even-skipped


TRANSCRIPTIONAL REGULATION (part 1/3)

Promoter Structure (part 1/2)

The eve promoter contains a TATA box (Frasch, 1987).

A 480 bp region of the eve promoter is both necessary and sufficient to direct a stripe of LacZ expression within the limits of the endogenous eve stripe 2. The maternal morphogen Bicoid and the gap proteins Hunchback, Krüppel and Giant all bind with high affinity to closely linked sites within this small promoter element. Activation appears to depend on cooperative interactions between BCD and HB proteins, since disrupting single binding sites causes catastrophic reductions in expression. GT, acting as a repressor, is directly involved in the formation of the anterior border, although additional repressors may also participate in this process. Forming the posterior border of the stripe involves a delicate balance between limiting amounts of the BCD activator and the KR repressor. Thus, clustering of activator and repressor binding sites in the stripe 2 element is required to bring these weakly interacting regulatory factors into close apposition so that they can function both cooperatively and synergistically to control transcription (Small, 1992).

Previous studies have provided a detailed model for the regulation of even-skipped stripe 2 expression in the Drosophila embryo. The strip 2 enhancer is inherently sensitized for repression by GT. Giant specifies the sharp anterior stripe border by blocking two tiers of transcriptional synergy, cooperative binding to DNA and cooperative contact of bound activators with the transcription complex. The synergistic activity of HB and BCD is promiscuous. For example, a maternally expressed Gal4-Sp1 fusion protein can functionally replace HB in the stripe 2 enhancer. This finding challenges previous proposals for dedicated HB and BCD interactions in the segmentation process (Arnosti, 1996).

An single enhancer sequence drives eve expression in stripes 3 and 7. The site consists of 500bp mapping 3.3kb upstream of the transciption start site. There are 5 Knirps binding sites in the 3 + 7 enhancer and 11 HB sites. HB and KNI act as repressors of stripe 3 expression, while the JAK kinase Hopscotch, acting through the Drosophila STAT protein Marelle, is involved in activation, with the KNI and HB sites closely linked to two STAT binding sites. In addition, Torso acts to restrict expression of stripe 3 in the anterior domain and stripe 7 in the posterior domain, while Bicoid restricts stripe 3 in the anterior domain. A model is presented in which the repressors provide short term quenching of widespread STAT activation (Small, 1996).

The stripe 3 enhancer is located 1.7 kb upstream of the stripe 2 enhancer. For proper stripe expression these enhancers must be physically separated by a minimum distance. For example, the levels of stripe 2 expression are augmented and there is a posterior expansion of the pattern when the stripe 3 enhancer is positioned immediately upstream of the stripe 2 enhancer. Despite this spacing requirement, the order of the enhancers within the eve promoter can be reversed without affecting the normal expression pattern. This suggests that spacing maintains the autonomous activities of the stripe enhancers and that interactions between enhancers can generate novel patterns of gene expression (Small, 1993).

Autoregulation of even-skipped is important for the specification of sharp stripes of gene expression at the onset of gastrulation. eve autoregulation is mediated, at least in part, by a 100-bp minimal autoregulatory sequence (MAS) located approximately 5 kb upstream from the eve transcription start site. Multimerization of a 200-bp DNA fragment that encompasses the MAS drives optimal autoregulatory activity, comparable to that obtained with the native distal enhancer element located between -5.9 and -5.2 kb. The MAS contains two EVE protein-binding sites (Jiang, 1991).

D. melanogaster and D. simulans stripe 2 enhancer regions have been compared. A relatively high level of variation in the stripe 2 enhancer region was found, including point substitutions and insertion/deletions in binding sites, and a comparable level of variation in the other noncoding regions. The patterns of variation fit a model of neutral molecular evolution. The multiplicity of binding sites in the enhancer apparently provides a redundancy in function that allows flexibility in the sequence requirements and structural design of the promoter (Ludwig, 1995).

Experimental investigations of eukaryotic enhancers suggest that multiple binding sites and trans-acting regulatory factors are often required for wild-type enhancer function. Genetic analysis of the stripe 2 enhancer of even-skipped (eve) provides support for this view. Given the importance of even-skipped expression in early Drosophila development, it is reasonable to predict that many structural features of the stripe 2 enhancer will be evolutionarily conserved, including the DNA sequences of protein binding sites and the spacing between them. To test this hypothesis, sequences of the stripe 2 enhancer were compared among four species of Drosophila: D. melanogaster, D. yakuba, D. erecta and D. pseudoobscura. This analysis reveals a surprising result that deviates from the expected results. There are a large number of nucleotide substitutions in regulatory protein binding sites for bicoid, hunchback, Kruppel and giant, as well as a systematic change in the size of the enhancer. Some of the binding sites in D. melanogaster are either absent or modified in other species. One functionally important Bicoid-binding site in D. melanogaster appears to be recently evolved. Possible functional consequences of sequence differences among these stripe 2 enhancers were investigated by P-element-mediated transformation. This analysis reveals that the eve stripe 2 enhancer from each of the four species drives reporter gene expression at the identical time and location in D. melanogaster embryos. Double staining of native Eve protein and transgene mRNA in early embryos shows that the reporter gene mimics native eve expression and, in every case, produces sharply defined stripes at the blastoderm stage that are coincident with Eve stripe 2 protein. It is argued that stripe 2 eve expression in Drosophila evolution can be viewed as being under constant stabilizing selection with respect to the location of the anterior and posterior borders of the stripe. It is further hypothesized that the stripe 2 enhancer is functionally robust, so that its evolution may be governed by the fixation of both slightly deleterious and adaptive mutations in regulatory protein binding sites as well as in the spacing between binding sites. This view allows for a slow but continual turnover of functionally important changes in the stripe 2 enhancer (Ludwig, 1998).

