Apterous

Lim homeodomain proteins of fish, frogs and birds

One of the first intercellular signalling events in the vertebrate embryo leads to mesoderm formation and axis determination. In the mouse, a gene encoding a new member of the TGF-beta superfamily, nodal, is disrupted in a mutant deficient in mesoderm formation. nodal mRNA is found in prestreak mouse embryos, consistent with a role in the development of the dorsal axis. Injection of nodal mRNA into zebrafish embryos causes the formation of ectopic axes that included notochord and somites. Axis duplication was preceded by the generation of an apparent ectopic shield (organizer equivalent) in nodal-injected embryos, as indicated by the appearance of a region over-expressing gsc and lim1. These results suggest a role for a nodal-like factor in pattern formation in zebrafish (Toyama, 1995b).

LIM homeodomain proteins are developmental regulators whose functions depend on synergism with LIM domain binding proteins (Ldb proteins); they are homologs of Drosophila Chip. Three members of the ldb gene family (Ldb1, Ldb2 and Ldb3) from the zebrafish, Danio rerio, share 95%, 73% and 62% amino acid identity with mouse Ldb1, respectively. In overlay assays, Ldb proteins bind LIM homeodomain proteins and LMO1, but not zyxin or MLP. Whole mount in situ hybridization showed that zebrafish ldb1 is expressed ubiquitously from gastrulation onward. Ldb2 is ubiquitous at gastrulation, and later is found in many tissues, especially the anterior central nervous system (CNS) and vasculature, but not all tissues; Ldb3 mRNA is expressed primarily in the anterior CNS. The expression of individual lbd and LIM proteins correlates in various regions and stages of embryogenesis. For example, lbd1 and ldh2 are expressed in the shield as are lim1 and lim6. Both lbd2 and lbd3 expression in the telencephalon overlap with that of lim5 and lim6, and three genes, lbd2, lim3 and lim5, are coexpressed in the epiphysis (Toyama 1998).

In zebrafish, individual primary motoneurons can be uniquely identified by their characteristic cell body positions and axonal projection patterns. The fate of individual primary motoneurons remains plastic until just prior to axogenesis when they become committed to particular identities. Distinct primary motoneurons express particular combinations of LIM homeobox genes. Expression precedes axogenesis as well as commitment, suggesting that LIM homeobox genes may contribute to the specification of motoneuronal fates. By transplanting them to new spinal cord positions, it can be demonstrated that primary motoneurons can initiate a new program of LIM homeobox gene expression, as well as the morphological features appropriate for the new position. The patterned distribution of different primary motoneuronal types within the zebrafish spinal cord follows the patterned expression of LIM homeobox genes, and this reflects a highly resolved system of positional information controlling gene transcription (Appel, 1995).

LIM class homeobox genes code for a family of transcriptional regulators that encode important determinants of cell lineage and cell type specificity. The lim3 gene from the zebrafish, Danio rerio, is highly conserved in sequence and expression pattern, as compared to its homologs in other vertebrates. Immunocytochemical analysis of Lim3 protein expression was performed in the pituitary, pineal, hindbrain, and spinal cord of the embryo, revealing an asymmetrical, lateral and late program of pituitary development in zebrafish, distinct from the pattern in other vertebrates. Lim3 expression was studied in no tail, floating head, and cyclops mutant embryos, all of which have midline defects, with special reference to spinal cord differentiation (where Lim3 marks mostly motoneurons). cyclops embryos show essentially normal Lim3 expression in the hindbrain and spinal cord despite the absence of the floor plate, while no tail (Drosophila homolog: brachyenteron) mutant embryos, which lack a differentiated notochord, display an excess of Lim3-expressing cells in a generally normal pattern. In contrast, Lim3-positive cells largely disappear from the posterior spinal cord in floating head (a homeodomain protein) mutants, except in patches that correlate with remnants of apparent floor plate cells. These results support the view that either notochord or floor plate signaling can specify Lim3-positive motoneurons in the spinal cord (Glasgow, 1997).

