Synergistic activation of transcription is mediated by the N-terminal domain of Drosophila fushi tarazu homeoprotein and can occur without DNA binding by the protein

Range of HOX/TALE superclass associations and protein domain requirements for HOXA13:MEIS interaction

AbdB-like HOX proteins form DNA-binding complexes with the TALE superclass proteins MEIS1A and MEIS1B, and trimeric complexes have been identified in nuclear extracts that include a second TALE protein, PBX. Thus, soluble DNA-independent protein-protein complexes exist in mammals. The extent of HOX/TALE superclass interactions, protein structural requirements, and sites of in vivo cooperative interaction have not been fully explored. We show that Hoxa13 and Hoxd13 expression does not overlap with that of Meis1-3 in the developing limb; however, coexpression occurs in the developing male and female reproductive tracts (FRTs). We demonstrate that both HOXA13 and HOXD13 associate with MEIS1B in mammalian and yeast cells, and that HOXA13 can interact with all MEIS proteins but not more diverged TALE superclass members. In addition, the C-terminal domains (CTDs) of MEIS1A (18 amino acids) and MEIS1B (93 amino acids) are necessary for HOXA13 interaction; for MEIS1B, this domain was also sufficient. We also show by yeast two-hybrid assay that MEIS proteins can interact with anterior HOX proteins, but for some, additional N-terminal MEIS sequences are required for interaction. Using deletion mutants of HOXA13 and HOXD13, we provide evidence for multiple HOX peptide domains interacting with MEIS proteins. These data suggest that HOX:MEIS interactions may extend to non-AbdB-like HOX proteins in solution and that differences may exist in the MEIS peptide domains utilized by different HOX groups. Finally, the capability of multiple HOX domains to interact with MEIS C-terminal sequences implies greater complexity of the HOX:MEIS protein-protein interactions and a larger role for variation of HOX amino-terminal sequences in specificity of function.

Candidate downstream regulated genes of HOX group 13 transcription factors with and without monomeric DNA binding capability

Hox genes encode transcription factors that regulate the morphogenesis of developing embryos. In mammals, knowledge of the genetic pathways, including the possible direct or indirect targets, regulated by HOX proteins is extremely limited. To identify the downstream genes regulated by posterior HOX proteins, we expressed HOXA13 in mouse embryonic fibroblasts lacking paralog group 13 expression using a bicistronic HOXA13/EGFP retroviral vector. Microarray analysis identified 68 genes with significant, reproducible RNA expression changes (50 activated; 18 repressed) in stable HOXA13-expressing cells. Genes with the GO annotation terms "extracellular matrix" and "basement membrane" were greatly overrepresented, and several were shown to be regulated by HOX proteins in other studies. Among the genes strongly activated by HOXA13 were Enpp2, a bifunctional enzyme known to modulate tumor and normal cell motility and which is expressed in precartilaginous condensations; Fhl1, a transcription factor implicated in muscle cell differentiation and development; and M32486, a putative integral membrane molecule expressed in the female reproductive tract. Expression differences in the HOXA13-expressing cells were confirmed for selected downstream genes using semi-quantitative RT-PCR, and in vivo coexpression with Hoxa13 in the limb interdigital mesenchyme was demonstrated for many. For two candidates, Igfbp4 and Fstl, interdigital limb bud expression was reduced in Hoxa13 mutants. To explore whether paralogous and nonparalogous HOX proteins could regulate the same genes, we created new HOX cell lines and examined the expression of selected genes identified by the HOXA13 screen. HOXD13 similarly activated/repressed 6 tested candidates, demonstrating that multiple downstream genetic pathways may be regulated by paralog HOX proteins. In contrast, HOXA9 was only able to repress expression of some gene targets. A HOXD13 mutant, HOXD13(IQN >)(AAA), incapable of monomeric DNA-binding, activated the expression of 5 HOXA13-upregulated genes; but was incapable of repressing the expression of Ngef and Casp8ap2. Our results suggest that HOX protein-protein interactions without direct HOX DNA-binding may play a larger role in HOX transcriptional regulation than generally assumed, and DNA-binding appears critical for repression.