Genetic experiments and a targeted misexpression approach have been combined to examine the role of the gap gene giant (gt) in patterning anterior regions of the Drosophila embryo. The results suggest that gt functions in the repression of three target genes, the gap genes Kruppel (Kr) and hunchback (hb), and the pair-rule gene even-skipped (eve). The anterior border of Kr, which lies 4-5 nucleus diameters posterior to nuclei that express GT mRNA, is set by a threshold repression mechanism involving very low levels of Gt protein. The gap gene Kr is activated in a broad central region of precellular embryos. Midway through cleavage cycle 14, this domain extends from 41-59% egg length. The initial positioning of the anterior border of this domain is thought to be controlled by repression involving a combination of maternal and zygotic hunchback transcripts. To test whether gt is also involved in setting or maintaining this border, the Kr expression pattern was analyzed in embryos containing the st2-gt transgene, a modified version of the 480 bp eve stripe 2 enhancer. These embryos show no changes in the initial positioning of the Kr expression domain early in cleavage cycle 14, but slightly later there is a dramatic retraction of the anterior Kr border. The delay in the observed repressive effect on the Kr anterior border is probably due to the fact that the Kr domain is expressed earlier than the st2-gt transgene. Higher levels of ectopic gt result in a more severe retraction, suggesting that Kr transcription is very sensitive to repression by gt. To test whether gt affects Kr expression during normal development, Kr expression was examined in embryos that carry a strong hypomorphic gt allele. The initial Kr expression pattern was correctly established in these gt hypomorphic embryos. However, slightly later, a significant anterior expansion (from 59% to 65% egg length) is observed, suggesting that gt-mediated repression is essential for maintaining the position of the anterior border of the Kr domain (Wu, 1998).

giant activity is required, but not sufficient, for the formation of the anterior border of eve stripe 2, which lies adjacent to nuclei that express GT mRNA. It is proposed that gt's role in forming this border is to potentiate repressive interaction(s) mediated by other factor(s) that are also localized to anterior regions of the early embryo. It is not clear whether gt is sufficient for repression of the in vivo eve stripe 2 response. To test this, eve expression was examined in embryos containing the st2-gt transgene, which extends the gt domain so that it overlaps the position of eve stripe 2. Surprisingly, the ectopic gt causes only a weak transient reduction of the stripe early in cycle 14. Later the stripe recovers to full strength, but expands toward the posterior by about two nucleus diameters. Double in situ hybridization experiments show that the timing and the extent of the expansion correlates well with the retraction of the Kr domain, suggesting that the expansion of eve stripe 2 is indirectly caused by relief from Kr repression. Doubling the ectopic gt expression levels still does not cause a significantly stronger repression, suggesting that eve stripe 2 is quite insensitive to gt repression. To test whether the effects of ectopic gt on eve stripe 2 are controlled by the early or late regulatory elements, the expression of lacZ reporter genes was examined in embryos containing st2-gt transgenes. It is likely that the posterior expansion of endogenous eve stripe 2 caused by the st2-gt transgene is mediated through the early acting enhancer (Wu, 1998).

The recalcitrance of the eve stripe 2 response to ectopic gt expression led to a reexamination of the eve expression pattern in gt mutant embryos. Early in cycle 14, these mutants show a derepression in the interstripe region between stripes 1 and 2. However, later in cycle 14, gt mutants show a dramatic reduction in stripe 2 expression levels, suggesting a role for gt in maintaining the stripe. Since Kr has been previously implicated as the repressor that forms the stripe 2 posterior border, it is possible that the stripe 2 reduction in gt mutants is indirectly caused by Kr, which expands anteriorly to completely overlap the diminishing stripe. The repression of eve stripe 2 observed in gt mutants can be relieved by reducing Kr levels. These results suggest that a major function of the anterior gt domain is to prevent Kr from expanding anteriorly, thus permitting the expression of eve stripe 2. Furthermore, since gt repression maintains the position of the anterior Kr border in wild-type embryos, it indirectly defines the position of the posterior border of eve stripe 2 (Wu, 1998).

In principle, the preceding experiments support the hypothesis that gt acts as a concentration-dependent repressor to set the anterior borders of the Kr and eve stripe 2 expression domains in different positions. Ectopic gt is an effective repressor of Kr, but has little effect on the activation of eve stripe 2. In situ hybridization experiments indicate that endogenous gt levels are significantly higher than the ectopic gt driven by even the strongest st2-gt transgenic lines. Perhaps these higher endogenous levels are required for effectively setting the anterior border of eve stripe 2. If this is the case, the early expansion of eve stripe 2 toward stripe 1 detected in gt mutants should not be affected in embryos in which the endogenous gt gene is replaced by the st2-gt misexpression domain. To test this, eve expression was examined in gt mutants that contained the st2-gt5 transgene. Surprisingly, a sharp anterior eve stripe 2 border is formed in these embryos, with a clear interstripe between eve stripes 1 and 2. Furthermore, the st2-gt domain rescues eve stripe 2 to full strength, with a posterior expansion that is probably due to repression of the anterior Kr border. The relatively low levels of ectopic gt driven by the st2-gt construct overlap the endogenous gt domain and extend 4-5 nucleus diameters posteriorly. The fact that a sharp anterior eve stripe 2 border is formed in embryos containing only this domain argues against a simple concentration-dependent mechanism for setting this border. Rather, it is proposed that other factor(s) are involved along with gt in defining the anterior border of eve stripe 2 in vivo. Thus, gt may act as a potentiator of repression mediated by these localized factors. Since gt encodes a putative leucine zipper (b-ZIP) protein, one possibility is that this activity is also a b-ZIP protein that can heterodimerize with gt as part of an effective repressor complex. Repressive function in the absence of gt would be provided by a homodimer of this protein. Alternatively, since the gt site deletions tested in previous experiments removed relatively long sequences (14-43 bp), it is possible that these deletions may have removed or interrupted binding sites for other protein(s) (Wu, 1998).