The results of recent studies have supported the idea that the ability to organize the formation of axes such as the anteroposterior and proximodistal axes corresponds to the limb regeneration ability in Xenopus. This study investigated the mechanism by which the dorsoventral (D-V) axis of regenerating Xenopus limbs is established and the relationships between D-V patterning and regenerative ability. Drosophila apterous, which specifies dorsal fate and organizes distal outgrowth of the wing during Drosophila wing development, has at least two vertebrate homologs: Lmx-1 and Lhx2. Two functions that apterous has are divided into these two genes: Lmx-1 is necessary and sufficient to specify dorsal fate, and Lhx2 is involved in limb outgrowth. These two vertebrate homologs share high sequence similarity with apterous (44% and 56% identity in the LIM domain to apterous, respectively. Therefore, based on sequence similarity alone, it was not possible to label the clone that was obtained from Xenopus as an amphibian homolog of Lmx-1. The expression profile of this clone strongly suggests that it is actually a partial clone of the Xenopus Lmx-1 homolog, and it has been concluded that this clone represents Xenopus Lmx-1. As in chick and mouse limb development, the Lmx-1 transcript starts and continues to be expressed throughout the dorsal mesenchyme in stage 51-53 limb buds of Xenopus tadpoles. At stage 55, this expression is restricted to the dorsal part of the presumptive digit region. The dorsal-sided expression suggests that Lmx-1 plays an important role in D-V patterning also in Xenopus limb development. In regenerating limb buds, Lmx-1 expression is first detectable in the dorsal mesenchyme of the blastema at 3 days after amputation of stage 52 limb buds. Reexpression of Lmx-1 also occurs in stage 55 blastemas, but it takes 7 days after amputation even when they are amputated at the same level (ankle level) as the stage 52 limb buds. Furthermore, Lmx-1 expression in the froglet blastema is undetectable for 14 days after amputation at the wrist level. These findings suggest that D-V axis formation in later stage limb buds, which have reduced regenerative ability, could be delayed. This delay may be because of the absence of epidermal signals directing D-V patterning. In any case, these results suggest that a signaling pathway mediated by Lmx-1 is involved in D-V patterning during limb regeneration (Matsuda, 2001).

Transplantation experiments were performed to study which epidermis or mesenchyme is responsible for the D-V patterning in regenerating limbs. Naked mesenchyme of a donor limb was rotated and implanted on a host opposite-side limb stump to make a reversed recombination about the D-V axis. The resultant regenerates had a normal-looking D-V pattern, including Lmx-1 expression, muscle pattern, and joints, in stage 52 recombinants and a reversed D-V pattern in stage 55 recombinants. Further experiments in recombination at stage 52 and stage 55 showed that the epidermal signal is responsible for producing the D-V pattern in the regenerating blastema. These results, together with the finding that Lmx-1 expression is absent in the froglet forelimb blastema, suggest that D-V axis formation is a key step in understanding the loss of regenerative ability (Matsuda, 2001).

Anteroposterior patterning of neural tissue is thought to be directed by the axial mesoderm, which is functionally divided into head (or precordal) and trunk organizer (notochord). In Xenopus the homeobox genes goosecoid (Drosophila homolog: Goosecoid) and Otx2 (Drosophila homolog: Orthodenticle) are expressed in the pre-cordal mesoderm; the LIM class homeobox gene Xlim-1 is expressed in the entire axial mesoderm, whereas the distinct Brachyury related transcription factor Xbra (Drosophila homolog: T-related gene) is expressed in the notochord but not in the procordal mesoderm. Messenger RNA injection experiments show that Xenopus animal pole explants (caps) expressing an activated form of Xlim-1 (a LIM domain mutant named 3m) induce anterior neural markers, whereas caps coexpressing Xlim-1/3m and Xbra induce posterior neural markers. These data indicate that in terms of neural inducing ability, Xlim-1/3m-expressing caps correspond to the head organizer and Xlim-1/3m plus Xbra-coexpressing caps to the trunk organizer. Thus the expression domains of Xlim-1 and Xbra correlate with, and possibly define, the functional domains of the organizer. In animal caps Xlim-1/3m initiates expression of a neuralizing factor chordin (Drosophila homolog: Short gastrulation, which counteracts the antineurogenic effects of Decapentaplegic), whereas Xbra activates embryonic fibroblast growth factor (eFGF expression) (See Drosophila FGF homolog Branchless); these factors could mediate the neural inducing and patterning effects that are observed. A dominant-negative FGF receptor (XFD) inhibits posteriorization by Xbra in a dose-dependent manner, supporting the suggestion that eFGF or a related factor has posteriorizing influence. Retinoic acid, postulated to be a posteriorizing factor based on the observations that RA treatment of embryos leads to truncation of anterior structures in Xenopus, can posteriorize neural tissue generated by Xlim-1. RA strongly inhibits Otx2 expression and induces Krox-20 and beta2-tubulin expression, indicating that RA can act as a posteriorizing factor for neural tissue in the absence of mesoderm (Taira, 1997).

