- Email: [email protected]
Plant development: Two sides to organ asymmetry
R756
Dispatch
Plant development: Two sides to organ asymmetry Andrew Hudson
Three gene families have been identified which interact to polarize plant lateral organs. The results suggest that organ polarity is initially determined by a signal from the shoot tip which specifies adaxial organ identity and results in repression of abaxial identity, thereby aligning the polarity of organs with the stem. Address: Institute of Cell and Molecular Biology, University of Edinburgh, Edinburgh, EH9 3JH, UK. E-mail: [email protected] Current Biology 2001, 11:R756–R758 0960-9822/01/$ – see front matter © 2001 Elsevier Science Ltd. All rights reserved.
The polarity of plant shoots and flowers is obvious in their direction of growth and in the morphology of their lateral organs, such as leaves and floral organs. Leaves tend to be flattened perpendicular to the stem axis and may show an asymmetric distribution of cell types, with tissues specialised for light harvesting on the ‘adaxial’ side — facing the shoot tip — and those specialised for gas exchange on the ‘abaxial’ side. Similar adaxial–abaxial polarity can be seen in floral organs and cotyledons. Recent studies [1–4] have revealed three families of genes which act to promote either adaxial or abaxial organ fate. The results suggest that adaxial fate is dependent on signals from the shoot apex which activate proteins needed to repress expression of abaxial genes. Experiments in the 1950s revealed that leaf asymmetry could be disrupted by inserting a barrier into the shoot apical meristem, on the adaxial side of a group of leaf initial cells [5,6]. The resulting leaf primordium grew with radial symmetry and lacked adaxial cell types, suggesting that a signal from the centre of the meristem promoted adaxial organ identity and that abaxial identity occurred by default. A modern interpretation would predict there are genes that specify adaxial fate early in organ development, in response to a signal in the shoot apical meristem, and that repress genes that promote abaxial organ fate. A family of regulatory genes fitting these criteria has now come to light in Arabidopsis. It was initially revealed by the phabulosa-1d (phb-1d) mutation, which causes dosedependent adaxialisation of all lateral organs [7]. This semi-dominance is consistent with PHB activity promoting adaxial fate and phb-1d being a gain-of-function mutation that causes ectopic activity. The recent isolation by McConnell et al. [1] of PHB has provided considerable support for this view. They found that PHB encodes a member of a small subfamily of homeodomain–leucine
zipper transcription factors — the HD-ZIP III proteins — which are know to bind specific DNA target sequences [8]. The other two members of this family in Arabidopsis — PHAVOLUTA (PHV) [1,8] and REVOLUTA (REV) [9,10] — have similar expression patterns to PHB [1,10,11], suggesting related roles. All three genes are normally expressed in the shoot apical meristem, uniformly within the organ initials. But their expression becomes restricted to an adaxial domain of each organ primordium as it initiates from the shoot apical meristem, and later expression is confined to developing vascular tissues. PHB and PHV show an equivalent expression in embryos — initially throughout the apical region of the globular embryo and then only in the shoot apical meristem, an adaxial domain of each developing cotyledon and developing vascular cells [1,10]. Analysis of phb-1d and equivalent phb mutations suggests the basis for their semi-dominance. PHB and the other HD-ZIP III proteins contain a so-called START domain. This motif is present in a range of functionally diverse proteins and in some has been found to bind a lipid ligand [12]. The best characterised version of this motif, in the human steroidogenic acute regulatory (StAR) protein, forms a hydrophobic pocket that binds cholesterol and transports it through the outer mitochondrial membrane [13]. Five independent phb mutations were found to alter PHB’s START domain: two cause an in-frame insertion of the same 10 or 11 amino acids near the protein’s amino terminus, while the others substituted glutamic acid for glycine at the adjacent position. Semi-dominant phv mutations, which have a similar adaxialising effect, were found to cause the equivalent substitution in PHV. One simple explanation for the effects of these mutations is that binding a lipid ligand in adaxial cells activates PHB and PHV proteins. Amino acid insertions or substitutions within the START domain might therefore