The nuclear receptor Ftz-F1 and homeodomain protein Ftz interact through evolutionarily conserved protein domains

The Drosophila homeodomain protein Fushi Tarazu (Ftz) and its partner, the orphan receptor Ftz-F1, are members of two distinct families of DNA binding transcriptional regulators. Ftz and Ftz-F1 form a novel partnership in vivo as a Hox/orphan receptor heterodimer. Here we show that the murine Ftz-F1 ortholog SF-1 functionally substitutes for Ftz-F1 in vivo, rescuing the defects of ftz-f1 mutants. This finding identified evolutionarily conserved domains of Ftz-F1 as critical for activity of this receptor in vivo. These domains function, at least in part, by mediating direct protein interactions with Ftz. The Ftz-F1 DNA binding domain interacts strongly with Ftz and dramatically facilitates the binding of Ftz to target DNA. This interaction is augmented by a second interaction between the AF-2 domain of Ftz-F1 and the N-terminus of Ftz via an LRALL sequence in Ftz that is reminiscent of LXXLL motifs in nuclear receptor coactivators. We propose that Ftz-F1 serves as a cofactor for Ftz by facilitating the selection of target sites in the genome that contain Ftz/Ftz-F1 composite binding sites. Ftz, on the other hand, influences Ftz-F1 activity by interacting with its AF-2 domain in a manner that mimics a nuclear receptor coactivator.

The identification of Prx1 transcription regulatory domains provides a mechanism for unequal compensation by the Prx1 and Prx2 loci

Transcription regulatory domains of the Prx1a and Prx1b homeoproteins were analyzed in transient transfection assays using artificial promoters as well as an established downstream target promoter (tenascin-c). Activation and repression domains were detected in their common amino end. In the carboxyl end of Prx1a an activation domain and an inhibition/masking region (OAR domain) were detected. The Prx1b isoform, generated by alternative splicing, does not contain these carboxyl activation or inhibition domains. Instead, the data demonstrate that the carboxyl tail of Prx1b contains a potent repressor region. This difference in the carboxyl tail accounts for a 45-fold difference observed in transcription regulatory activity between Prx1a and Prx1b. The data also support the likelihood that this difference between Prx1a and Prx1b is higher in the presence of still undetermined cofactors. DNA binding affinities of Prx1a, Prx1b, and a series of truncation mutants were also examined. The carboxyl tail of Prx1a, which inhibited transcription activation in the transfection assays, also inhibited DNA binding. These differences in biochemical function between Prx1a and Prx1b, as well as the recently described activities of Prx2, provide a mechanism for the unequal compensation between the Prx1 and Prx2 loci.

Goosecoid suppresses cell growth and enhances neuronal differentiation of PC12 cells

In all vertebrate species, the homeobox gene goosecoid serves as a marker of the Spemann organizer tissue. One function of the organizer is the induction of neural tissue. To investigate the role of goosecoid in neuronal differentiation of mammalian cells, we have introduced goosecoid into PC12 cells. Expression of goosecoid resulted in reduced cell proliferation and enhanced neurite outgrowth in response to NGF. Expression of goosecoid led to a decrease in the percentage of S-phase cells and to upregulation of the expression of the neuron-specific markers MAP-1b and neurofilament-L. Analysis of goosecoid mutants revealed that these effects were independent of either DNA binding or homodimerization of Goosecoid. Coexpression of the N-terminal portion of the ets transcription factor PU.1, a protein that can bind to Goosecoid, repressed neurite outgrowth and rescued the proliferation of PC12 cultures. In contrast, expression of the bHLH transcription factor HES-1 repressed goosecoid-mediated neurite outgrowth without changing the proportion of S-phase cells. These results suggest that goosecoid is involved in neuronal differentiation in two ways, by slowing the cell cycle and stimulating neurite outgrowth, and that these two events are separately regulated.