The Xlim-1 gene is activated in the late blastula stage of Xenopus embryogenesis in the mesoderm; its RNA product becomes concentrated in the Spemann organizer at early gastrula stage. A major regulator of early expression of Xlim-1 is activin or an activin-like signal. The 5' flanking region of Xlim-1 contains a constitutive promoter that is not activin responsive, whereas sequences in the first intron mediate repression of basal promoter activity and stimulation by activin. An intron-derived fragment of 212 nt is the smallest element that can mediate activin responsiveness. Nodal and act-Vg1, factors with signaling properties similar to activin, also stimulate Xlim-1 reporter constructs, whereas BMP-4 neither stimulates nor represses the constructs. The mechanism of activin regulation of Xlim-1 and the sequence of the response element are distinct from the activin response elements of other genes studied to date (Rebbert,1997).

Recently, a model to explain the mechanism of Xenopus tail bud formation has been proposed. The NMC model proposes that three regions around the late blastopore lip are required to initiate tail formation. These are the posterior-most neural plate, fated to form tail somites (M), the neural plate (N), immediately anterior to M, and the underlying caudal notochord (C). To initiate tail formation, C must underlie (and presumably signal to) the junction of N and M, which subsequently forms the tip of the tail. During normal development, the NMC interaction leading to specification of the tail bud occurs at the end of gastrulation. Outgrowth of the tail bud commences much later, becoming clearly visible by stage 30 (Beck, 1998 and references).

Several domains of the Xenopus tail bud are defined by two phases of gene expression. The first group of genes are already expressed in the tail bud region before its determination at stage 13 and are subsequently restricted in the extending tail bud by stage 30. This group, the early genes, includes the Notch ligand X-delta-1, the lim domain homeobox factor Xlim1, the T-box factor Xbra, and the homeobox factor Xnot2 and Xcad2, a member of the caudal family. X-delta-1 is expressed specifically in the posterior wall of the neuroenteric canal but is excluded from the chordoneural hinge at stage 30, thus maintaining its earlier expression in the lateral and ventral blastopore lips. Xim1 is expressed in the notochord and dorsal blastopore lip at the end of gastrulation, and is maintained in the chordoneural hinge and posterior tip of the differentiated notochord in later stages. Xnot2 is expressed in the ventral neural tube and chordoneural hinge, but not in the posterior notochord. The posterior notochord therefore represents a novel tail bud region by stage 30, marked by Xlim but not Xnot transcripts, whereas the posterior ventral neural tube is marked by Xnot but not Xbra or Xlim1. Xbra is expressed in the chordoneural hinge and posterior wall. Xcad3 expression in the posterior neural plate is later maintained in the posterior wall and posterior dorsal neural tube. Xpo is expressed in all tissues of the tail bud with the exception of the chordoneural hinge, and is expressed in the fin and epidermis (Beck 1998).