render the proteins constitutively active, thereby causing them to specify adaxial identity ectopically. Further support for regulation of this kind has come from the finding that ectopic expression of the wildtype PHB gene had no effect, presumably because distribution of the putative ligand remained limiting. Analysis of phb-1d expression also proved informative. Transcripts were found in the shoot apical meristem and organ initials, as in the case of wild-type PHB, but they did not become restricted to adaxial cells as organ primordia initiated. Because phb mutations are likely to affect protein activity, this suggested that active PHB promotes accumulation of PHB RNA. Such autoregulation, coupled with the ability of activated PHB protein to promote availability of its ligand, could provide a mechanism by
Dispatch
which adaxial fate is maintained in growing primordia. It also suggests that PHB RNA levels may reflect the concentration of the activating ligand. PHB transcripts are abundant in the centre of the shoot apical meristem and show graded expression in organ primordia, with highest levels adaxially. Such a pattern suggests the existence of a morphogen which forms a gradient from the centre to the periphery of the shoot apical meristem and specifies adaxial fate in a concentration-dependent manner. It is also consistent with the loss of adaxial fate observed in leaf initials surgically isolated from the centre of the shoot apical meristem, which is proposed to be the source of the morphogen. Although it is appealing to speculate that the potential lipid ligand of PHB might be this morphogen, the ligand might alternatively be produced as a second messenger within cells that have been activated by a different signal. The higher levels of PHB expression detected in shoot apical meristem cells adaxial to organ initials are further suggestive of reverse signalling from organs to the shoot apical meristem. Because the related REV gene is needed to promote axillary shoot apical meristem formation [11,14], and phb-1d mutants have larger shoot apical meristems than wild-type [7], the three HD-ZIP III genes might function redundantly to promote shoot apical meristem activity. Such roles should be revealed in plants carrying multiple loss-of-function mutations of PHV, PHB or REV. The HD-ZIP III genes are not alone in promoting adaxial cell fate in higher plants. For example, PINHEAD (PNH), which encodes a potential transcription elongation factor, is expressed in a similar domain to PHB and promotes adaxial fate and shoot apical meristem activity redundantly with the related AGONAUTE1 gene [15,16]. It should now be possible to test how this role is related to that of the HD-ZIP III genes. On the other side of the organ, two Arabidopsis gene families promote abaxial fate. These include all five members of the YABBY (YAB) family, which encode likely transcription factors. [17–19]. Each organ expresses at least two YAB genes, first in internal organ initials and then in an abaxial domain once primordia initiate. Although this pattern suggested a role in promoting abaxial fate, leaves of plants mutant for the two most similar YAB genes, FILAMENTOUS FLOWER (FIL) and YAB3, showed alterations to abaxial cell types that were considerably less severe than in plants carrying the phb-1d allele. Eshed et al. [2] therefore proposed that additional genes must act independently to promote abaxial fate, and that these might be identified by mutations that enhance yab mutant phenotypes. Their approach led to identification of the KANADI1 (KAN1) gene, which had also been found independently by Kerstetter et al. [3]. A further screen for
R757
enhancers of the kan1 phenotype identified additional genes, including KAN2 [4]. The recent isolation of KAN1 and KAN2 has shown that they encode two of the three Arabidopsis members of a protein subfamily implicated in transcriptional regulation. In wild-type plants, KAN1 expression is detected in an abaxial domain of each organ primordium, consistent with a role in promoting abaxial fate [3]. KAN1 expression in the embryo is established in a domain that includes the abaxial part of the developing cotyledons and periphery of the hypocotyl, the apparent complement of the PHB and REV domain. This embryonic domain of expression also provides further evidence for a relationship between the abaxial side of an organ and the peripheral stem. The enhanced mutant phenotypes produced by combining kan and yab mutations suggested that the genes are expressed independently. Loss of KAN2 activity has no effect in a wild-type background [3], while kan1 mutations alone have subtle and semi-dominant effects on leaf and carpel polarity [4]. Organs of plants mutant for both KAN1 and KAN2 show more severe adaxialisation — leaves are narrow, contain adaxial cell types abaxially and produce adaxial outgrowths consisting of abaxial and marginal cell types. Additional loss of YAB activity in the kan1 kan2 fil yab3 quadruple mutant causes further loss of abaxial fate, so organs approached those of the phb-1d mutant in phenotype. Residual differences were proposed to result from the activity of other genes, possibly including the remaining YAB or KAN family members [4]. Interaction between cells with adaxial and abaxial identities has been suggested to induce the lateral growth of leaves that flattens them perpendicular to the stem axis [20]. The expression of FIL in the abaxial outgrowth of kan1 kan2 mutant leaves, and the loss of these outgrowths with additional loss of FIL and YAB3 activity, led Eshed et al. [4] to propose that the outgrowth resulted from juxtaposition of domains of high and low YAB gene expression. This is consistent with the observed correlation of the boundaries between high and low (or no) YAB expression with outgrowth of the normal leaf blade, and with the reduction in blade growth that results from uniform YAB expression [17]. How the KAN and YAB genes interact is not yet clear, partly because of genetic redundancy. Enhancement of yab mutant phenotypes by kan mutations suggests that KAN and YAB genes act independently. Similarly, the abaxialising effect of ectopic KAN1 or KAN2 expression is not dependent on YAB3 and FIL activity [4].. However, the abaxial expression domain of FIL is reduced in organs of kan1 kan2 mutants [17], consistent with KAN genes promoting YAB gene expression. Similarly, uniform
R758
Current Biology Vol 11 No 18
expression of either KAN1 or KAN2, unlike ectopic YAB expression, appears sufficient for complete abaxialisation of cotyledons [4]. Several other genes that promote abaxial organ fate were also identified by genetic enhancement of yab or kan mutant phenotypes. These included PICKLE (PKL), which is likely to encode part of a chromatin remodelling complex involved in the repression of target genes [2,21]. One explanation for the involvement of PKL, which is consistent with its role in repressing shoot apical meristempromoting homeobox genes in organs [21], is that it maintains organ fate and therefore the correct activity of organ polarity genes. Adaxial and abaxial organ fates appear mutually exclusive: a gain of adaxial identity in phb mutants occurs at the expense of abaxial identity, whereas ectopic KAN gene expression has the opposite effect. This suggests that adaxial and abaxial genes might repress each other’s expression. Although current evidence has been limited by redundancy within gene family, available data generally support this view. Thus, the domain of PHB and REV expression in wild-type embryos appears complementary to that of KAN1 [1,4]; the YAB gene FIL is not expressed in severely adaxialised leaves of phb-1d [17]; and PHV and REV show ectopic abaxial expression in kan1 kan2 fil yab3 mutants [4]. An organ initial cell can express both HDZIPIII and YAB genes, however, suggesting that additional factors that are specific to organ primordia must be needed for mutual repression to occur. What might be the significance of mutual repression? Current evidence supports the view that activation of HD-ZIPIII proteins is the primary step in determination of organ polarity, and repression of YAB and KAN genes is a secondary consequence. Subsequent repression of HDZIPIII proteins by KAN (and possibly YAB) genes might thus serve to reinforce the distinction between adaxial and abaxial domains. Reinforcement might also involve abaxial cells signalling to adaxial cells to promote their identity, as suggested by the mild adaxial defects of kan1 mutants [3]. In this scenario, abaxial fate represents the absence of adaxial specification. The available evidence does not, however, rule out the possibility that abaxial fate is itself specified independently, for example by signals from the periphery of the shoot apical meristem. In this case, mutual repression could act to partition primordium cells to the two alternative fates, as well as to reinforce the distinction between them. Current evidence also cannot determine the extent to which HD-ZIP III genes directly promote adaxial fate, rather than repressing activity of abaxial genes (and vice versa). It has, however, provided the tools to allow these outstanding issues to be investigated.
