 35S::LFY-GR mostly rescues the floralmorphology of lfy mutants. In addition, an early flowering AP1 is also a direct target of LFY
phenotype and the shoot-to-flower conversions were AP1 is both a meristem identity gene and an A function observed in these dexamethasone treated plants; there- organ identity gene, as mentioned above. Plants homozy- fore, 35S::LFY-GR has the same activities as 35S::LFY gous for strong ap1 alleles develop bracts in the first floral [18••,29]. To evaluate whether AP1 is a direct target of whorl, usually lack petals, and have secondary flowers in LFY, AP1 expression was analyzed in lfy mutants carrying the axils of the first floral organs [3,7]. AP1 is initially the 35S::LFY-GR transgene treated with dexamethasone expressed throughout the floral meristem from floral stages and cycloheximide. Cycloheximide treatment, which pre- 1 to 3 (Figure 1, [6]). Expression then abates in the two vents protein synthesis, ensures that only genes directly central whorls because of negative regulation by AG [6,26].
activated by LFY are induced upon dexamethasone addi- To explain how AP1 is expressed in the outer two whorls of tion. AP1 RNA could be detected in young flower a mature flower, therefore, one needs to explain how AP1 primordia of these plants eight hours after dexamethasone is initially expressed throughout the floral meristem.
Numerous experiments suggest that AP1 activity is mostly Activation of AP3 by LFY and UFO
downstream of LFY. AP1 expression is significantly Plants carrying mutations in AP3 or PI have sepals in the delayed and reduced in lfy mutants [27•,28•]. Constitutive second and carpels in the third whorl [5,8]. AP3 starts to be expression of LFY (35S::LFY) leads to precocious expres- expressed in the second and third whorls from floral stage sion of AP1 [29]. In addition, ap1 mutants attenuate the 3 (Figure 1, [5]). Whereas the initiation of AP3 expression shoot-to-flower conversion phenotype of the 35S::LFY is unchanged in ap3 and pi mutants, continued expression plants, whereas the gain-of-function phenotype of 35S::AP1 of AP3 after floral stage 6 depends on wild-type activities plants is mostly unaffected by mutations in LFY [27•,29,30]. These observations prompted Parcy et al. [16••]to examine AP1 expression in LFY:VP16 plants, although Several lines of evidence suggest that LFY and UFO are the phenotype of these plants suggests no a priory reason key upstream regulators of AP3. Flowers from strong lfy for altered AP1 expression. They found that the level of and ufo mutants lack petals and stamens [12–14,20,21]; early AP1 expression is greatly elevated, although its early AP3 expression is significantly reduced in strong lfy and ufo pattern of expression is unaltered. In vitro DNA-binding mutants [13,22]; and constitutive expression of AP3 and PI assays showed that a high-affinity LFY binding site is pre- partially restores stamens and petals in lfy and ufo mutants, sent in the AP1 promoter [16••,17••]. Although this site is whereas constitutive expression of UFO does not rescue present in a minimal AP1 promoter, its in vivo function is ap3 and pi mutants [15,35].
unknown [31]. Consistent with the idea that AP1 expres-sion is directly regulated by LFY, LFY:VP16 activates the Consistent with its upstream regulatory role, UFO RNA expression of a reporter gene in yeast under the control of accumulates in the floral meristem before the onset of AP3 an AP1 promoter containing the LFY binding site [16••].
expression [15,36]. UFO is first expressed in the centraldome, including the presumptive third and fourth whorls, Another recent study provides evidence that LFY is a during floral stage 2 (Figure 1, [15]). During floral stage 3, direct transcriptional activator of AP1 in vivo [18••].
its expression domain broadens, with the concomitant loss Growth and development
of UFO RNA in the center. During late floral stage 3, UFO identified, therefore, one might also start looking for LFY- RNA can be detected in the second and third whorls, sim- binding sites in the AP3 promoter [38•,39•].
ilar to the AP3 expression domain at the same stage.
