The homeobrain (hbn) gene is a new paired-like homeobox gene that is expressed in the embryonic brain and the ventral nerve cord. Expression of homeobrain initiates during the blastoderm stage in the anterior dorsal head primordia and the gene is persistently expressed in these cells that form parts of the brain during later embryonic stages. An additional weaker expression pattern is detected in cells of the ventral nerve cord from stage 11 on. The homeodomain in the Homeobrain protein is most similar to the Drosophila proteins Rx, Aristaless and Munster. In addition, the localized brain expression patterns of homeobrain and Rx resemble each other. Two other homeobox genes, orthopedia and Rx are clustered in the 57B region along with homeobrain. The current evidence indicates that homeobrain, Rx and orthopedia form a homeobox gene cluster in which all the members are expressed in specific embryonic brain subregions (Walldorf, 2000).
In contrast to paired-type homeodomains that have serine at position 50, the genes similar to homeobrain have paired-like homeodomains of the subclass that possess a glutamine (Rx, Al, Mu, Repo, Otp) or lysine (D-Gsc) at position 50. Another common motif that is encoded by Rx, homeobrain and the Gsc genes is the octapeptide/GEHdomain, a transcriptional repression domain. The OAR domain, an activation domain found only in some homeoproteins of the paired class is not present in the Homeobrain protein sequence. It is interesting that nearly all of the genes in this subclass (with the exception of repo) reside in two gene complexes at 21C (al, mu and Gsc) and 57B (otp, Rx and homeobrain) (Walldorf, 2000).
A phylogenetic tree indicates that Hbn is only slightly more closely related to Mu and Al proteins than to Rx, Gsc and Otp in the paired-like homeobox family. These data do not support the idea that there was one ancestral complex of brain/tail homeobox genes that then duplicated and diverged to form the two clusters of hbn related genes in the current Drosophila genome. A comparison of the chromosomal location of vertebrate orthologs might give hints concerning the evolution of the two gene clusters. Although Otp, Gsc and Rx have obvious vertebrate orthologs, there is not yet strong evidence that there are distinct orthologs for Hbn, Mu and Al proteins in vertebrate genomes (Walldorf, 2000).
Homeobrain transcript expression starts at the syncytial blastoderm stage (stage 4) as a horseshoe-like pattern in the dorsal head region. During the cellular blastoderm stage (stage 5) the expression domain retracts from the ventrolateral side. A second more laterally located expression domain becomes visible during early gastrulation. In addition, the horseshoe pattern resolves into two distinct domains, a dorsal domain showing weaker staining and dorsal±lateral domains with stronger staining. During embryonic stages 9 and 10, transcript expression is detected in three domains on either side of the dorsal posterior head and in a dorsal-anterior spot. When the germband is fully extended at stage 11, a few cells in each neuromere of the CNS express homeobrain transcripts and shortly afterwards each of the three dorsal±lateral expression domains in the head split, resulting in six dorsal±lateral patches of expression. Another expression domain in the posterior region of the embryo (hindgut/midgut boundary) appears during germband retraction. Discrete localized expression domains in the brain and (at lower levels) in the ventral nerve cord are detected until the end of embryogenesis, whereas the expression in the posterior (gut) region is absent at the terminal stages of embryonic development (Walldorf, 2000).
Planarians are the free-living members (order Tricladida) of the phylum Platyhelminthes. They are triploblastic, acoelomate, unsegmented and located at the base of the Lophotrochozoa clade. Besides their huge regenerative capacity, planarians have simple eyes, considered similar to the prototypic eye suggested by Charles Darwin in his book 'On the Origin of Species'. The conserved genetic network that determines the initial steps of eye development across metazoans supports a monophyletic origin of the various eye types present in the animal kingdom. This study summarizes the pattern of expression of certain genes involved in the eye network that have been isolated in planarians, such as Otx, Pax-6, Six, Rax and opsin. The effects of RNA interference-mediated loss of function on eye regeneration are described. Finally, the relevance of these findings for the evolution of the eye gene network is discussed (Salo, 2002).