References 1. McConnell, Emery J, Eshed Y, Bao N, Bowman J, Barton MK: Role of PHABULOSA and PHAVOLUTA in determining radial patterning in shoots. Nature 2001, 411:709-713. 2. Eshed Y, Baum SF, Bowman JL: Distinct mechanisms promote polarity establishment in carpels of Arabidopsis. Cell 1999, 99:199-209. 3. Kerstetter RA, Bollman K, Taylor RA, Bomblies K, Poethig RS: KANADI regulates organ polarity in Arabidopsis. Nature 2001, 411:706-709. 4. Eshed Y, Baum SF, Perea JV, Bowman JL: Establishment of polarity in lateral organs of plants. Curr Biol 2001, 11:1251-1260. 5. Sussex IM: Experiments on the cause of dorsiventrality in leaves. Nature 1954, 174:351-352. 6. Sussex IM: Morphogenesis in Solanum tuberosum L.: experimental investigation of leaf dorsiventrality and orientation in the juvenile shoot. Phytomorphology 1955, 5:286-300. 7. McConnell JR, Barton MK: Leaf polarity and meristem formation in Arabidopsis. Development 1998, 125:2935-2942. 8. Sessa G, Steindler C, Morelli G, Ruberti I: The Arabidopsis Athb-8, –9 and –14 genes are members of a small family coding for highly related HD-ZIP proteins. Plant Mol Biol 1998, 38:609-622. 9. Ratcliffe OJ, Riechman JL, Zhang JZ: INTERFASCICULAR FIBERLESS1 is the same gene as REVOLUTA. Plant Cell 2000, 12:315-317. 10. Zhong R, Ye ZH: IFL1, a gene regulating interfascicular fibre differentiation in Arabidopsis, encodes a homeodomain-leucine supper protein. Plant Cell 1999, 11:2139-2152. 11. Otsuga D, DeGuzman B, Prigge MJ, Drews GN, Clark SE: REVOLUTA regulates meristem initiation at lateral positions. Plant J 2001, 25:223-236. 12. Ponting CP, Aravind L: START: a lipid-binding domain in StAR, HD-ZIP and signalling proteins. Trends Biochem Sci 1999, 24:130-132. 13. Tsujishita Y, Hurley JH: Structure and lipid transport mechanism of a StAR-related domain. Nat Struct Biol 2000, 7:408-414. 14. Talbert PB, Alder HT, Parks DW, Comai L: The REVOLUTA gene is necessary for apical meristem development and for limiting cell divisions in the leaves and stems of Arabidopsis thaliana. Development 1995, 121:2723-2735. 15. Lynn K, Fernandez A, Aida M, Sedbrook J, Tasaka M, Masson P, Barton MK: The PINHEAD/ZWILLE gene acts pleiotropically in Arabidopsis and has overlapping functions with the ARGONAUTE1 gene. Development 1999, 126:1-13. 16. Moussian B, Schoof H, Haecker A, Jürgens G, Laux T: Role of the ZWILLE gene in the regulation of central shoot meristem cell fate during Arabidopsis embryogenesis. EMBO J 1998, 17:1799-1809. 17. Siegfried KR, Eshed Y, Baum S, Otsuga D, Drews GN, Bowman JL: Members of the YABBY gene family specify abaxial cell fate in Arabidopsis. Development 1999, 126:4117-4128. 18. Sawa S, Watanabe K, Goto K, Kanaya E, Morita EH, Okada K: FILAMENTOUS FLOWER, a meristem and organ identity gene of Arabidopsis, encodes a protein with a zinc finger and HMGrelated domains. Genes Dev 1999, 13:1079-1088. 19. Bowman JL, Smyth DR: CRABS CLAW, a gene that regulates carpel and nectary development in Arabidopsis, encodes a novel protein with zinc finger and helix-loop-helix domains. Development 1999, 126:2387-2396. 20. Waites R, Hudson A: phantastica: a gene required for dorsoventrality of leaves in Antirrhinum majus. Development 1995, 121:2143-2154. 21. Ogas J, Cheng J, Sung ZR, Somerville C: Cellular differentiation regulated by gibberellin in the Arabidopsis thaliana pickle mutant. Science 1997, 277:91-94. 22. Ori N, Eshed Y, Chuck G, Bowman JL, Hake S: Mechanisms that control knox gene expression in the Arabidopsis shoot. Development 2000, 127:5523-5532.