Unlike the early genes, the regional expression of the second group of genes in the extended tail bud can not be traced back to the stage of tail bud initiation. These genes have a late onset of localized expression in the tail bud, corresponding to the beginning of tail outgrowth, although they may be expressed elsewhere in the embryo at stage 13. The dorsal roof domain of the tail bud is marked by expression of Xwnt3a and lunatic fringe. Xwnt5a expression is restricted to the tail bud roof. The distal tip of the tail, which comprises part of the posterior wall, is marked by expression of Xhox3, which marks the distal cells of the tail bud. Xhox3 is a vertebrate homolog of Drosophila evenskipped. Other late genes include BMP-4, X-serrate-1 and BMP-2 (Beck, 1998).

The existence of distinct domains in the positions predicted for C and M is proposed. The restriction of Xcad3 and Xlim1 transcripts to the posterior of the notochord in the early neurula demonstrates that the posterior part of the notochord differs from the crest, corresponding to the C region. Novel domains of the tail bud are proposed to express different combinations of genes. These domains include the dorsal roof of the tail bud, the distal tip of the tail, marked by Xhox3, the chordoneural hinge, the posterior tip of the chordoneural hinge, the posterior wall domain, the tip of the posterior wall, the posterior notochord, the posterior wall of the neuroenteric canal and the ventral neural tube (Beck, 1998).

The Xenopus LIM homeodomain protein Xlim-1 is specifically expressed in the Spemann organizer region and assumed to play a role in the establishment of the body axis as a transcriptional activator. To further elucidate the mechanism underlying the regulation of its transcriptional activity, a focus was placed on the region C-terminal to the homeodomain of Xlim-1 (CT239-403), dividing it into five regions, [CCR1-5 (C-terminal conserved regions)], based on similarity between Xlim-1 and its paralog, Xlim-5. The role of Xlim-1 CT239-403 in the Spemann organizer was analyzed by assaying the axis-forming ability of a series of CCR-mutated constructs in Xenopus embryos. High doses of Xlim-1 constructs deleted of CCR1 or CCR2 initiate secondary axis formation in the absence of its coactivator Ldb1 (LIM-domain-binding protein 1), suggesting that CCR1 and CCR2 are involved in negative regulation of Xlim-1. In contrast, while Xlim-1 is capable of initiating secondary axis formation at low doses in the presence of Ldb1, deletion of CCR2 (aa 275-295) or substitution of five conserved tyrosines in CCR2 with alanines (CCR2-5YA) abolishes the activity. In addition, UAS-GAL4 one-hybrid reporter assays in Xenopus show that CCR2, but not CCR2-5YA, with its flanking regions (aa 261-315) functions as a transactivation domain when fused to the GAL4 DNA-binding domain. Finally, none of the known transcriptional coactivators tested (CBP, SRC-1, and TIF2) interacts with the Xlim-1 transactivation domain (aa 261-315). Thus, Xlim-1 not only contains a unique tyrosine-rich activation domain but also contains a negative regulatory domain in CT239-403, suggesting a complex regulatory mechanism underlying the transcriptional activity of Xlim-1 in the organizer (Hiratani, 2001).

Xlim-5 in Xenopus and lim5 in the zebrafish are highly similar in sequence but quite distinct in expression pattern from the previously described Xlim-1/lim1 gene. In both species studied the lim5 gene is expressed in the entire ectoderm in the early gastrula embryo. The Xlim-5 gene is activated in a cell autonomous manner in ectodermal cells, and this activation is suppressed by the mesoderm inducer activin. During neurulation, expression of the lim5 gene in both the frog and fish embryo is rapidly restricted to an anterior region in the developing neural plate/keel. In the 2-day Xenopus and 24-hr zebrafish embryo, this region becomes more sharply defined, forming a strongly lim5-expressing domain in the diencephalon anterior to the midbrain-forebrain boundary. In addition, regions of less intense lim5 expression are seen in the zebrafish embryo in parts of the telencephalon, in the anterior diencephalon coincident with the postoptic commissure, and in restricted regions of the midbrain, hindbrain, and spinal cord. Expression in ventral forebrain is abolished from the 5-somite stage onward in cyclops mutant fish. These results imply a role for lim5 in the patterning of the nervous system, in particular in the early specification of the diencephalon (Toyama, 1995a).