The homeobox Ol-Rx3 gene is a medaka gene homologous to the mouse, Xenopus, zebrafish and Drosophila Rx genes. Ol-Rx3 starts to be expressed, at late gastrula stages, in the presumptive territories of the anterior brain. Subsequently, transcripts are localized in an antero-ventral region of the prosencephalon and in the primordia of the optic vesicles. During organogenesis, distribution of Ol-Rx3 transcripts are gradually restricted to the floor of the diencephalon, the prospective territory of the hypothalamus and the neurohypophysis. During late development and in adult, Ol-Rx3 expression is maintained in hypothalamic nuclei bordering the third ventricle. In the optic vesicles, Ol-Rx3 expression is temporarily switched off when the eye cup morphogenesis is complete, but it is turned on again in the inner nuclear layer of the retina. Thus, the early expression pattern of Ol-Rx3 is in agreement with a conserved role in the specification of the ventral forebrain and eye field. Putative functions linked to late expression domains are discussed in light of the different hypothesis concerning the involvement of vertebrate Rx genes in the maintenance of particular cell fate (Deschet, 1999).
The teleost Astyanax mexicanus exhibits eyed surface dwelling (surface fish) and blind cave dwelling (cavefish) forms. Despite the lack of functional eyes as adults, cavefish embryos form eye primordia, which later arrest in development, degenerate and sink into the orbit. The expression patterns of various eye regulatory genes during surfacefish and cavefish development have been compared to determine the cause of eye degeneration. Rx and Chx/Vsx family homeobox genes, which have a major role in cell proliferation in the vertebrate retina, have been examined in this study. A full-length RxcDNA clone (As-Rx1) and part of a Chx/Vsx(As-Vsx2) gene, which appear to be most closely related to the zebrafish Rx1 and Alx/Vsx2 genes respectively, were isolated and sequenced. In situ hybridization shows that these genes have similar but non-identical expression patterns during Astyanax eye development. Expression is first detected in the optic vesicle, then throughout the presumptive retina of the optic cup, and finally in the ciliary marginal zone (CMZ), the region of the growing retina where most new retinoblasts are formed. In addition, As-Rx1 is expressed in the outer nuclear layer (ONL) of the retina, which contains the photoreceptor cells, and As-Vsx2 is expressed in the inner nuclear layer, probably in the bipolar cells. With the exception of reduced As-Rx-1 expression in the ONL, the As-Rx1 and As-Vsx2 expression patterns were unchanged in the developing retina of two different cavefish populations, suggesting that cell proliferation is not inhibited. These results were confirmed by using PCNA and BrdU markers for retinal cell division. It is concluded that the CMZ is active in cell proliferation long after eye growth is diminished and is therefore not the major cause of eye degeneration (Strickler, 2000).
In early vertebrate eye development, the retinal anlage is specified in the anterior neuroectoderm. During neurulation, the optic vesicles evaginate from the lateral wall of the prosencephalon. The temperature-sensitive mutation eyeless is described in the Japanese medakafish. Marker gene analysis indicates that, whereas specification of two retinal primordia and proximodistal patterning takes place in the mutant embryo, optic vesicle evagination does not occur and subsequent differentiation of the retinal primordia is not observed. The mutation eyeless thus uncouples patterning and morphogenesis at early steps of retinal development. Temperature-shift experiments indicate a requirement for eyeless activity prior to optic vesicle evagination. Cell transplantation shows that eyeless acts cell autonomously (Winkler, 2000).
Eye development and brain structures of a mutant teleost fish were investigated. The el (eyeless) mutation in medaka (Oryzias atipes) is recessive and affects eye formation; in the most severe cases, it results in the absence of eyes. Developmental studies reveal that normal eyeballs are not formed in the el mutant embryos, but small optic cup-like structures differentiate in situ in the walls of the prosencephalon without evagination. The anophthalmic el homozygous fish hatch normally, although they do not respond behaviorally to visual stimuli. A small fraction of these fish grow to adulthood. In the adult anophthalmic el homozygous fish, the brain exhibits abnormalities in several subdivisions. A pair of small abnormal protrusions is observed on the surface of the ventral telencephalon and preoptic area. Immunocytochemistry using a rhodopsin monoclonal antibody shows that opsin-positive cells are present in the abnormal structures. Bodian staining showed that the optic nerves are present near the abnormal structures, although the number of optic nerve fibers is extremely small. The optic tectum is extremely small, and the thickness of the stratum opticum and stratum fibrosum et griseum superficiale is reduced. These behavioral and morphological observations suggest that the adult anophthalmic el homozygous fish are functionally blind, although small retina-like structures are partially differentiated and persist in the adult fish brain. Moreover, the adult anophthalmic el homozygous fish are infertile, and the sizes of the hypophysis and the hypothalamus are reduced. Thus, the el mutation affects not only the brain structures that are related to the visual system but also those related to the reproductive system (Ishikawa, 2001).