During floral stage 4, the UFO domain becomes mainly As UFO is not a transcription factor, it is unlikely to direct- restricted to the petal primordia. From embryonic to repro- ly activate AP3 transcription. Analyses of the Antirrhinum ductive phases of development, UFO is also expressed at UFO orthologue FIMBRIATA (FIM) provide good insights high levels at the periphery and at low levels at the center into how UFO may function [40]. Antirrhinum proteins with strong similarity to Skp1 from yeast and animals have beenshown to interact with FIM [40]. In yeast and humans, How do LFY and UFO activate AP3 expression? Clearly, Skp1 proteins interact with F-box containing proteins to the simple hierarchical models of UFO acting downstream form a complex targeted for degradation, which is required of LFY and LFY acting downstream of UFO are incorrect.
for cell cycle progression [41–43]. Furthermore, it has also This is because constitutive expression of UFO fails to res- been shown that an F-box protein, E3RSIκB, targets IκB, a cue lfy mutants, and, conversely, constitutive expression of repressor of the transcription factor NK-κB, for ubiquitin- LFY does not rescue ufo mutants [15,29]. Plants doubly proteasome-mediated degradation [44]. On the basis of transgenic for 35S::LFY and 35S::UFO have ubiquitous these results, it is tempting to speculate that a protein com- expression of AP3 throughout the developmentally arrest- plex formed by UFO and the Arabidopsis Skp1-like ed seedings [16••]. In contrast, AP3 cannot be detected in proteins might act by promoting degradation of a tran- seedlings expressing either LFY or UFO. On the basis of scriptional repressor of AP3. The absence of the repressor these results, Parcy et al. [16••] have suggested that during in the second and third whorls, and the presence of the normal development the expression domain of AP3 is activator LFY throughout the floral meristem may be nec- defined by LFY, which is expressed throughout the devel- essary for spatially restricted AP3 expression.
oping flower, and UFO, which is expressed in the emergingpetal and stamen primordia. In this context, LFY provides Conclusions
the floral meristem specificity and UFO provides the While significant progress has been made toward elucidat- regional specificity for AP3 expression, analogous to the ing how the floral homeotic genes are expressed in spatially proposal that LFY and the unknown factor ‘X’ activate AG restricted patterns, our understanding of the transcriptional expression in the center of the flower.
regulation of these genes is clearly incomplete. We nowknow that the LFY meristem identity gene directly acti- Although this model provides a good framework for AP3 vates AP1 and AG, and perhaps AP3, although these studies activation, some results cannot be easily reconciled with it.
indicate that additional factors (e.g., UFO, factor ‘X’) must Much evidence has shown that co-expression of LFY and also interact with LFY in this process. We also know that UFO is not sufficient for AP3 expression. First, the shoot many other genes are involved in regulating the ABC apex of 35S::LFY plants, which expresses endogenous genes, including AP1, AP2, CAULIFLOWER, CURLY UFO, does not express AP3 [16••]; similarly, AP3 cannot be LEAF, SUPERMAN and LEUNIG, although their mecha- detected in the shoot apex of plants expressing LFY under nistic roles have yet to be clearly defined [7,22,45–50].
the control of UFO promoter [15]. Second, constitutive Given the long-term goal of elucidating the cascade of tar- expression of UFO fails to initiate AP3 expression during get gene regulation beginning in the floral meristem with floral stage 1 of the floral primordia, which express high genes such as LFY, and ending with fully differentiated flo- levels of LFY [15]. These observations led Lee et al. [15] ral organs, it is clear that we are only scratching the surface to propose that induction of AP3 expression by UFO and of what promises to be a very deep and interesting story.
LFY is dependent on additional factors. It should beemphasized that the model proposed by Parcy et al. [16••] Acknowledgements
is not necessarily rejected by the results described above.
We thank François Parcy and Detlef Weigel for comments. MN received along-term post-doctorate fellowship from The Human Frontier Science For example, high levels of both LFY and UFO may be Program Organization (LT-367/97). Research in the laboratory of MFY is required for AP3 expression, and this condition may be supported by grants from the National Science Foundation and the achieved only in seedlings expressing LFY and UFO under the control of the 35S promoter, and in wild-type flowers.
References and recommended reading
At present, it is unclear whether LFY directly activates AP3 Papers of particular interest, published within the annual period of review, transcription. However, it should be very straightforward to address this issue by analyzing the lfy 35S::LFY-GR plants. If these plants carry an AP3::GUS reporter gene, they will stainpositive for GUS activity after dexamethasone treatment Coen ES, Meyerowitz EM: The war of the whorls: genetic interactions
controlling flower development. Nature 1991, 353:31-37.