The predicted Xenopus Xlim-3 mRNA is detected in dorsal regions at neural tube and tailbud stages and in adults predominantly in the pituitary gland and weakly in the eye and brain. Whole mount in situ hybridization revealed that Xlim-3 mRNA is first detectable at the neural plate stage in the stomodeal-hypophyseal (pituitary) anlage and in the neural plate where labeled cells were found adjacent to the forming floor plate. In situ hybridization analysis on serial sections at later stages shows that embryonic Xlim-3 expression persists in the pituitary and pineal, as well as in some cells of the retina, hindbrain, and spinal cord. In the retina, Xlim-3 mRNA was detected only in a distinct sublamina of the inner nuclear layer, but not in dividing cells of ciliary margin. This discrete manner of Xlim-3 expression, especially persistent expression in the pituitary (before morphogenesis of the gland to adult), supports a role in the specification and maintenance of differentiation of distinct neuronal and neuroendocrine tissues (Taira, 1993).

A new LIM-domain-binding factor, Ldb1, an novel protein, has been isolated on the basis of its ability to interact with the LIM-HD protein Lhx1 (Lim1). High-affinity binding by Ldb1 requires paired LIM domains and is restricted to the related subgroup of LIM domains found in LIM-HD and LMO proteins (See Drosophila Muscle LIM ptotein at 60A). The highly conserved Xenopus Ldb protein XLdb1, interacts with Xlim-1, the Xenopus orthologue of Lhx1. When injected into Xenopus embryos, XLdb1 (or Ldb1) can synergize with Xlim-1 in the formation of partial secondary axes and in activation of the genes encoding goosecoid, chordin, NCAM and XCG7, demonstrating a functional as well as a physical interaction between the two proteins (Agulnick, 1996).

The chicken homeobox genes LH-2A and LH-2B that encode two related LIM domain-containing homeodomain proteins have been isolated and their expression pattern during chick limb development examined. LH-2A is most closely related to human and rat LH-2, while LH-2B is less well conserved. Although both LH-2A and LH-2B are expressed in the limb mesenchyme throughout stage 16 to stage 32, LH-2A transcripts are detectable in the distal limb bud and LH-2B transcripts are detectable in the anterior limb bud. Signals from the apical ectodermal ridge positively regulate LH-2A expression, since removal of the apical ectoderm results in the rapid reduction of LH-2A expression in the distal limb mesenchyme. Ectopic expression of the sonic hedgehog gene in the anterior margin of the limb bud results in the rapid reduction of LH-2B expression accompanying respecification of the positional value to the posterior phenotype. These results suggest that LH-2A and LH-2B play important roles in the determination and specification of the proximal-distal and anterior-posterior positional values, respectively (Nohno, 1997).

The homeobox genes Xlim-1 and goosecoid are coexpressed in the Spemann organizer and later in the prechordal plate that acts as head organizer. Since gsc is a possible target gene for Xlim-1, the regulation of gsc transcription by Xlim-1 and other regulatory genes expressed at gastrula stages was studied by using gsc-luciferase reporter constructs injected into animal explants. A 492-bp upstream region of the gsc promoter responds to Xlim-1/3m, an activated form of Xlim-1, and to a combination of wild-type Xlim-1 and Ldb1, a LIM domain binding protein, supporting the view that gsc is a direct target of Xlim-1. Footprint and electrophoretic mobility shift assays with GST-homeodomain fusion proteins and embryo extracts overexpressing FLAG-tagged full-length proteins show that the Xlim-1 homeodomain and the Xlim-1/Ldb1 complex recognize several TAATXY core elements in the 492-bp upstream region, where XY is TA, TG, CA, or GG. Some of these elements are also bound by the ventral factor PV.1, whereas a TAATCT element does not bind Xlim-1 or PV.1 but does bind the anterior factors Otx2 and Gsc. These proteins modulate the activity of the gsc reporter in animal caps: Otx2 activates the reporter synergistically with Xlim-1 plus Ldb1, whereas Gsc and PV.1 strongly repress reporter activity. Using animal cap assays, it has been shown that the endogenous gsc gene is synergistically activated by Xlim-1, Ldb1, and Otx2 and that the endogenous otx2 gene is activated by Xlim-1/3m, and this activation is suppressed by the posterior factor Xbra. Based on these data, a model is proposed for gene interactions in the specification of dorsoventral and anteroposterior differences in the mesoderm during gastrulation (Mochizuki, 2000).