The characterization of a 16-kb region sufficient for all known aspects of even-skipped expression and the rescue of an eve null mutation are described. 45 kb surrounding the eve coding sequence were examined for DNaseI hypersensitive sites and other transcription units. The previously identified eve regulatory elements, those for early stripes 2, 3, and 7 and the late element (all of which are included within the 16-kb region), do not generate prominent hypersensitive sites. However, strong, constitutive DNaseI hypersensitive sites flank a 16-kb region, within which one developmentally regulated site is found at the eve promoter region. P-element transformation of this 16-kb domain into eve mutants rescues them to adult viability. This 16-kb domain contains regulatory elements for all known features of eve expression: the seven major blastoderm stripes (stripes 2, 3 and 7 are controlled by 5' sequences; stripes 1, 4, 5 and 6 are controlled by 3' sequences); minor stripe expression during germ band extension (controlled by 5' sequences); later expression in the lateral mesodermal muscle precursor cells (controlled by 5' sequences); expression in the central nervous system (controlled by 3' sequences); expression adjacent to the invaginating proctodeum and in a ring around the anal pad (controlled by 3' and 5' sequences). A preliminary dissection of the 16-kb domain into its constituent regulatory elements is reported. Other major findings include the following: (1) there is a second element for late stripe expression adjacent to the traditional late element (present in 5' sequences); (2) a stripe element 3' of the gene interacts with the late element to give rise to the minor stripes seen in the even-numbered parasegments; (3) expression in the proctodeum and anal pad is driven by sequences both 5' and 3' of the gene and (4) expression in different sites in the central nervous system is driven by separable elements widely dispersed throughout 8 kb 3' of the gene (Sackerson, 1999).

Evidence for stabilizing selection in a eukaryotic enhancer element

Eukaryotic gene expression is mediated by compact cis-regulatory modules, or enhancers, which are bound by specific sets of transcription factors. The combinatorial interaction of these bound transcription factors determines time- and tissue-specific gene activation or repression. The even-skipped stripe 2 element controls the expression of the second transverse stripe of Even-skipped messenger RNA in Drosophila melanogaster embryos, and is one of the best characterized eukaryotic enhancers. Although even-skipped stripe 2 expression is strongly conserved in Drosophila, the stripe 2 element itself has undergone considerable evolutionary change in its binding-site sequences and the spacing between them. This apparent contradiction has been investigated, and two chimaeric enhancers, constructed by swapping the 5' and 3' halves of the native stripe 2 elements of two species, have been shown to no longer be able to drive expression of a reporter gene in the wild-type pattern. Sequence differences between species have functional consequences, therefore, but they are masked by other co-evolved differences. On the basis of these results, a model for the evolution of eukaryotic regulatory sequences is presented (Ludwig, 2000).

Multiple binding sites for each of four transcription factors, the activators Bicoid and Hunchback, and the repressors Kruppel and Giant, have been physically localized to the stripe 2 element (S2E). Genetic and experimental evidence indicates that concentration gradients of Bicoid and Hunchback can activate eve in a broad domain, whereas the more localized expression of the S2E is determined by the repressors Giant anteriorly and Kruppel posteriorly. Experimental elimination, addition or augmentation of both repressor and activation sites produces predictable changes in reporter-gene expression, and provides evidence for the mechanisms of enhancer function. This model emphasizes the functional importance of binding-site sequences, as well as the number and spatial configuration of binding sites within an element (Ludwig, 2000 and references therein).

In a comparison of 13 species, none of the 16 surveyed D. melanogaster binding sites is completely conserved. Most experimentally verified binding sites have accumulated point substitutions, and three are recognizable in only a subset of taxa. Each S2E also differs in the spacing between binding sites. Despite these differences, experiments with reporter constructs of native S2Es of four species have shown that each enhancer drives reporter-gene expression at the identical time and location in early D. melanogaster blastoderm embryos (Ludwig, 2000).

Does this mean that the substitutional differences between species are functionally inconsequential? To answer this question, investigation focused on comparison of native and chimaeric stripe 2 elements of D. melanogaster and D. pseudoobscura, whose most recent common ancestor occurred 40-60 million years ago. These species were chosen for this analysis because a comparative binding-site prediction method indicated potentially important differences in the S2Es, including the absence of the bcd-3 site, the presence of a new Kr site and reductions in likelihood potentials for bcd-4, bcd-2, Kr-4 and hb-3 sites in D. pseudoobscura relative to D. melanogaster (Ludwig, 2000).

The two native S2E sequences S2E(m) and S2E(p), where m and p refer to D. melanogaster and D. pseudoobscura respectively, were placed in a reporter-gene construct that also included an internal control for position effect. These two native constructs expressed in transformed embryos at a time point and location along the A-P axis that is indistinguishable from native eve expression. This result indicates either that the mutational differences between these S2Es are functionally unimportant (to the level of resolution allowed by the experimental procedure), or that the changes, if functional, balance one another in such a way as to produce no net functional change in expression (Ludwig, 2000).

It was predicted that any chimaeric enhancer that contains the conserved sequences of the S2E would have wild-type function if these sequences only are important for eve stripe 2 expression. However, if substitutional differences between lineages are functional then it should be possible to create chimaeric sequences that disrupt wild-type function. To test these alternatives, two complementary chimaeric enhancers were created in which the distal or the proximal parts of D. melanogaster S2E were substituted with corresponding parts of the D. pseudoobscura enhancer. The break lies within a conservative block of seven bases midway through the S2E, which allowed the creation of a 'seamless' connection between homologous parts in the chimaeric constructs. The chimaeric enhancers are designated S2E(m1-p2) and S2E(p1-m2), where (x-y) corresponds to species x distal (5') segment connected to species y proximal (3') segment (Ludwig, 2000).