[18••], and, if AP3 is a direct target of LFY, then AP3 and Weigel D, Meyerowitz E: The ABCs of floral homeotic genes. Cell
GUS RNA should be detected in these plants after treatment 1994, 78:203-209.
with dexamethasone and cycloheximide. Regulatory ele- Irish VF, Sussex IM: Function of the apetala1-1 gene during
ments crucial for AP3 expression are beginning to be Arabidopsis floral development. Plant Cell 1990, 2:741-751.
Three ways to learn the ABCs Ng and Yanofsky 51
Yanofsky MF, Ma H, Bowman JL, Drews GN, Feldmann KA, Meyerowitz 25. Sieburth LE, Meyerowitz EM: Molecular dissection of the
EM: The protein encoded by the Arabidopsis homeotic gene
AGAMOUS control region shows that cis elements for spatial
agamous resembles transcription factors. Nature 1990, 346:35-39.
regulation are located intragenically. Plant Cell 1997, 9:355-365.
Jack T, Brockman LL, Meyerowitz EM: The homeotic gene
26. Gustafson-Brown C, Savidge B, Yanofsky MF: Regulation of the
APETALA3 of Arabidopsis thaliana encodes a MADS box and is
Arabidopsis floral homeotic gene APETALA1. Cell 1994, 76:131-143.
expressed in petals and stamens. Cell 1992, 68:683-697.
Liljegren SJ, Gustafson-Brown C, Pinyopich A, Ditta GS, Mandel MA, Gustafson-Brown C, Savidge B, Yanofsky MF: Molecular
Yanofsky MF: Interactions among APETALA1, LEAFY, and
characterization of the Arabidopsis floral homeotic gene
TERMINAL FLOWER1 specify meristem fate. Plant Cell 1999,
APETALA1. Nature 1992, 360:273-277.
11:1007-1018.
Interactions among APETALA1, LEAFY and TERMINAL FLOWER1 are Bowman JL, Alvarez J, Weigel D, Meyerowitz EM, Smyth DR: Control
described in this paper. The authors show that AP1 can activate LFY expres- of flower development in Arabidopsis thaliana by APETALA1 and
sion, and vice versa. By contrast, AP1 and TFL1 negatively regulate each interacting genes. Development 1993, 119:721-743.
other’s expression. It is concluded that these interactions contribute to thesharp transition that occurs from vegetative to reproductive growth phases.
Goto K, Meyerowitz EM: Function and regulation of the Arabidopsis
floral homeotic gene PISTILLATA. Genes Dev 1993, 8:1548-1560.
28. Ratcliffe OJ, Bradley DJ, Coen ES: Separation of shoot and floral
identity in Arabidopsis. Development 1999, 126:1109-1120.
Riechmann JL, Meyerowitz EM: MADS domain proteins in plant
This paper describes the hierarchy among several genes (APETALA1, CAU- development. Biol Chem 1997, 378:1079-1101.
LIFLOWER, LEAFY and TERMINAL FLOWER 1) which affect the shootmeristems and floral meristems identities. The meristem identity genes (AP1, 10. Schultz EA, Haughn GW: LEAFY, a homeotic gene that regulates
CAL and LFY) prevent TFL1 transcription in the floral meristems; converse- inflorescence development in Arabidopsis. Plant Cell 1991,
ly, TFL1 delays the upregulation of the meristem identity genes, and prevents 3:771-781.
the shoot meristems from responding to LFY and AP1. The authors con- 11. Huala E, Sussex IM: LEAFY interacts with floral homeotic genes to
clude that the relative timing of upregulating TFL1 and the meristem identity regulate Arabidopsis floral development. Plant Cell 1992, 4:901-913.
genes determine their final expression patterns.
12. Weigel D, Alvarez J, Smyth DR, Yanofsky MF, Meyerowitz EM: LEAFY
29. Weigel D, Nilsson O: A developmental switch sufficient for flower
controls floral meristem identity in Arabidopsis. Cell 1992,
initiation in diverse plants. Nature 1995, 377:495-500.
69:843-859.
30. Mandel MA, Yanofsky MF: A gene triggering flower formation in
Arabidopsis. Nature 1995, 377:522-524.