Dispatch
Plant development: Two sides to organ asymmetry Andrew Hudson
Three gene families have been identified which interact to polarize plant lateral organs. The results suggest that organ polarity is initially determined by a signal from the shoot tip which specifies adaxial organ identity and results in repression of abaxial identity, thereby aligning the polarity of organs with the stem. Address: Institute of Cell and Molecular Biology, University of Edinburgh, Edinburgh, EH9 3JH, UK. E-mail: [email protected] Current Biology 2001, 11:R756–R758 0960-9822/01/$ – see front matter © 2001 Elsevier Science Ltd. All rights reserved.
The polarity of plant shoots and flowers is obvious in their direction of growth and in the morphology of their lateral organs, such as leaves and floral organs. Leaves tend to be flattened perpendicular to the stem axis and may show an asymmetric distribution of cell types, with tissues specialised for light harvesting on the ‘adaxial’ side — facing the shoot tip — and those specialised for gas exchange on the ‘abaxial’ side. Similar adaxial–abaxial polarity can be seen in floral organs and cotyledons. Recent studies [1–4] have revealed three families of genes which act to promote either adaxial or abaxial organ fate. The results suggest that adaxial fate is dependent on signals from the shoot apex which activate proteins needed to repress expression of abaxial genes. Experiments in the 1950s revealed that leaf asymmetry could be disrupted by inserting a barrier into the shoot apical meristem, on the adaxial side of a group of leaf initial cells [5,6]. The resulting leaf primordium grew with radial symmetry and lacked adaxial cell types, suggesting that a signal from the centre of the meristem promoted adaxial organ identity and that abaxial identity occurred by default. A modern interpretation would predict there are genes that specify adaxial fate early in organ development, in response to a signal in the shoot apical meristem, and that repress genes that promote abaxial organ fate. A family of regulatory genes fitting these criteria has now come to light in Arabidopsis. It was initially revealed by the phabulosa-1d (phb-1d) mutation, which causes dosedependent adaxialisation of all lateral organs [7]. This semi-dominance is consistent with PHB activity promoting adaxial fate and phb-1d being a gain-of-function mutation that causes ectopic activity. The recent isolation by McConnell et al. [1] of PHB has provided considerable support for this view. They found that PHB encodes a member of a small subfamily of homeodomain–leucine
zipper transcription factors — the HD-ZIP III proteins — which are know to bind specific DNA target sequences [8]. The other two members of this family in Arabidopsis — PHAVOLUTA (PHV) [1,8] and REVOLUTA (REV) [9,10] — have similar expression patterns to PHB [1,10,11], suggesting related roles. All three genes are normally expressed in the shoot apical meristem, uniformly within the organ initials. But their expression becomes restricted to an adaxial domain of each organ primordium as it initiates from the shoot apical meristem, and later expression is confined to developing vascular tissues. PHB and PHV show an equivalent expression in embryos — initially throughout the apical region of the globular embryo and then only in the shoot apical meristem, an adaxial domain of each developing cotyledon and developing vascular cells [1,10]. Analysis of phb-1d and equivalent phb mutations suggests the basis for their semi-dominance. PHB and the other HD-ZIP III proteins contain a so-called START domain. This motif is present in a range of functionally diverse proteins and in some has been found to bind a lipid ligand [12]. The best characterised version of this motif, in the human steroidogenic acute regulatory (StAR) protein, forms a hydrophobic pocket that binds cholesterol and transports it through the outer mitochondrial membrane [13]. Five independent phb mutations were found to alter PHB’s START domain: two cause an in-frame insertion of the same 10 or 11 amino acids near the protein’s amino terminus, while the others substituted glutamic acid for glycine at the adjacent position. Semi-dominant phv mutations, which have a similar adaxialising effect, were found to cause the equivalent substitution in PHV. One simple explanation for the effects of these mutations is that binding a lipid ligand in adaxial cells activates PHB and PHV proteins. Amino acid insertions or substitutions within the START domain might therefore render the proteins constitutively active, thereby