In separate experiments involving multiple independent transformants of each chimaeric construct, consistent evidence has been found for a posterior shift of about the width of two cells in reporter-gene expression in the S2E(m1-p2) construct. In some embryos, and in some locations along the circumference of the stripe, this posterior expression appears to involve an expansion of the stripe rather than a simple shift. These two phenotypic defects, posterior shifting and expansion, suggest both a change in the sensitivity to the bcd/hb activation gradients, and a reduction in the overall effectiveness of Kr repression of the posterior stripe margin. The absence of a putative D. pseudoobscura-specific Kr site in the S2E(m1-p2) chimaera, on the basis of binding-site prediction, provides a testable hypothesis to explain the posterior shift in reporter-gene expression. The complementary construct, S2E(p1-m2), does not show the same defect. Instead, it appears to be defective in that it leads to subtle expansion of both the anterior and posterior borders of the stripe, when compared with native eve expression (Ludwig, 2000).

To explain these results, it is proposed that stabilizing selection has maintained phenotypic constancy for eve expression but has allowed mutational turnover of functionally important sites. Stripe 2 expression can be viewed as a quantitative character in which stabilizing selection has preserved eve expression to a specific band of cells and a particular time in embryogenesis. Mutational changes in the element, including some base substitutions within binding sites and short insertions or deletions in the spacer regions between binding sites, are proposed to have only weak functional (and therefore selective) effects. Theoretical models of phenotypic traits under stabilizing selection show that nearly neutral variation (that is, slightly deleterious and advantageous mutations) reaches fixation at an appreciable rate by the process of genetic drift. The model implies that each species lineage will differ by many functionally compensatory mutations. It is predicted that such a pattern of substitution will be a common theme in cis-regulatory evolution (Ludwig, 2000).

These results have implications for understanding changes in gene expression and morphological evolution. If this model of enhancer evolution is correct, then weakly selected mutations in regulatory elements will be present in natural populations and will be available for directional change by natural selection. Consistent with this prediction, two traits in Drosophila that differ between species but that are likely to be under stabilizing selection within a species -- abdominal bristle number and wing morphology -- are reported to have significant associations between quantitative trait variation and nucleotide polymorphism in noncoding regions of candidate loci. But as the results reported here indicate, selection can maintain functional conservation of gene expression for long periods of evolutionary time despite binding site turnover, which may make it difficult to identify homologous elements in different species groups by sequence comparison alone (Ludwig, 2000).

Role of CtBP in transcriptional repression by the Drosophila Giant protein

The Giant protein is a short-range transcriptional repressor that refines the expression pattern of gap and pair-rule genes in the Drosophila blastoderm embryo. Short-range repressors including Knirps, Krüppel, and Snail utilize the CtBP cofactor for repression, but it is not known whether a functional interaction with CtBP is a general property of all short-range repressors. Giant repression activity was studied in a CtBP mutant and it has been found that this cofactor is required for Giant repression of some, but not all, genes. While targets of Giant such as the even-skipped stripe 2 enhancer and a synthetic lacZ reporter show clear derepression in the CtBP mutant, another Giant target, the hunchback gene, is expressed normally. A more complex situation is seen with regulation of the Krüppel gene, in which one enhancer is repressed by Giant in a CtBP-dependent manner, while another is repressed in a CtBP-independent manner. These results demonstrate that Giant can repress both via CtBP-dependent and CtBP-independent pathways, and that promoter context is critical for determining giant-CtBP functional interaction. To initiate mechanistic studies of the Giant repression activity, a minimal repression domain within Giant has been identified that encompasses residues 89-205, including an evolutionarily conserved region bearing a putative CtBP binding motif (Stunk, 2001).

What characteristics of a regulatory region dictate CtBP-dependent or CtBP-independent repression? In considering which features of a gene determine CtBP-dependence or -independence, the structure of the basal promoter cannot be the deciding factor, for the same Kr promoter is regulated by distinct elements, some that exhibit CtBP-dependence and some that show CtBP-independence. Similarly, the eve gene is repressed by Knirps via CtBP-dependent and CtBP-independent regulatory elements. While the eve enhancers in question are kilobases apart, the Kr regulatory elements driving anterior and central domain (CD) expression are closely intertwined, and appear to share at least some of the same activator binding sites, suggesting that subtle differences in enhancer architecture or differences in levels of regulatory proteins interacting with those elements may dictate CtBP dependence. The Giant binding site in the Kr CD2 enhancer site was shown to be of higher affinity than the gt1 site in the eve stripe 2 enhancer. Thus, there may be a correlation between Giant binding site affinity and the requirement for CtBP, with elements containing Giant sites of lower affinity showing CtBP-dependence. A consensus has been derived for the Giant protein by aligning binding sites for Giant from eve, Kr, and the recently identified abdA iab-2 enhancer site. The consensus features an extended half-site inverted repeat TNTTAC, consistent with the dimeric nature of basic zipper proteins, and a central ACGT core common to recognition motifs for many basic zipper proteins. The higher affinity sequences from the CtBP-independent Kr CD element are closer to the consensus than those of the CtBP-dependent eve stripe 2 enhancer. Weaker sites may only be partially occupied, resulting in an overall lower level of Giant mediated repression. A loss of CtBP might further depress repression activity below a critical threshold, leading to the derepression observed. Repression of the lacZ reporter containing the giant CD1 site from Kr is CtBP-dependent, a result that contrasts with the CtBP independence of the CD itself, but this particular site may not be optimal, since it contains two mismatches. Full Giant activity may also be mediated on the native CD element through the additional high-affinity CD2 site (Stunk, 2001).