Plasticity of Drosophila paired function

The Drosophila Paired (Prd) transcription factor has homeodomain (HD) and paired domain (PD) DNA-binding activities required for in vivo function. Correspondingly, Prd activation of late even-skipped (eve) expression occurs through a conserved target sequence (PTE) with HD and PD half sites, both of which are required for activation. To investigate the relationship between the HD and PD, and their roles in conferring specificity to Prd function, we tested altered versions of the Prd protein and of the PTE target site using in vivo assays in embryos. We found that function through PTE was constrained by the targeting specifications of both the HD and PD as well as the spatial relationship between these two domains. PTE function was also constrained by the spacing between the target half sites for the PD and HD, although surprisingly, late eve activation was retained when PTE was replaced by in vitro optimized binding sites for either the PD alone or for an HD dimer. In contrast to late eve regulation, other Prd targets tolerated more changes in the Prd protein, suggesting that their target sequences may be qualitatively different from PTE.

How does the fushi tarazu gene activate engrailed in the Drosophila embryo?

In the even-numbered parasegments of the Drosophila embryo, expression of the fushi tarazu (ftz) gene is necessary for transcription of engrailed (en). Yet those cells expressing ftz+ in a stripe, only the anteriormost come to express en. One explanation is that the level of ftz+ might be graded across the stripe and in order to express en, it would be sufficient for cells to exceed a threshold concentration of Ftz protein. We use photographs and microspectrophotometry to measure differences in Ftz antigen concentration; we do not find a gradient within the Ftz stripe. Rather, the stripe appears to contain cells with similar amounts of antigen plus a few weakly staining cells that are usually at the posterior edge. Further, varying the amount of Ftz protein has no effect on en expression. Finally, embryos lacking the even-skipped gene have normal levels of Ftz but do not express en. Our observations appear to rule out the threshold hypothesis.

Combinatorial activity of pair-rule proteins on the Drosophila gooseberry early enhancer

The early expression of the Drosophila segment polarity gene gooseberry (gsb) is under the control of the pair-rule genes. We have identified a 514-bp enhancer which reproduces the early gsb expression pattern in transgenic flies. The transcription factor Paired (Prd) is the main activator of this enhancer in all parasegments of the embryo. It binds to paired- and homeodomain-binding sites, which are segregated on the enhancer. Using site-directed mutagenesis, we have identified sites critical for Prd activity. Negative regulation of this enhancer is mediated by the Even-skipped protein (Eve) in the odd-numbered parasegments and by the combination of Fushi-tarazu (Ftz) and Odd-skipped proteins in the even-numbered parasegments. The organisation of the Prd-binding sites, as well as the necessity for intact DNA binding sites for both paired- and homeodomains, suggests a molecular model whereby the two DNA-binding domains of the Prd protein cooperate in transcriptional activation of gsb. This positive activity appears to be in competition with Eve and Ftz on Prd homeodomain-binding sites.

Divergent roles for NK-2 class homeobox genes in cardiogenesis in flies and mice

Recent evidence suggests that cardiogenesis in organisms as diverse as insects and vertebrates is controlled by an ancient and evolutionarily conserved transcriptional pathway. In Drosophila, the NK-2 class homeobox gene tinman (tin) is expressed in cardiac and visceral mesodermal progenitors and is essential for their specification. In vertebrates, the tin homologue Nkx2-5/Csx and related genes are expressed in early cardiac and visceral mesodermal progenitors. To test for an early cardiogenic function for Nkx2-5 and to examine whether cardiogenic mechanisms are conserved, we introduced the mouse Nkx2-5 gene and various mutant and chimeric derivatives into the Drosophila germline, and tested for their ability to rescue the tin mutant phenotype. While tin itself strongly rescued both heart and visceral mesoderm, Nkx2-5 rescued only visceral mesoderm. Other vertebrate ‘non-cardiac’ NK-2 genes rescued neither. We mapped the cardiogenic domain of tin to a unique region at its N terminus and, when transferred to Nkx2-5, this region conferred a strong ability to rescue heart. Thus, the cardiac and visceral mesodermal functions of NK-2 homeogenes are separable in the Drosophila assay. The results suggest that, while tin and Nkx2-5 show close functional kinship, their mode of deployment in cardiogenesis has diverged possibly because of differences in their interactions with accessory factors. The distinct cardiogenic programs in vertebrates and flies may be built upon a common and perhaps more ancient program for specification of visceral muscle.