13. Levin JZ, Meyerowitz EM: UFO: an Arabidopsis gene involved in
both floral meristem and floral organ development. Plant Cell
31. Hempel FD, Weigel D, Mandel MA, Ditta G, Zambryski PC, 1995, 7:529-548.
Feldman L, Yanofsky MF: Floral determination and expression of
floral regulatory genes in Arabidopsis. Development 1997,
14. Wilkinson MD, Haughn GW: UNUSUAL FLORAL ORGANS controls
124:3845-3853.
meristem identity and organ primordia fate in Arabidopsis.
Plant Cell 1995, 7:1485-1499.
32. Lloyd AM, Schena M, Walbot V, Davis RW: Epidermal cell fate
determination in Arabidopsis: patterns defined by a steroid-
15. Lee I, Wolfe DS, Nilsson O, Weigel D: A LEAFY co-regulator
inducible regulator. Science 1994, 266:436-439.
encoded by UNUSUAL FLORAL ORGANS. Curr Biol 1997, 7:95-104.
33. Sablowski RWM, Meyerowitz EM: A homolog of NO APICAL
16. Parcy F, Nilsson O, Busch MA, Lee I, Weigel D: A genetic framework
MERISTEM is an immediate target of the floral homeotic genes
for floral patterning. Nature 1998, 395:561-566.
APETALA3/PISTILLATA. Cell 1998, 92:93-103.
This paper describes the phenotype of plants carrying a gain-of-function APETALA3 and PISTILLATA function are placed under post-translational allele of LEAFY, called LFY:VP16, which was constructed by inserting the control by use of a steroid-inducible AP3 (35S::AP3-GR). Using differential transcriptional activation domain of VP16 into LFY. It was found that display, three direct targets of AP3/PI are identified and one of them is LFY:VP16 has different effects on the expression of the A, B and C class flo- called NAP (NO APICAL MERISTEM-LIKE, ACTIVATED BY AP3/PI). The ral homeotic genes. The authors conclude that LFY activates different floral expression pattern of NAP and the phenotypes caused by its mis-expression homeotic genes using different mechanisms.
suggest that NAP is important for cell division and cell expansion in stamensand petals.
Busch MA, Bomblies K, Weigel D: Activation of a floral homeotic
gene in Arabidopsis. Science 1999, 285:585-587.
34. Jack T, Fox GL, Meyerowitz EM: Arabidopsis homeotic gene
This is an extension of the previous study [16••]. The authors analyze the APETALA3 ectopic expression: transcriptional and
enhancer elements controlling AGAMOUS expression. They find that two posttranscriptional regulation determine floral organ identity. Cell
LEAFY binding sites in the first intron of AG are necessary for LFY directed 1994, 76:703-716.
AG expression in vivo. Therefore, LFY is formally a direct upstream activatorof AG expression.
35. Krizek BA, Meyerowitz EM: The Arabidopsis homeotic genes
APETALA3 and PISTILLATA are sufficient to provide the B class
18. Wagner D, Sablowski RWM, Meyerowitz EM: Transcriptional
organ identity function. Development 1996, 122:11-22.
activation of APETALA1 by LEAFY. Science 1999, 285:582-584.
By using a steroid-inducible LFY, called 35S::LFY-GR, the authors show that 36. Ingram GC, Goodrich J, Wilkinson MD, Simon R, Haughn GW, early, but not late, expression of APETALA1 results from direct transcrip- Coen ES: Parallels between UNUSUAL FLORAL ORGANS and
FIMBRIATA, genes controlling flower development in Arabidopsis
and Antirrhinum. Plant Cell 1995, 7:1501-1510.
19. Drews GN, Bowman JL, Meyerowitz EM: Negative regulation of the
Long JA, Barton MK: The development of apical embryonic pattern
Arabidopsis homeotic gene AGAMOUS by the APETALA2 product.
in Arabidopsis. Development 1998, 125:3027-3035.
Cell 1991, 65:991-1002.
38. Hill TA, Day CD, Zondlo SC, Thackeray AG, Irish VF: Discrete spatial
20. Schultz EA, Haughn GW: LEAFY, a homeotic gene that regulates
and temporal cis-acting elements regulate transcription of the
inflorescence development in Arabidopsis. Plant Cell 1991,
Arabidopsis floral homeotic gene APETALA3. Development 1998,
3:771-781.
125:1711-1721.