Anterior repression of a Drosophila stripe enhancer requires three position-specific mechanisms

The striped expression pattern of the pair-rule gene even skipped (eve) is established by five stripe-specific enhancers, each of which responds in a unique way to gradients of positional information in the early Drosophila embryo. The enhancer for eve stripe 2 (eve 2) is directly activated by the morphogens Bicoid (Bcd) and Hunchback (Hb). Since these proteins are distributed throughout the anterior half of the embryo, formation of a single stripe requires that enhancer activation is prevented in all nuclei anterior to the stripe 2 position. The gap gene giant (gt) is involved in a repression mechanism that sets the anterior stripe border, but genetic removal of gt (or deletion of Gt-binding sites) causes stripe expansion only in the anterior subregion that lies adjacent to the stripe border. A well-conserved sequence repeat, (GTTT)4 has been identified that is required for repression in a more anterior subregion. This site is bound specifically by Sloppy-paired 1 (Slp1), which is expressed in a gap gene-like anterior domain. Ectopic Slp1 activity is sufficient for repression of stripe 2 of the endogenous eve gene, but is not required, suggesting that it is redundant with other anterior factors. Further genetic analysis suggests that the (GTTT)4-mediated mechanism is independent of the Gt-mediated mechanism that sets the anterior stripe border, and suggests that a third mechanism, downregulation of Bcd activity by Torso, prevents activation near the anterior tip. Thus, three distinct mechanisms are required for anterior repression of a single eve enhancer, each in a specific position. Ectopic Slp1 also represses eve stripes 1 and 3 to varying degrees, and the eve 1 and eve 3+7 enhancers each contain GTTT repeats similar to the site in the eve 2 enhancer. These results suggest a common mechanism for preventing anterior activation of three different eve enhancers (Andrioli, 2002).

Previous DNA-binding experiments identified twelve sites for the genetically defined regulators within the eve 2 minimal stripe element (MSE). To identify other sequences important for eve 2 regulation, a series of mutant enhancers were constructed that contain deletions (D1-D5) of the regions between the known binding sites. Each deletion was tested separately in the context of an eve 2 lacZ fusion gene that also contains the eve 3+7 MSE as an internal control for levels of expression. Several independent lines for each construct were obtained by P-element mediated transformation. Embryos were collected from these lines, and examined by in situ hybridization for expression of lacZ mRNA (Andrioli, 2002).

All five deletions disrupt the normal function of the eve 2 enhancer. Four (D1, D3, D4, and D5) cause a significant reduction in the level of stripe activation. It is not clear whether these reductions are caused by removing discrete activator sites or by changing the spacing between the known sites. By contrast, the D2 deletion, which removes a 62 bp sequence between the Bcd 2 and the Kr 4 sites, causes an apparent strengthening of stripe 2 activation, and ectopic expression in more anterior regions. The derepression is quite broad early in nuclear cycle 14, but refines later to an ectopic anterior stripe. These results suggest that the D2 region contains sequences required for repression in the region of the ectopic stripe (Andrioli, 2002).

Because important functional binding sites are likely to be evolutionarily conserved, the eve 2 sequence from D. melanogaster was compared with published sequences from four other Drosophila species. The Kr4 and Bcd2 sites are well-conserved among all five species, and thus represent excellent anchors for the careful analysis of the intervening region. The best-conserved sequence block in this region is a 16 bp sequence that consists of four repeats of the sequence GTTT. Although this type of repeat is unusual for a functional binding site, it was tested by deletion or mutagenesis in the context of an eve 2-lacZ reporter gene. Both disruptions cause severe anterior derepressions. By contrast, a deletion that removes the rest of the D2 sequence (46 bp), but leaves the (GTTT)4 sequence intact, does not cause any detectable change in enhancer activity. Thus, the (GTTT)4 sequence is the major binding site for a repressive activity that prevents expression of the eve 2 enhancer in a specific anterior region (Andrioli, 2002).

Previous experiments suggested that the gap gene gt and the Gt-binding sites are required for the correct positioning of the anterior eve 2 border. To test the relationship between the (GTTT)4-binding activity and Gt-mediated repression, the eve2Delta(GTTT)4-lacZ construct was crossed into a gt mutant background. If the two repression mechanisms are independent, an additive effect would be expected from combining these two perturbations. If, however, Gt-mediated repression is partially redundant with the (GTTT)4-binding activity, removing both might cause a more severe derepression. In this cross there is an anterior shift and slight expansion of stripe 2 that is similar to the effects on the wild-type eve 2 transgene in gt mutants. No new effect is detected on the band of derepression created by deleting the (GTTT)4 site, and a small repressed area is still maintained between the two parts of the pattern. This result is consistent with an additive effect, and suggests that the (GTTT)4-binding activity functions independently of Gt-mediated repression. The failure to derepress in the region between the two parts of the pattern probably reflects the activity of the unknown protein X, which normally participates with Gt in repression (Andrioli, 2002).

The eve2Delta(GTTT)4-lacZ transgene is also repressed at the anterior tip, even in gt mutants, suggesting that yet another mechanism prevents activation in this region. This mechanism could work through another localized repressor activity, or by modifying Bcd, the major activator of eve 2. Consistent with the latter possibility, it has been previously shown that Bcd-dependent activation of hb and orthodenticle (otd) is downregulated by the Tor phosphorylation cascade at the anterior tip. To test whether tor controls the ability of Bcd to activate eve 2, the eve2Delta(GTTT)4-lacZ transgene was crossed into embryos lacking tor activity. This causes a significant derepression at the anterior tip, suggesting that tor-mediated modification of bcd activity is important for preventing activation in this region. A similar derepression is not detected with the wild-type eve2-lacZ transgene in tor mutants, suggesting that Tor-mediated repression is dependent on the (GTTT)4-binding activity. In summary, these results suggest that multiple activities are required for anterior repression of eve 2, and that three different mechanisms prevent activation in different anterior regions (Andrioli, 2002).