Thus Gsc protein is capable of inhibiting the activity of its own promoter in assays using reporters activated by Xlim-1, Ldb1, and Otx2. Otx2 and Gsc belong to the same homeodomain group in that both have a lysine residue at position 50 of the homeodomain and share binding specificity for TAATCT and TAATCC. Since these two proteins recognize similar target sequences, there may be competition between Otx2 and Gsc for binding to the C site of distal element, with Otx2 having activating and Gsc inhibiting effects. Inhibition of the mouse and human gsc promoter by Gsc requires the proximal element, suggesting that Gsc inhibition, just like Xlim-1 activation, involves multiple sites in the complex gsc promoter. Repression of the gsc promoter by Gsc and PV.1 proteins is similarly effective under the experimental conditions employed, but the biological roles of the two proteins are different. In the case of Gsc autoinhibition, the rationale may be to provide a feedback loop to limit gsc expression. In contrast, PV.1, closely related to Xvent-1, is expressed ventrally as a consequence of BMP signaling in a region of the embryo where gsc is not expressed. It appears that PV.1 is a repressor protein whose function is to maintain the character of ventral mesoderm by inhibiting gsc expression in the non-organizer regions of the marginal zone. Similarly, Xbra may inhibit Gsc function in the notochord where gsc expression diminishes during gastrulation. The ability of Xbra to repress otx2 expression and of Gsc to repress Xbra expression may play a role in restricting gsc expression to the prechordal plate and Xbra expression to the notochord at mid- to late gastrulation. However, because Xbra is a transcriptional activator, it is assumed that otx2 repression is indirect (Mochizuki, 2000).

These transcription factor interactions have been incorporated into a model of dorsoventral and anteroposterior patterning in the gastrula embryo. In the prechordal plate, Xlim-1 and Ldb1, in addition to contributing to chordin induction, maintain the expression of otx2 and of gsc; the autoinhibitory action of the latter is counteracted by the activating function of Otx2, while Xbra expression is suppressed by Gsc. In the notochord, the high initial level of Xbra prevents otx2 gene activation by Xlim-1 plus Ldb1, and in the absence of Otx2, the gsc gene turns itself off by autorepression. Note that in the early gastrula, gsc is active in the entire organizer, but its expression fades in posterior axial mesoderm as gastrulation proceeds. In ventral mesoderm, the strong repression of gsc and otx2 by PV.1/Xvent-1 and Xbra maintains the ventral character of this tissue. Clearly, this scheme is incomplete in that additional factors are undoubtedly involved, yet it provides a cogent model for the interactions of the factors considered in this paper during axial patterning in the gastrula (Mochizuki, 2000).

A comparative analysis of LIM-homeodomain (LIM-hd) expression patterns in the developing stage 32 Xenopus brain is presented. x-Lhx2, x-Lhx7, and x-Lhx9 were isolated and their expression, together with that of x-Lhx1 and x-Lhx5, was analyzed in terms of prosomeric brain development and LIM-hd combinatorial code and compared with mouse expression data. The results show an almost complete conservation of expression patterns in the diencephalon. The Lhx1/5 and Lhx2/9 subgroups label the pretectum/ventral thalamus/zona limitans versus the dorsal thalamus, respectively, in alternating stripes of expression in both species. Conversely, strong divergences in expression patterns are observed between the telencephalon of the two species for Lhx1/5 and Lhx2/9. Lhx7 exhibits particularly conservative patterns and is proposed as a medial ganglionic eminence marker. The conservation of diencephalic segments is proposed to mirror the conservative nature of diencephalic structures across vertebrates. In contrast, the telencephalic divergences are proposed to reflect the emergence of significant novelty in the telencephalon (connectivity changes) at the anamniote/amniote transition. Moreover, the data allow the new delineation of pallial and subpallial domains in the developing frog telencephalon; these are compared with mouse subdivisions. In the pallium, the mouse combinatorial expression of LIM-hd is notably richer than in the frog, again possibly reflecting evolutionary changes in cortical connectivity (Bachy, 2001).