The complete absence of eyes in the medaka fish mutation eyeless is the result of defective optic vesicle evagination. The eyeless mutation is caused by an intronic insertion in the Rx3 homeobox gene resulting in a transcriptional repression of the locus that is rescued by injection of plasmid DNA containing the wild-type locus. Functional analysis reveals that Six3- and Pax6- dependent retina determination does not require Rx3. However, gain- and loss-of-function phenotypes show that Rx3 is indispensable to initiate optic vesicle evagination and to control vesicle proliferation, by that regulating organ size. Thus, Rx3 acts at a key position coupling the determination with subsequent morphogenesis and differentiation of the developing eye (Loosli, 2001).
The paired-class homeobox gene, Rx, is important in eye development. Expression patterns of three zebrafish Rx genes (Zrx1, 2, 3) has been studied in embryos and adults. All three genes show dynamic spatiotemporal patterns of expression. Zrx3 is expressed earliest, in the anteriormost region of the neural plate, in regions that give rise to ventral diencephalon and retinae. As development proceeds, Zrx3 expression is reduced in the lateral optic primordia, and is absent in the optic cup, but is retained at the ventral midline of the diencephalon, and is expressed in hypothalamus in the adult. As the neural retina begins to differentiate, Zrx3 is re-expressed in a subset of cells in the inner nuclear layer, presumably bipolar cells, and this expression is retained in the adult. In contrast, Zrx1/2 have a slightly later onset of expression, are initially coincident with Zrx3, but then become complementary, remaining on in the optic primordia but disappearing from the ventral midline of the diencephalon. Zrx1/2 are down-regulated as the retina differentiates, except in the outer nuclear layer where they continue to be expressed at high levels in cone, but not rod, photoreceptors. This is the first transcription factor described that distinguishes between cone and rod photoreceptors (Chuang, 1999).
Zebrafish retinal homeobox genes rx1 and rx2 are expressed exclusively in the optic primordia and then in cone photoreceptors of the differentiated neural retina. The rx expression domain is coextensive with the region identified as the retinal field in published fate maps of the neural plate in zebrafish embryos. Analysis of the spatiotemporal relationships between retinal and forebrain precursors suggests that lateral movement of retinal precursors is responsible for evagination of the optic primordia. Overexpression of either rx1 or rx2 results in the loss of forebrain tissue and the ectopic formation of retinal tissue. It was asked whether the deletion of forebrain and expansion of retinal tissue could be explained by the death of telencephalic precursors and enhanced proliferation of retinal precursors, and it was found that it could not. Instead, the data are consistent with a change in cell fate of forebrain precursors associated with reduced expression of telencephalic markers (emx1 and BF-1) and ectopic expression of retinal markers (rx1/2/3, pax6, six6, and vsx2) at the neural keel stage. The rx homeodomain alone is sufficient to induce ectopic retinal tissue, although weakly so, and this observation, together with results from deletion constructs, suggests that interactions with unidentified transcriptional regulators are important for rx1 and rx2 function during early eye development. It is concluded that regulated expression of zebrafish rx1 and rx2 helps to define the region of the forebrain fated to give rise to retinal tissue and may be involved in the cellular migrations that lead to splitting of the retinal field and formation of the optic primordia (Chuang, 2001).
Hedgehog (Hh) signaling is required for eye development in vertebrates; known roles in the zebrafish include regulation of eye morphogenesis and ganglion cell and photoreceptor differentiation. A temporally selective Hh signaling knockdown strategy was used (either antisense morpholino oligonucleotides or the teratogenic alkaloid cyclopamine) in order to dissect the separate roles of Hh signaling arising from specific sources. Also, the eye phenotype was examined of zebrafish slow muscle-omitted (smu) mutants, which lack a functional smoothened gene which encodes a component of the Hh signal transduction pathway. Hh signaling from extraretinal sources is found to be required for the initiation of retinal differentiation, but this involvement may be independent of the effects of Hh signaling on optic stalk development. Hh signals from ganglion cells participate in propagating expression of ath5. It is suggested that the effects of Hh signals from the retinal pigmented epithelium on photoreceptor differentiation may be mediated by the transcription factor rx1 (Stenkamp, 2003).