Because of the peculiar sequence of the (GTTT)4 site, gel shift experiments were performed using nuclear extracts from 0- to 12-hour-old wild-type embryos. These experiments showed the formation of several specific protein-DNA complexes. To identify specific proteins that bind the (GTTT)4 sequence, a yeast one-hybrid assay was constructed with constructs containing four tandem copies of the intact site. From an initial screen of ~500,000 clones, 66 true positives were obtained. These clones were then transformed into a yeast strain containing identical reporters except for base pair substitutions in the (GTTT)4 sequence. Forty-nine clones also activated one of the mutant constructs, leaving only 16 that activated the (GTTT)4 constructs, but not the negative controls. Among these 16 were two clones that encode histone H1 and a single clone that encodes the forkhead domain (FD) protein Slp1. slp was originally classified as a pair-rule mutation, but the slp locus contains two tightly linked genes, slp1 and slp2. These genes are related in their primary structure, and their expression patterns overlap significantly. However, slp1 is expressed much earlier in an anterior 'gap gene-like' domain, which first appears as an anterior cap, and then evolves into a broad stripe at approximately 80% egg length. Double staining experiments with gt show that both genes are expressed at the same time, and that slp1 expression overlaps the anterior part of the gt expression domain. Thus, the temporal and spatial expression patterns of slp1 are consistent with a role in anterior repression of eve 2 (Andrioli, 2002).

The (GTTT)4 sequence bears little resemblance to the only previously defined Slp1-binding site (TCTTCGATGTCAACACACC). Thus, tests were performed to see whether bacterially expressed Slp1 can bind directly to the (GTTT)4 sequence in vitro. These experiments show that Slp1 binds specifically to this sequence, suggesting that it may directly interact with this sequence in vivo. Similar results were obtained using a fragment of the Slp1 protein that contains only the forkhead domain (Andrioli, 2002).

If Slp1 acts as an anterior repressor of eve 2, genetically removing it might cause an anterior derepression of the eve2-lacZ expression pattern. To test this, the reporter was crosssed into a slp deletion mutant that completely removes the slp1-coding region and disrupts slp2. No anterior expansion was detected in this experiment. Endogenous eve expression was also analyzed in this mutant background, and slight anterior shifts of stripes 1 and 2 were detected, but no significant derepression in anterior regions. To test whether Slp1-mediated repression requires gt, eve and the eve2-lacZ reporter gene were both examined in gt; slp double mutant embryos. The double mutant shows no increase in the anterior derepression over that caused by removal of gt alone. These results argue against a role for Slp1 in anterior repression of eve, but do not rule out the possibility that Slp1 is one of several redundant proteins that repress through the (GTTT)4 site. Two other FD proteins, Fkh and Crocodile (Croc), are expressed in anterior regions of the embryo. However, the expression domains of both proteins are located very near the anterior pole, making it unlikely that either gene is involved in this repression mechanism. To make sure, eve and eve 2-lacZ expression was examined in each mutant; neither shows an anterior derepression. Thus these two genes are unlikely to play important roles in this repression mechanism (Andrioli, 2002). To further test the roles of Slp1 and Gt in eve patterning, a fragment of the snail (sna) promoter and the yeast FLP-FRT system were used to drive ectopic domains of each gene along the ventral surface of the embryo. This method is an efficient way to test whether any gene is sufficient for repression of individual stripes because the ventral expression domain intersects all seven eve stripes. In this assay, Slp1 expression alone distorts the expression of eve 1 in ventral regions by shifting it posteriorly, and causes a strong repression of eve 2 and a weaker repression of eve 3. By contrast, there is no detectable effect on the posterior stripes. Thus, Slp1 activity is sufficient for repression of specific anterior stripes including eve 2. By contrast, ventrally misexpressed Gt causes only a weak repression of eve 1 and 2, but strongly affects eve 5, a repression target of the posterior gt expression domain. The minor effect of Gt on eve 2 is transient, and the stripe recovers and expands posteriorly later in cycle 14. This expansion is probably caused by repression of Kr, which forms the posterior border of eve 2. These results confirm that Gt is not sufficient for effective repression of eve 2, and that its effect is much weaker than Slp1-mediated repression. Embryos that contain ventral expression domains of both Slp1 and Gt were constructed. While effects of both genes are detected within the same embryos, there is no evidence of synergistic repression activity in these experiments. This is consistent with the demonstration that the (GTTT)4 site is independent of Gt-mediated repression (Andrioli, 2002).

The repressive effects of ectopic Slp1 on the three anterior eve stripes suggest a common mechanism for repression in anterior regions of the embryo. To test this, the eve locus was examined for the presence of binding sites similar to the (GTTT)4 site in the eve 2 enhancer. Interestingly, there are only two other such sites in the eve locus; these are located within the boundaries of the stripe 1 and stripe 3+7 enhancers. Whether repression by ectopic Slp1 is mediated through these enhancers and the eve 2 MSE was examined. In these experiments, ectopic Slp1 causes a ventral repression of stripe 1 in the context of an eve1+5-lacZ transgene. A similar repression of stripe 3 was observed in the context of an eve3+7-lacZ transgene. These results are consistent with the idea of a common mechanism. By contrast, no ventral repression of the eve2-lacZ transgene was detected, which is surprising in light of the fact that eve 2 is the most strongly affected stripe in the context of the endogenous gene (Andrioli, 2002).

The reason for the discrepancy between the reporter and the endogenous gene is not clear. It is possible that Slp1-mediated repression of eve 2 requires eve sequences outside the minimal enhancer. For example, the late element (LE) mediates the refinement of all seven eve stripes after they are initially positioned. Perhaps interactions between the LE and/or other cis sequences are required for effective repression by Slp1. To test this, a larger reporter gene (-7.8 eve-lacZ) was used that contains all native sequences from the 5' border of the locus to the transcription start site. This transgene contains the LE, the eve 2 and eve 3+7 enhancers, and all native sequences that lie between these elements, and drives expression of stripes 2, 3 and 7, and a single line of nuclei located within the normal position of stripe 1. Expression from this reporter is effectively repressed at the position of stripes 1 and 3 in embryos containing ventrally expressed Slp1, but there is still no effect on the stripe 2 response. Thus, the addition of these extra sequences does not restore the sensitivity of eve 2 to Slp1-mediated repression. This suggests that undefined properties of the endogenous eve locus are required for Slp1-mediated repression of eve 2 (Andrioli, 2002).