Molecular cloning and characterization of human cardiac homeobox gene CSX1

Accumulating evidence has suggested that homeo-domain-containing proteins play critical roles in regulating the tissue-specific gene expression essential for tissue differentiation and in determining the temporal and spatial patterns of development. In order to elucidate the mechanisms of human heart development, we have isolated a human homologue of the murine cardiac homeobox gene Csx (also called Nkx-2.5) and denoted it as CSX1. The amino acid sequence of the CSX1 homeodomain is 100% and 67% identical to that of murine Csx/Nkx-2.5 and Drosophila tinman, respectively. CSX1 has at least three isoforms generated by an alternative splicing mechanism. One of these isoforms (CSX1a) encodes a protein of approximately 35 kD that possesses the homeodomain, whereas the other two (CSX1b and CSX1c) encode a truncated protein of approximately 12 kD that is identical to the CSX1a protein at the amino-terminal 112 amino acids but lacks the homeodomain. Northern blot analysis showed that CSX1 transcripts are abundantly expressed in both fetal and adult hearts, but no signal was detected in other human tissues examined. Amplification of each isoform by reverse transcriptase-polymerase chain reaction revealed that all of the three isoforms are expressed in fetal and adult hearts and that the homeobox-containing isoform CSX1a is most abundant. The homeodomain-containing protein encoded by CSX1a binds to Csx/Nkx-2.5 binding sequences and transactivates the sequence-containing luciferase reporter gene. Unexpectedly, the homeodomain-lacking protein encoded by CSX1b also transactivates the reporter gene, although CSX1b does not bind to the Csx/Nkx-2.5 binding sequences. The highly conserved homeodomain sequence in evolution and the restricted expression in the heart suggest that CSX1 plays an important role in the development and differentiation of the human heart and that there may be two different mechanisms in transcriptional regulation by the CSX1 protein, homeodomain-dependent and -independent mechanisms.

Both the paired domain and homeodomain are required for in vivo function of Drosophila Paired

Drosophila paired, a homolog of mammalian Pax-3, is key to the coordinated regulation of segment-polarity genes during embryogenesis. The paired gene and its homologs are unusual in encoding proteins with two DNA-binding domains, a paired domain and a homeodomain. We are using an in vivo assay to dissect the functions of the domains of this type of molecule. In particular, we are interested in determining whether one or both DNA-binding activities are required for individual in vivo functions of Paired. We constructed point mutants in each domain designed to disrupt DNA binding and tested the mutants with ectopic expression assays in Drosophila embryos. Mutations in either domain abolished the normal regulation of the target genes engrailed, hedgehog, gooseberry and even-skipped, suggesting that these in vivo functions of Paired require DNA binding through both domains rather than either domain alone. However, when the two mutant proteins were placed in the same embryo, Paired function was restored, indicating that the two DNA-binding activities need not be present in the same molecule. Quantitation of this effect shows that the paired domain mutant has a dominant-negative effect consistent with the observations that Paired protein can bind DNA as a dimer.

Drosophila Paired regulates late even-skipped expression through a composite binding site for the paired domain and the homeodomain

The even-skipped (eve) pair-rule gene plays a key role in the establishment of the anterior-posterior segmental pattern of the Drosophila embryo. The continuously changing pattern of eve expression can be resolved into two phases. Early expression consists of seven broad stripes in the blastoderm embryo, while late expression, which occurs after cellularization, consists of narrow stripes with sharp anterior borders that coincide with the odd-numbered parasegment boundaries. Previous studies have shown that these two phases are controlled by separate classes of cis elements in the eve promoter. Early stripes are expressed by multiple stripe-specific elements under the control of maternal-effect genes and gap genes, while late stripes are expressed by a single regulatory element, the ‘late element’, under the control of pair-rule genes including eve itself. We report here that paired (prd), a pair-rule gene which had been considered to be below eve in the regulatory hierarchy of pair-rule genes, in fact plays a critical role in the regulation of late eve expression. Transgenic analysis shows that this regulation is largely mediated by an evolutionarily conserved sequence within the late element termed PTE (Paired Target Element). In vitro analysis shows that the Prd protein binds strongly to this sequence. Interestingly, PTE contains juxtaposed binding sites for the two DNA-binding domains of the Prd protein, the paired domain and the homeodomain. Mutagenesis of either binding site leads to significant reduction in the activity of the late element, indicating that both DNA-binding domains in the Paired protein are required for regulation.