The results of this study suggest that the effects of Hh signaling on photoreceptor development may involve the transcription factor rx1, and further confirm that Hh signaling from the retinal pigment epithelium (RPE) is primarily implicated. Antisense injections delivered at 51 hpf generated some of the same retinal phenotypes as antisense injections delivered at earlier time points, indicating that interference with Hh signaling rather late in development is sufficient to interfere with photoreceptor differentiation. Knockdown of Hh signaling with antisense-MO consistently resulted in failed rx1 expression in the outer nuclear layer (ONL), while crx expression was unaffected, further supporting the hypothesis that Hh signaling may influence photoreceptor differentiation via the transcription factor rx1. The rx gene product has been shown to participate in regulating photoreceptor-specific gene expression in cell-free systems. The chicken homolog of zebrafish rx1/2, RaxL, is involved in the early stages of photoreceptor differentiation. To confirm the proposed interaction in zebrafish it will be important to demonstrate that rx1 expression regulates photoreceptor differentiation (opsin expression) in vivo. One alternative to this hypothesis is that effects of reduced Hh signaling on rx and opsin genes are related manifestations of a photoreceptor maturation defect (Stenkamp, 2003).
The smu-/- embryos similarly show reduced expression of photoreceptor markers, and lack of rx1 in the photoreceptor layer. Interestingly, many of the smu-/- embryos develop normally laminated retinas. It is suspected that these mutants are those that had sufficient maternal smoothened expression to initiate retinal retinal differentiation, but lacked functional (zygotic) Smoothened at the time of Hh signaling from the RPE. In these mutants, it would be predicted that the only notable retinal defects would be those related to Hh signaling from ganglion cells and RPE. A fraction of the mutants showed a small patch of rx1 and rod opsin expression in the ventronasal ONL, suggesting that this region of retina may have requirements for cell differentiation that are distinct from the rest of the retina. This is consistent with the proposal that the ventral retina of the embryonic zebrafish comprises a discrete domain, influenced primarily by signals originating outside the eye, while the differentiation of the remainder of the retina requires the propagation of additional Hh, and other signals, from within the eye (Stenkamp, 2003).
A novel Xenopus homeobox gene, Xrx1, belonging to the paired-like class of homeobox genes, has been isolated. Xrx1 is expressed in the anterior neural plate, and subsequently in the neural structures of the developing eye (neural retina and pigmented epithelium), and in other forebrain structures deriving from the anterior neural plate: in the pineal gland, throughout its development; in the diencephalon floor, and in the hypophysis. Xrx1's rostral limit of expression corresponds to the chiasmatic ridge, which some authors consider as the anteriormost limit of the neural tube: thus, Xrx1 may represent one of the most anteriorly expressed homeobox genes reported to date. Moreover, its expression in organs implicated in the establishment of circadian rhythms, may suggest for Xrx1 a role in the genetic control of this function. Finally, analysis of Xrx1 expression in embryos subjected to various treatments, or microinjected with different dorsalizing agents (noggin, Xwnt-8), suggests that vertical inductive signals leading to head morphogenesis are required to activate Xrx1 (Casarosa, 1997).
The anteriormost part of the neural plate is fated to give rise to the retina and anterior brain regions. In Xenopus, this territory is initially included within the expression domain of the bicoid-class homeobox gene Xotx2 but very soon, at the beginning of neurulation, it becomes devoid of Xotx2 transcripts in spatiotemporal concomitance with the transcriptional activation of the paired-like homeobox gene Xrx1. By use of gain- and loss-of-function approaches, the role played by Xrx1 in the anterior neural plate and its interactions with other anterior homeobox genes were studied. At early neurula stage Xrx1 is able to repress Xotx2 expression, thus first defining the retina-diencephalon territory in the anterior neural plate. Overexpression studies indicate that Xrx1 possesses a proliferative activity that is coupled with the specification of anterior fate. Expression of a Xrx1 dominant repressor construct (Xrx1-EnR) results in a severe impairment of eye and anterior brain development. Analysis of several brain markers in early Xrx1-EnR-injected embryos reveals that anterior deletions are preceded by a reduction of anterior gene expression domains in the neural plate. Accordingly, expression of anterior markers is abolished or decreased in animal caps coinjected with the neural inducer chordin and the Xrx1-EnR construct. The lack of expansion of mid-hindbrain markers, and the increase of apoptosis in the anterior neural plate after Xrx1-EnR injection, indicate that anterior deletions result from an early loss of anterior neural plate territories rather than posteriorization of the neuroectoderm. Altogether, these data suggest that Xrx1 plays a role in assigning anterior and proliferative properties to the rostralmost part of the neural plate, thus being required for eye and anterior brain development (Andreazzoli, 1999).