The results presented here indicate that three distinct mechanisms are required for anterior repression of eve 2, with each activity functioning within a specific subregion (Andrioli, 2002).

In subregion III, the Gt-binding sites are crucial for repression -- deletion of these sites leads to an anterior expansion of the stripe. However, it is clear that Gt does not act alone, and that at least one other factor (X) must be involved in repressing through these sites. The identity of X is not clear, but genetic studies have localized a Gt-like patterning activity to the left arm of chromosome II. Segmental aneuploids that remove this arm show an expansion similar to that seen in gt mutants (Andrioli, 2002).

In subregion II, repression of eve 2 is mediated by the (GTTT)4 site described in this paper. A candidate protein, Slp1, has been identified that is expressed at the right time and place for the repression activity and binds specifically to this site in the yeast 1-hybrid experiment and in vitro. The (GTTT)4 site shows little similarity to the other known Slp1-binding site, but is quite similar to sites bound by other members of the FD protein family. For example, a 115 amino acid FD fragment of Fkh binds specifically to the site CTTTGTAAA, which bears some resemblance to the (GTTT)4 site. Also, the hepatocyte FD protein HNF-3 binds to a site (TGTTTGTTTTAGTT) that contains two perfect GTTT repeats. Ventrally expressed Slp1 specifically represses eve 2, strongly supporting a role in regulation of the endogenous eve gene. However, there is no effect on eve 2 in slp mutants, suggesting that Slp1 is redundant with at least one other protein (Y), which also mediates repression through the (GTTT)4 site. The existence of multiple complexes in gel shifts with embryo extracts is consistent with this, but the identity of Y is still unknown (Andrioli, 2002).

In subregion I, eve 2 repression is controlled by Tor, which may act by downregulating Bcd-dependent activation. This is consistent with the previous demonstration that Tor interferes with Bcd-dependent activation of hb, otd and slp1 (Andrioli, 2002).

In summary, at least five different protein activities are involved in three distinct mechanisms that repress eve 2 in anterior regions. Interestingly, it seems that all aspects of eve 2 regulation are controlled, directly or indirectly, by the Bcd morphogen gradient. The eve 2 enhancer is directly activated by Bcd, but activation is prevented near the anterior pole by Tor. The anterior expression patterns of the defined repressors of eve 2 (Slp1 and Gt) are also activated by Bcd. It is proposed that the relative positions of these domains, and the ultimate position of eve 2, are controlled by differential sensitivity to the Bcd concentration gradient. Future experiments on the cis-elements that regulate slp1 and gt transcription will be required to test this (Andrioli, 2002).

The study of eve regulation is an excellent paradigm for how complex promoters integrate the activities of multiple enhancers. Previous studies suggest that all five enhancers function independently in the segmented part of the embryo. The autonomy of each enhancer depends on short-range repression mechanisms and sufficient linear spacing between enhancers along the DNA sequence. These factors are crucial for creating the pattern of seven eve stripes because they permit different enhancers to be in different transcriptional states within the same nuclei. For example, in nuclei at the position of stripe 1, the eve 1 enhancer will activate transcription even though the eve 2 enhancer is in a repressed state (Andrioli, 2002).

A different scenario exists in regions anterior to the striped pattern, where none of the eve enhancers are activated. The genetic removal of various repression activities suggests that at least two eve enhancers (eve 2 and eve 3+7) can be activated in this region. The mechanism of eve 1 activation is still unknown, but it is reasonable to suggest that its activators are also distributed in this region. Thus, mechanisms must be in place to prevent activation by each enhancer. Alternatively, repression could occur by an anterior repression activity that directly contacts the basal transcription complex. The (GTTT)4 site in the eve 2 enhancer mediates anterior repression, and there are similar binding sites in the eve 1 and eve 3+7 enhancers. Furthermore, ectopic Slp1 expression represses all three stripes. These results argue against the mechanism of direct contact with the basal machinery, and suggest that these three enhancers share a common mechanism for repression in a specific anterior region of the embryo. Genetic experiments also show that the Tor phosphorylation cascade participates in polar repression of the eve 2 and eve 3+7 enhancers, which suggests another common repression mechanism that is shared by at least two of these enhancers (Andrioli, 2002).

By contrast, the eve 5 and eve 4+6 enhancers do not appear to contain sites similar to the (GTTT)4 sites, and they are immune to repression by ectopic Slp1. These enhancers are expressed in posterior regions of the embryo, and thus may be activated by factors localized there, making anterior repression unnecessary for their function (Andrioli, 2002).

slp was originally classified as a pair-rule gene based on its cuticular phenotype, and has been shown to function at both the level of the pair-rule genes and the segment polarity genes. However, specific patterning functions for the early anterior Slp1 expression domain have remained unclear, although strong alleles of slp1 exhibit severe defects in the mandibular lobe. The results presented in this paper suggest that Slp1 acts at the level of the gap genes by repressing enhancer elements that control the initial eve stripes. Thus, Slp1 function is required at three different levels of the segmentation hierarchy (Andrioli, 2002).

The mechanism involved in Slp1-mediated repression of eve is unknown, but may involve an interaction with the co-repressor Groucho (Gro). The Slp1 protein sequence contains a motif (FSIDAIL), which is very similar to the EH1 Gro-binding consensus (FSIDNIL) and Slp1 has been shown to bind Gro in vitro (Andrioli, 2002).