Patterning of the Drosophila embryo by a homeodomain-deleted Ftz polypeptide

Homeodomain proteins regulate diverse developmental processes in a wide range of organisms, yet bind in vitro to DNA sequences that are remarkably similar. This has raised the fundamental question of how target gene specificity is achieved in vivo. The Drosophila fushi tarazu protein (Ftz) contains a homeodomain and is required for the formation of alternate segments. We have shown previously that a homeodomain-deleted Ftz polypeptide (Ftz delta HD), incapable of binding DNA in vitro, could regulate endogenous ftz gene expression. Here we test Ftz delta HD activities in a ftz mutant background and find that, surprisingly, Ftz delta HD can directly regulate ftz-dependent segmentation, suggesting that it can control target gene expression through interactions with other proteins. A likely candidate is the pair-rule protein Paired (Prd). Ftz delta HD bound directly to Prd in vitro and required Prd to repress wingless in vivo. These results emphasize the pivotal importance of protein-protein interactions in homeodomain protein function.

Transcriptional repression by the human homeobox protein EVX1 in transfected mammalian cells

The human homeobox protein EVX1 (EVX1) is thought to play an important role during embryogenesis. In this study, the effect of EVX1 on gene transcription has been investigated in transfected mammalian cells. EVX1 expression represses transcription of a reporter gene directed by either cell-specific or viral promoter/enhancer sequences in a variety of mammalian cell lines and in a concentration-dependent manner. Transcriptional repression is independent of the presence of DNA-binding sites for EVX1 in all the promoters we tested. Furthermore, repression by EVX1 is evident also using a TATA-less minimal promoter in the reporter construct. A carboxyl-terminal proline/alanine-rich region of EVX1 seems to be responsible for the transcriptional repression activity, as suggested by transfection of EVX1 mutants. We speculate that the repressor function of EVX1 contributes to its proposed role in embryogenesis.

Dissection of the Drosophila paired protein: functional requirements for conserved motifs

The Drosophila paired gene encodes three conserved motifs: a homeodomain, paired domain and PRD (his/pro) repeat. To investigate the functional importance of the PRD repeat and paired domain, we tested deletion mutants using an ectopic expression assay in embryos. Our results suggest that the PRD repeat is not required for the in vivo regulation of the target genes, engrailed and gooseberry. However, the PRD repeat appears to be embedded within a proline-rich transcriptional activation domain required for the regulation of these genes. Our analysis of the paired domain indicated that its N-terminal half, which is required for DNA binding in vitro, is also required for in vivo function, whereas surprisingly, the C-terminal half is dispensable for the regulation of engrailed and gooseberry.

Abnormal embryonic cerebellar development and patterning of postnatal foliation in two mouse Engrailed-2 mutants

The cerebellum is an ideal system to study pattern formation in the central nervous system because of its simple cytoarchitecture and regular organization of folds and neural circuitry. Engrailed-2 (En-2) is expressed in a spatially restricted broad band around the mesencephalic-metencephalic junction, a region from which the cerebellum is derived. Mice homozygous for a targeted deletion of the En-2 homeobox, En-2hd, previously have been shown to have an altered adult cerebellar foliation pattern. To address whether the En-2hd allele was hypomorphic, we generated a putative null mutation that makes an N-terminal deletion (ntd). Mice homozygous for this new mutation, En-2ntd, display an identical cerebellar patterning defect, suggesting that both alleles represent null alleles. We also examined the developmental profile of En-2 homozygous mutant cerebellar foliation. This revealed a complex phenotype of general developmental delay and abnormal formation of specific fissures with the most severe morphological disruptions being limited to the posterior region of the cerebellum. The expression of two transgenes, which express lacZ in lobe-specific patterns in the cerebellum, also was found to be altered in En-2 homozygotes, suggesting possible lobe transformations. Finally, during embryogenesis there was a clear delay in fusion of the cerebellar rudiments at the midline by 15.5 d.p.c. This and the expression pattern of En-2 suggests that although cerebellar foliation is largely a postnatal process, the patterning of the cerebellum may begin during embryogenesis and that En-2 plays a critical role in this early process.