It is suggested that LIM-hd expression defines telencephalic subdivisions in developing Xenopus. Such subdivisions defined by gene expression become more and more precise in the developing telencephalon of mice or birds. In contrast, they are poorly known in frogs and fishes, probably because of the fact that their telencephalon is small and less differentiated. It is suggested that stage 32 Xenopus telencephalon is delimited by a line drawn from the optic stalk and running dorsally orthogonal to the brain axis. Inside the telencephalon, concurrent expression of LIM-hd factors and other genes such as x-Dll3 define pallial and subpallial compartments. Inside these, LIM-hd expression defines two pallial and three subpallial divisions (Bachy, 2001).

It is suggested that the x-Lhx7-expressing area corresponds to the medial ganglionic eminence. The frog pallidum is histologically poorly delineated. Only connectivity and immunohistochemical data suggest the existence of a pallidum in amphibians, but GABAergic neurons have never been found. The finding of an Lhx7-positive domain localized inside the distalless-positive subpallium is an additional excellent argument in favor of the existence of this structure. The mammalian mge also expresses Lhx6 and Lhx8 (two paralogs of Lhx7; Lhx8 probably has its origin in a rodent-specific duplication). It is not known whether the frog mge expresses any other x-Lhx7 paralog. However, it is noteworthy that the mammalian mge also expresses Lhx2 and therefore presents a richer LIM-hd code. Functionally, members of the Lhx6/7/8 group might be involved in the tangential migration of GABA interneurons from the mge to the striatum and cortex in rodents. The x-Lhx7 expression pattern might suggest that similar migrations occur in the amphibian telencephalon. In another respect, Lhx2 is strongly expressed in the proliferative zone of the rodent basal ganglia, which are hypoplasic in Lhx2^-/- mice. The absence of x-Lhx2 in the Xenopus cell-poor pallidum therefore would agree with a role for Lhx2 in cell proliferation control (Bachy, 2001).

Two other subdivisions, expressing x-Lhx2 and x-Lhx1/2/9, can be delineated from LIM-hd expression in Xenopus subpallium. Altogether, the three LIM-hd-deduced subpallial compartments might correspond to the three subdivisions proposed as the striatal, pallidal, and telencephalic stalk divisions of the basal forebrain. Among them, only the Lhx7-positive region can be attributed to the mge with some confidence. In the two other compartments the LIM-hd combinations are clearly different between Xenopus and mouse and might reflect the many differences in cell types and connectivity found in the basal ganglia of the two species. However, the possibility that cell migrations occur in frog telencephalon, as described in mouse, and could impair the interpretation of the results, cannot be excluded (Bachy, 2001).

The Xenopus LIM homeodomain (LIM-HD) protein, Xlim-1, is expressed in the Spemann organizer and cooperates with its positive regulator, Ldb1, to activate organizer gene expression. While this activation is presumably mediated through Xlim-1/Ldb1 tetramer formation, the mechanisms regulating proper Xlim-1/Ldb1 stoichiometry remain largely unknown. The Xenopus ortholog (XRnf12) of the RING finger protein Rnf12/RLIM has been isolated and its functional interactions with Xlim-1 and Ldb1 have been explored. Although XRnf12 functions as an E3 ubiquitin ligase for Ldb1 and causes proteasome-dependent degradation of Ldb1, co-expression of a high level of Xlim-1 suppresses Ldb1 degradation by XRnf12. This suppression requires both the LIM domains of Xlim-1 and the LIM interaction domain of Ldb1, suggesting that Ldb1, when bound to Xlim-1, escapes degradation by XRnf12. A high level of Ldb1 suppresses the organizer activity of Xlim-1/Ldb1, suggesting that excess Ldb1 molecules disturb Xlim-1/Ldb1 stoichiometry. Consistent with this, Ldb1 overexpression in the dorsal marginal zone suppresses expression of several organizer genes including postulated Xlim-1 targets, and importantly, this suppression is rescued by co-expression of XRnf12. These data suggest that XRnf12 confers proper Ldb1 protein levels and Xlim-1/Ldb1 stoichiometry for their functions in the organizer. Together with the similarity in the expression pattern of Ldb1 and XRnf12 throughout early embryogenesis, Rnf12/RLIM is proposed as a specific regulator of Ldb1 to ensure its proper interactions with LIM-HD proteins and possibly other Ldb1-interacting proteins in the organizer as well as in other tissues (Hiratani, 2003).