causing them to specify adaxial identity ectopically. Further support for regulation of this kind has come from the finding that ectopic expression of the wildtype PHB gene had no effect, presumably because distribution of the putative ligand remained limiting. Analysis of phb-1d expression also proved informative. Transcripts were found in the shoot apical meristem and organ initials, as in the case of wild-type PHB, but they did not become restricted to adaxial cells as organ primordia initiated. Because phb mutations are likely to affect protein activity, this suggested that active PHB promotes accumulation of PHB RNA. Such autoregulation, coupled with the ability of activated PHB protein to promote availability of its ligand, could provide a mechanism by
Dispatch
which adaxial fate is maintained in growing primordia. It also suggests that PHB RNA levels may reflect the concentration of the activating ligand. PHB transcripts are abundant in the centre of the shoot apical meristem and show graded expression in organ primordia, with highest levels adaxially. Such a pattern suggests the existence of a morphogen which forms a gradient from the centre to the periphery of the shoot apical meristem and specifies adaxial fate in a concentration-dependent manner. It is also consistent with the loss of adaxial fate observed in leaf initials surgically isolated from the centre of the shoot apical meristem, which is proposed to be the source of the morphogen. Although it is appealing to speculate that the potential lipid ligand of PHB might be this morphogen, the ligand might alternatively be produced as a second messenger within cells that have been activated by a different signal. The higher levels of PHB expression detected in shoot apical meristem cells adaxial to organ initials are further suggestive of reverse signalling from organs to the shoot apical meristem. Because the related REV gene is needed to promote axillary shoot apical meristem formation [11,14], and phb-1d mutants have larger shoot apical meristems than wild-type [7], the three HD-ZIP III genes might function redundantly to promote shoot apical meristem activity. Such roles should be revealed in plants carrying multiple loss-of-function mutations of PHV, PHB or REV. The HD-ZIP III genes are not alone in promoting adaxial cell fate in higher plants. For example, PINHEAD (PNH), which encodes a potential transcription elongation factor, is expressed in a similar domain to PHB and promotes adaxial fate and shoot apical meristem activity redundantly with the related AGONAUTE1 gene [15,16]. It should now be possible to test how this role is related to that of the HD-ZIP III genes. On the other side of the organ, two Arabidopsis gene families promote abaxial fate. These include all five members of the YABBY (YAB) family, which encode likely transcription factors. [17–19]. Each organ expresses at least two YAB genes, first in internal organ initials and then in an abaxial domain once primordia initiate. Although this pattern suggested a role in promoting abaxial fate, leaves of plants mutant for the two most similar YAB genes, FILAMENTOUS FLOWER (FIL) and YAB3, showed alterations to abaxial cell types that were considerably less severe than in plants carrying the phb-1d allele. Eshed et al. [2] therefore proposed that additional genes must act independently to promote abaxial fate, and that these might be identified by mutations that enhance yab mutant phenotypes. Their approach led to identification of the KANADI1 (KAN1) gene, which had also been found independently by Kerstetter et al. [3]. A further screen for
R757
enhancers of the kan1 phenotype identified additional genes, including KAN2 [4]. The recent isolation of KAN1 and KAN2 has shown that they encode two of the three Arabidopsis members of a protein subfamily implicated in transcriptional regulation. In wild-type plants, KAN1 expression is detected in an abaxial domain of each organ primordium, consistent with a role in promoting abaxial fate [3]. KAN1 expression in the embryo is established in a domain that includes the abaxial part of the developing cotyledons and periphery of the hypocotyl, the apparent complement of the PHB and REV domain. This embryonic domain of expression also provides further evidence for a relationship between the abaxial side of an organ and the peripheral stem. The enhanced mutant phenotypes produced by combining kan and yab mutations suggested that the genes are expressed independently. Loss of KAN2 activity has no effect in a wild-type background [3], while kan1 mutations alone