21. Haula E, Sussex IM: LEAFY interacts with floral homeotic genes to
This paper describes molecular dissection of the APETALA3 promoter. The regulate Arabidopsis floral development. Plant Cell 1992,
authors identify regions of the promoter require for petal-specific and sta- 4:901-913.
men-specific expression, and show that AP3, PISTILLATA, UNUSUAL FLO-RAL ORGANS and APETALA1 are required for AP3::GUS expression in 22. Weigel D, Meyerowitz EM: Activation of floral homeotic genes in
the petals. The authors also demonstrate that AP1, AP3, PI and AG bind to Arabidopsis. Science 1993, 261:1723-1726.
three sequence elements, called CArG boxes, present in the AP3 promoter.
See also [39•].
23. Mizukami Y, Ma H: Ectopic expression of the floral homeotic gene
AGAMOUS in transgenic Arabidopsis plants alters floral organ
39. Tilly JJ, Allen DW, Jack T: The CArG boxes in the promoter of the
identity. Cell 1992, 71:119-131.
Arabidopsis floral organ identity gene APETALA3 mediate diverse
regulatory effects. Development 1998, 125:1647-1657.
24. Bowman JL, Smyth DR, Meyerowitz EM: Genetic interactions
To understand how the expression of APETALA3 is regulated, the authors ana- among floral homeotic genes of Arabidopsis. Development 1991,
lyze the AP3 promoter using AP3::GUS fusions. A 496 bp fragment of the AP3 112:1-20.
promoter, or a synthetic AP3 promoter containing three tandem repeats of a Growth and development
143 bp fragment, directs GUS activity in the same spatial and temporal expres- receptor component of the IκBα-ubiquitin ligase. Nature 1998,
sion pattern as the endogenous AP3 gene. Mutations of the three CArG 396:590-594.
(CArG1–CArG3) boxes, which are present in the 143 bp fragment, affectsGUS expression in the context of the synthetic promoter. The authors conclude 45. Bowman JL, Sakai H, Jack T, Weigel D, Mayer U, Meyerowitz EM: that CArG1 binds a positively acting factor(s), CArG2 is required for petal spe- SUPERMAN, a regulator of floral homeotic genes in Arabidopsis.
cific expression, and CArG3 binds a negatively acting factor(s). See also [38•].
Development 1992, 114:599-615.
40. Ingram GC, Doyle S, Carpenter R, Schultz EA, Simon R, Coen ES: 46. Jofuku DK, den Boer BGW, Van Montagu M, Okamuro JK: Control of
Dual role for fimbriata in regulating floral homeotic genes and cell
Arabidopsis flower and seed development by the homeotic gene
division in Antirrhinum. EMBO J 1997, 16:6521-6534.
APETALA2. Plant Cell 1994, 6:1211-1225.
41. Zhang H, Kobayashi R, Galaktionov K, Beach D: p19Skp1 and p45Skp2
Kempin SA, Savidge B, Yanofsky MF: Molecular basis of the
are essential elements of the cyclinA-CDK2 S phase kinase. Cell
cauliflower phenotype in Arabidopsis. Science 1995, 267:522-525.
1995, 82:915-925.
48. Liu Z, Meyerowitz EM: LEUNIG regulates AGAMOUS expression in
42. Bai C, Sen P, Hofmann K, Ma L, Goebl M, Harper JW, Elledge S: SKP1
Arabidopsis. Development 1995, 121:975-991.
connects cell cycle regulators to the ubiquitin proteolysis
machinery through a novel motif, the F-box. Cell 1996, 86:263-274.
49. Sakai H, Medrano LJ, Meyerowitz EM: Role of SUPERMAN in
43. Connelly C, Hieter P: Budding yeast SKP1 encodes an
maintaining Arabidopsis floral whorl boundaries. Nature 1995,
evolutionarily conserved kinetochore protein required for cell
378:199-203.
cycle progression. Cell 1996, 86:275-285.
50. Goodrich J, Puangsomlee P, Martin M, Long D, Meyerowitz EM, 44. Yaron A, Hatzubai A, Davis M, Lavon I, Amit S, Manning AM, Coupland G: A Polycomb-group gene regulates homeotic gene
Andersen JS, Mann M, Mercurio F, Ben-Neriah Y: Identification of the
expression in Arabidopsis. Nature 1997, 386:44-51.

Source: http://jpkc.sdau.edu.cn/zhiwushengwuxue/tuozhan/tuijian/Three%20ways%20to%20learn%20the%20ABCs.pdf