In the telencephalic pallium, four major subdivisions are found in birds and mammals: medial (hippocampus), dorsal (isocortex), lateral (olfactory cortex), and ventral (amygdala/claustrum) pallium. These pallial divisions can be deduced from LIM-hd expression in mice, by comparing mediolateral extent and laminar patterns of Lhx1/2/5/9 expression. In Xenopus, only two pallial subdivisions hae been found: one expressing only Lhx2, the other expressing Lhx2/5. Functionally, Lhx2 in mice regulates the formation of the cortical hem and Lhx5 controls neural patterning in the hippocampus, implying crucial roles for the LIM-hd family in patterning pallial subdivisions. The presence in the small Xenopus pallium of two LIM-hd-defined compartments suggests that only two distinct functional areas are found in anamniotes, when using these specific markers. An intermediate territory has been observed between Dlx- and Emx-positive domains in frog telencephalon, which is likely to correspond to the ventral pallium in amniotes. This intermediate territory does not correspond to one of the LIM-hd-defined compartments, because none of them is found in the Emx-negative, Dlx-negative portion of the frog telencephalon (Bachy, 2001).

In conclusion, richer LIM-hd combinatorial expression in the mammalian pallium could reflect an enrichment in cortical connectivity. This will have to be functionally tested by overexpression experiments in Xenopus, or, conversely, by analyzing in these terms the mouse lines in which LIM-hd genes have already been inactivated. Finally, the topological relationships of the deduced telencephalic subdivisions are organized along the anteroposterior axis of the brain, with the 'basal ganglia' complex in the anterior position (apparently ventral, because of the brain curvature) and the pallium in a more caudal position (apparently dorsal). This appears strongly similar to the situation shown for the chicken telencephalic fate map (Bachy, 2001).

To elucidate the molecular basis of organizer functions in Xenopus, target genes were sought of the LIM homeodomain protein Xlim-1, which is one of the organizer-specific transcriptional activators. An activated form of Xlim-1, Xlim-1/3m, initiates ectopic expression of the head-inducing organizer factor BMP antagonist gene cerberus in animal caps. Thus, the cerberus promoter was analyzed using reporter assays. Three consecutive TAAT motifs of the homeodomain-binding sites between positions -141 and -118, collectively designated the '3×TAAT element', are crucial for the response of the cerberus promoter to Xlim-1/3m, and for its activation in the dorsal region of the embryo. Because cooperative activation of the cerberus promoter by Xnr1 and Xwnt8 also requires the 3×TAAT element, focus was place on homeodomain transcriptional activators downstream from either Nodal or Wnt signaling. Wild-type Xlim-1 was found to synergistically activate the cerberus promoter with paired-like homeodomain transcription factor Mix.1 and paired-type homeodomain protein Siamois through the 3×TAAT element, and this synergy requires the LIM domains of Xlim-1. In contrast, Xotx2 acts synergistically with Mix.1, and Siamois through the TAATCT sequence at -95. Electrophoretic mobility shift assays reveal that Xlim-1, Siamois, and Mix.1 are likely to bind as a complex, in a LIM domain-dependent manner, to the region containing the 3×TAAT element. These data suggest that cerberus is a direct target for Xlim-1, Mix.1, Siamois, and Xotx2. Therefore, a model is proposed for the molecular link in the inductive sequence from the formation of the organizer to anterior neural induction (Yamamoto, 2003).

Table of contents

The Interactive Fly resides on the
Society for Developmental Biology's Web server.