Chick rax/rx cDNAs, cRaxL (chick Rax/Rx-like) and cRax, (chick Rax) have been isolated and their expression patterns have been examined during early eye and brain development. The cRaxL cDNA encodes a 228 amino acid protein that is most closely related to the zebrafish Rx1 and Rx2. The cRax cDNA encodes a 317 amino acid protein, which shares higher homology with the Xenopus Rx. In addition to the homeodomain, the octapeptide and paired tail domains are conserved between the cRax and other vertebrate Rax/Rx, while cRaxL lacks the octapeptide containing N-terminal region, which is conserved among all other members of the rax/rx gene family identified so far. The chick rax/rx genes are expressed in overlapping domains in the anterior neural ectoderm, which corresponds to the forebrain and retina field, and later in the optic vesicle. cRax mRNA can be detected earlier than cRaxL prior to the formation of the notochord and its expression domain appears broader than that of cRaxL (Ohuchi, 1999).
A conserved vertebrate homeobox gene, Rx, is essential for normal eye development, and its misexpression has profound effects on eye morphology. Xenopus embryos injected with synthetic Rx RNA develop ectopic retinal tissue and display hyperproliferation in the neuroretina. Mouse embryos carrying a null allele of this gene do not form optic cups and so do not develop eyes. The Rx gene family plays an important role in the establishment and/or proliferation of retinal progenitor cells (Mathers, 1997).
Development of the vertebrate eye has been found to require the activity of several genes encoding homeodomain proteins. Some of these genes, or portions thereof, are highly conserved across phyla. A novel homeobox gene, rax (retina and anterior neural fold homeobox), has been identified; rax expression pattern suggests an important role in eye development. The predicted amino acid sequence of Rax comprises a protein with a paired-type homeobox, as well as the octapeptide that is found in many paired-type homeobox genes. In addition, in the C terminus of Rax, a 15-aa domain was found that has been termed the OAR domain. This domain is also found in several other homeobox genes. In the early mouse embryo, rax is expressed in the anterior neural fold, including areas that will give rise to the ventral forebrain and optic vesicles. Once the optic vesicles form, rax expression is restricted to the ventral diencephalon and the optic vesicles. At later stages, rax expression is found only in the developing retina. After birth, the expression of rax is restricted to the zone of proliferating cells within the retina, and expression gradually decreases as proliferation declines. These findings suggest that rax is one of the molecules that define the eye field during early development and that it has a role in the proliferation and/or differentiation of retinal cells (Furukawa, 1997).
RX, a homeodomain-containing protein essential for proper eye development, binds to the photoreceptor conserved element-1 (PCE-1/Ret 1) in the photoreceptor cell-specific arrestin promoter and stimulates gene expression. RX is found in many retinal cell types including photoreceptor cells. Another homeodomain-containing protein, CRX, which binds to the OTX element to stimulate promoter activity, is found exclusively in photoreceptor cells. Binding assay and cell culture studies indicate that both PCE-1 and OTX elements and at least two different regulatory factors RX and CRX are necessary for high level, photoreceptor cell-restricted gene expression. Thus, photoreceptor specificity can be achieved by multiple promoter elements interacting with a combination of both photoreceptor-specific regulatory factors and factors present in closely related cell lineages (Kimura, 2000).