have subtle and semi-dominant effects on leaf and carpel polarity [4]. Organs of plants mutant for both KAN1 and KAN2 show more severe adaxialisation — leaves are narrow, contain adaxial cell types abaxially and produce adaxial outgrowths consisting of abaxial and marginal cell types. Additional loss of YAB activity in the kan1 kan2 fil yab3 quadruple mutant causes further loss of abaxial fate, so organs approached those of the phb-1d mutant in phenotype. Residual differences were proposed to result from the activity of other genes, possibly including the remaining YAB or KAN family members [4]. Interaction between cells with adaxial and abaxial identities has been suggested to induce the lateral growth of leaves that flattens them perpendicular to the stem axis [20]. The expression of FIL in the abaxial outgrowth of kan1 kan2 mutant leaves, and the loss of these outgrowths with additional loss of FIL and YAB3 activity, led Eshed et al. [4] to propose that the outgrowth resulted from juxtaposition of domains of high and low YAB gene expression. This is consistent with the observed correlation of the boundaries between high and low (or no) YAB expression with outgrowth of the normal leaf blade, and with the reduction in blade growth that results from uniform YAB expression [17]. How the KAN and YAB genes interact is not yet clear, partly because of genetic redundancy. Enhancement of yab mutant phenotypes by kan mutations suggests that KAN and YAB genes act independently. Similarly, the abaxialising effect of ectopic KAN1 or KAN2 expression is not dependent on YAB3 and FIL activity [4].. However, the abaxial expression domain of FIL is reduced in organs of kan1 kan2 mutants [17], consistent with KAN genes promoting YAB gene expression. Similarly, uniform
R758
Current Biology Vol 11 No 18
expression of either KAN1 or KAN2, unlike ectopic YAB expression, appears sufficient for complete abaxialisation of cotyledons [4]. Several other genes that promote abaxial organ fate were also identified by genetic enhancement of yab or kan mutant phenotypes. These included PICKLE (PKL), which is likely to encode part of a chromatin remodelling complex involved in the repression of target genes [2,21]. One explanation for the involvement of PKL, which is consistent with its role in repressing shoot apical meristempromoting homeobox genes in organs [21], is that it maintains organ fate and therefore the correct activity of organ polarity genes. Adaxial and abaxial organ fates appear mutually exclusive: a gain of adaxial identity in phb mutants occurs at the expense of abaxial identity, whereas ectopic KAN gene expression has the opposite effect. This suggests that adaxial and abaxial genes might repress each other’s expression. Although current evidence has been limited by redundancy within gene family, available data generally support this view. Thus, the domain of PHB and REV expression in wild-type embryos appears complementary to that of KAN1 [1,4]; the YAB gene FIL is not expressed in severely adaxialised leaves of phb-1d [17]; and PHV and REV show ectopic abaxial expression in kan1 kan2 fil yab3 mutants [4]. An organ initial cell can express both HDZIPIII and YAB genes, however, suggesting that additional factors that are specific to organ primordia must be needed for mutual repression to occur. What might be the significance of mutual repression? Current evidence supports the view that activation of HD-ZIPIII proteins is the primary step in determination of organ polarity, and repression of YAB and KAN genes is a secondary consequence. Subsequent repression of HDZIPIII proteins by KAN (and possibly YAB) genes might thus serve to reinforce the distinction between adaxial and abaxial domains. Reinforcement might also involve abaxial cells signalling to adaxial cells to promote their identity, as suggested by the mild adaxial defects of kan1 mutants [3]. In this scenario, abaxial fate represents the absence of adaxial specification. The available evidence does not, however, rule out the possibility that abaxial fate is itself specified independently, for example by signals from the periphery of the shoot apical meristem. In this case, mutual repression could act to partition primordium cells to the two alternative fates, as well as to reinforce the distinction between them. Current evidence also cannot determine the extent to which HD-ZIP III genes directly promote adaxial fate, rather than repressing activity of abaxial genes (and vice versa). It has, however, provided the tools to allow these outstanding issues to be investigated.
References 1. McConnell, Emery J, Eshed Y, Bao N, Bowman J, Barton MK: Role of PHABULOSA and PHAVOLUTA in determining radial patterning in shoots. Nature 2001, 411:709-713. 2. Eshed Y, Baum SF, Bowman JL: Distinct mechanisms promote polarity establishment in carpels of Arabidopsis. Cell 1999, 99:199-209. 3. Kerstetter RA, Bollman K, Taylor RA, Bomblies K, Poethig RS: KANADI regulates organ polarity in Arabidopsis. Nature 2001, 411:706-709. 4. Eshed Y, Baum SF, Perea JV, Bowman JL: Establishment of polarity in lateral organs of plants. Curr Biol 2001, 11:1251-1260. 5. Sussex IM: Experiments on the cause of dorsiventrality in leaves. Nature 1954, 174:351-352. 6. Sussex IM: Morphogenesis in Solanum tuberosum L.: experimental investigation of leaf dorsiventrality and orientation in the juvenile shoot. Phytomorphology 1955, 5:286-300. 7. McConnell JR, Barton MK: Leaf polarity and meristem formation in Arabidopsis. Development 1998, 125:2935-2942. 8. Sessa G, Steindler C, Morelli G, Ruberti I: The Arabidopsis Athb-8, –9 and –14 genes are members of a small family coding for highly related HD-ZIP proteins. Plant Mol Biol 1998, 38:609-622. 9. Ratcliffe OJ, Riechman JL, Zhang JZ: INTERFASCICULAR FIBERLESS1 is the same gene as REVOLUTA. Plant Cell 2000, 12:315-317. 10. Zhong R, Ye ZH: IFL1, a gene regulating interfascicular fibre differentiation in Arabidopsis, encodes a homeodomain-leucine supper protein. Plant Cell 1999, 11:2139-2152. 11. Otsuga D, DeGuzman B, Prigge MJ, Drews GN, Clark SE: REVOLUTA regulates meristem initiation at lateral positions. Plant J 2001, 25:223-236. 12. Ponting CP, Aravind L: START: a lipid-binding domain in StAR, HD-ZIP and signalling proteins. Trends Biochem Sci 1999, 24:130-132. 13. Tsujishita Y, Hurley JH: Structure and lipid transport mechanism of a StAR-related domain. Nat Struct Biol 2000, 7:408-414. 14. Talbert PB, Alder HT, Parks DW, Comai L: The REVOLUTA gene is necessary for apical meristem development and for limiting cell divisions in the leaves and stems of Arabidopsis thaliana. Development 1995, 121:2723-2735. 15. Lynn K, Fernandez A, Aida M, Sedbrook J, Tasaka M, Masson P, Barton MK: The PINHEAD/ZWILLE gene acts pleiotropically in Arabidopsis and has overlapping functions with the ARGONAUTE1 gene. Development 1999, 126:1-13. 16. Moussian B, Schoof H, Haecker A, Jürgens G, Laux T: Role of the ZWILLE gene in the regulation of central shoot meristem cell fate during Arabidopsis embryogenesis. EMBO J 1998, 17:1799-1809. 17. Siegfried KR, Eshed Y, Baum S, Otsuga D, Drews GN, Bowman JL: Members of the YABBY gene family specify abaxial cell fate in Arabidopsis. Development 1999, 126:4117-4128. 18. Sawa S, Watanabe K, Goto K, Kanaya E, Morita EH, Okada K: FILAMENTOUS FLOWER, a meristem and organ identity gene of Arabidopsis, encodes a protein with a zinc finger and HMGrelated domains. Genes Dev 1999, 13:1079-1088. 19. Bowman JL, Smyth DR: CRABS CLAW, a gene that regulates carpel and nectary development in Arabidopsis, encodes a novel protein with zinc finger and helix-loop-helix domains. Development 1999, 126:2387-2396. 20. Waites R, Hudson A: phantastica: a gene required for dorsoventrality of leaves in Antirrhinum majus. Development 1995, 121:2143-2154. 21. Ogas J, Cheng J, Sung ZR, Somerville C: Cellular differentiation regulated by gibberellin in the Arabidopsis thaliana pickle mutant. Science 1997, 277:91-94. 22. Ori N, Eshed Y, Chuck G, Bowman JL, Hake S: Mechanisms that control knox gene expression in the Arabidopsis shoot. Development 2000, 127:5523-5532.