The eyeless inbred mouse strain ZRDCT has long served as a spontaneous model for human anophthalmia and the evolutionary reduction of eyes that occurs in some naturally blind mammals. ZRDCT mice have orbits but lack eyes and optic tracts and have hypothalamic abnormalities. Segregation data suggest that a small number of interacting genes are responsible, including at least one major recessive locus, ey1. Although predicted since the 1940s, these loci were never identified. ey1 was mapped to chromosome 18 using an F2 genome scan and there a Met10-->Leu mutation was found in Rx/rax, a homeobox gene that is expressed in the anterior headfold, developing retina, pineal, and hypothalamus and is translated via a leaky scanning mechanism. The mutation affects a conserved AUG codon that functions as an alternative translation initiation site and consequently reduces the abundance of Rx protein. In contrast to a targeted Rx null allele, which causes anophthalmia, central nervous system defects, and neonatal death, the hypomorphic M10L allele is fully viable (Tucker, 2000).
Rx plays a critical role in eye formation. Targeted elimination of Rx results in embryos that do not develop eyes. In this study, the expression of Otx2, Six3, and Pax6 was examined in Rx deficient embryos. These genes show normal activation in the anterior neural plate in Rx-/- embryos, but they are not upregulated in the area of the neural plate that would form the primordium of the optic vesicle. In contrast, in homozygous Small eye embryos that lack Pax6 function, Rx shows normal activation in the anterior neural plate and normal upregulation in the optic vesicle/retinal progenitor cells. This suggests that neither Rx expression nor the formation of retinal progenitor cells is dependent on a functional copy of the Pax6 gene, but that Pax6 expression and the formation of the progenitor cells of the optic cup is dependent on a functional copy of the Rx gene (Zhang, 2000).
Despite the growing information concerning the developmental importance of the Prx2 protein, the structural determinants of Prx2 function are poorly understood. To gain insight into the transcription regulatory regions of the Prx2 protein, a series of truncation mutants were generated. Both the Prx2 response element (PRE) and a portion of the tenascin promoter, a downstream target of Prx2, were used as reporters in transient transfection assays. This analysis shows that a conserved domain (PRX), found in both Prx1 and Prx2, activates transcription in NIH 3T3 cells. This PRX domain, as well as other functional regions of Prx2, demonstrate both cell-specific and promoter-dependent transcriptional regulation. A second important region, the OAR (aristaless) domain, which is conserved among 35 Paired-type homeodomain proteins, is observed to inhibit transcription. Deletion of this element results in a 20-fold increase of transcription from the tenascin reporter in NIH 3T3 cells but not in C2C12 cells. The OAR domain does not function as a repressor in chimeric fusions with the Gal4 DNA binding domain in either cell type, characterizing it as an inhibitor instead of a repressor. These results give insight into the function of the Prx2 transcription factor while establishing the framework for comparison with the two isoforms of Prx1 (Norris, 2001).
Mechanisms of glial cell development in the vertebrate central nervous system have been examined. Genes have been identified that can direct the formation of glia in the retina. rax, a homeobox gene, Hes1, a basic helix-loop-helix gene, and notch1, a transmembrane receptor gene, are expressed in retinal progenitor cells, downregulated in differentiated neurons, and expressed in Muller glia. Retroviral transduction of any of these genes results in expression of glial markers. In contrast, misexpression of a dominant-negative Hes1 gene reduces the number of glia. Cotransfection of rax with reporter constructs containing the Hes1 or notch1 regulatory regions leads to the upregulation of reporter transcription. These data suggest a regulatory hierarchy that controls the formation of glia at the expense of neurons (Furukawa, 2000).
Homeobox genes are important regulators of cellular identity. Several homeobox genes are known to be specifically expressed in subsets of neurons in the forebrain, exclusively, or in distinct combinations. This study explores the expression of Homeobox genes in the forebrain of the adult rat, using a degenerate polymerase chain reaction cloning strategy. The expression of 12 homeobox genes was identified, several of which display a remarkable restricted expression pattern in the adult brain. The expression of goosecoid is demonstrated in a very small set of neurons in the hypothalamus. By using Otp as a marker, these goosecoid-positive cells were found to constitute a small area just beside the paraventricular nucleus. Furthermore, expression of Rx was found in the pineal gland, along with Alx4. Rx was additionally found in the posterior pituitary and in cells aligning the bottom of the third ventricle. These findings form a starting point to reveal functions of the described homeobox genes in the forebrain (Asbreuk, 2002).
Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.