It has been proposed that Gro mediates repression by creating a direct physical link between DNA-bound proteins and components of the basal transcription machinery. As such, repressors that act through Gro can function over very long distances, and have been classified as long-range repressors. The ability of Slp1 to interact with Gro suggests such a long-range mechanism, but several considerations are not consistent with this model. For example, ventral expression of a long-range repressor that 'locks' the basal transcription machinery should repress all seven eve stripes, not just the anterior three. Also, the three (GTTT)4 sites described in this study are all located within minimal enhancer elements that control specific stripes. In a long-range mechanism, these sites could be located anywhere in the promoter, and need not be associated with specific enhancers (Andrioli, 2002).

One of the most intriguing findings of this study is that ectopic Slp1 represses eve 2 in the endogenous gene, but not in the context of several lacZ reporter genes. Despite much effort, this discrepancy has not been resolved, but it is informative to compare the structural differences between the endogenous gene and the tested transgenes. One obvious difference between the lacZ reporter genes tested in these experiments and the endogenous eve gene is copy number. Perhaps Slp1-mediated repression requires two copies of the enhancer in a homozygous situation. Consistent with this hypothesis, a pairing-sensitive element (PSE) that reduces marker gene expression in homozygotes has been identified in the far 3' region of the eve gene. Two experiments argue against this hypothesis: (1) ectopic Slp1 still represses endogenous eve in Df (eve)/+ heterozygotes and (2) Slp1 fails to repress eve 2-containing transgenes when they are homozygosed (Andrioli, 2002).

Another difference between the endogenous gene and the reporters is that the endogenous gene contains genomic regions outside those tested in the reporter genes, and is located in a different genomic position. Perhaps control sequences in the 3' region of the gene, or further 5' are required for this repression mechanism. As mentioned above, the eve 1 enhancer, which is located in the 3' region, contains a (GTTT)4-binding site. Perhaps effective repression of eve 2 requires all three sites contained in the three different enhancers. This will be tested in future experiments (Andrioli, 2002).

Finally, it is possible that the native eve locus is organized in a specific chromatin conformation that permits repression by Slp1, and this configuration is not maintained when eve 2 transgenes are inserted into ectopic genomic locations. The fact that Slp1 protein contains an FD DNA-binding domain is interesting in this regard. Structural studies suggest that FD domains form a 'winged-helix' are very similar to the globular DNA-binding domain of the linker histone H1. Furthermore, it has been shown that the mammalian FD protein hepatic nuclear factor 3 (HNF3) competes with H1 for binding to specific sites, and that this competition is critical for the in vivo regulation of the albumin liver-specific enhancer. Such a mechanism may be involved in Slp1-mediated repression of eve 2. Consistent with this, two clones have been isolated that encode histone H1 in the one-hybrid experiment with the (GTTT)4 site. This suggests that both proteins can bind to this site, and supports the idea that regulation of chromatin structure may be an important part of Slp1-mediated repression of eve 2. More experiments will be required to test this hypothesis (Andrioli, 2002).

The eve 2 enhancer is one of the best-characterized patterning elements in Drosophila development. Proteins involved in activation and repression have been identified, and a simple model has emerged that explains the basic activity of the enhancer. Anterior repression of this element requires at least three position-specific mechanisms -- this fact significantly extends understanding of this aspect of enhancer function. These results also suggest that the current model for activation of this enhancer is also incomplete. The deletion analysis identified four regions that are required for efficient activation of the enhancer. The effects of these deletions may be caused by changing the spacing between known activator and/or repressor sites within the enhancer. However, it possible that these regions contain specific binding sites required for activation. Consistent with this, there are several well-conserved sequence blocks that might represent specific sites required for activation. Base-pair substitutions that disrupt the conserved sequences without changing site spacing will be used to initially test this (Andrioli, 2002).

Specification of motoneuron fate in Drosophila: Integration of positive and negative transcription factor inputs by a minimal eve enhancer

The mechanisms that generate neuronal diversity within the Drosophila central nervous system (CNS), and in particular in the development of a single identified motoneuron called RP2, are of great interest. Expression of the homeodomain transcription factor Even-skipped (Eve) is required for RP2 to establish proper connectivity with its muscle target. The mechanisms by which eve is specifically expressed within the RP2 motoneuron lineage have been examined. Within the NB4-2 lineage, expression of eve first occurs in the precursor of RP2, called GMC4-2a. A small 500 base pair eve enhancer has been identified that mediates eve expression in GMC4-2a. Four different transcription factors (Prospero, Huckebein, Fushi tarazu, and Pdm1) are all expressed in GMC4-2a, and are required to activate eve via this minimal enhancer; one transcription factor (Klumpfuss) represses eve expression via this element. All four positively acting transcription factors act independently, regulating eve but not each other. Thus, the eve enhancer integrates multiple positive and negative transcription factor inputs to restrict eve expression to a single precursor cell (GMC4- 2a) and its RP2 motoneuron progeny (McDonald, 2003).

GMC4-2a forms at stage 9, becomes Eve+ at stage 11, and generates the Eve+ RP2/sib neurons at late stage 11. The second-born Eve-negative GMC4-2b forms at stage 10, and generates an unknown pair of neurons. The first transcription factors detected in GMC4-2a are Pros and Hkb, due to inheritance of the proteins from the neuroblast. The next transcription factors detected in GMC4-2a are Ftz and Pdm1. Ftz is first detected at stage 10, and Pdm1 is first detected at stage 11. The de novo expression of Pdm1 is distinct from its inheritance in GMCs produced by Pdm+ neuroblasts during the assignment of temporal identity. The last protein to be detected is Eve, which appears only at late stage 11. Pros, Hkb, Ftz, and Pdm1 are each expressed transiently in the RP2/sib neurons at stage 12, but by stage 16 none of these proteins is detectable in the mature RP2 neuron. It is concluded that there is a temporal sequence of transcription factor expression in GMC4-2a: first Pros and Hkb, then Ftz, then Pdm1, and that Eve is detected only after all of these proteins are present (McDonald, 2003).

GMC4-2b forms at late stage 10, never expresses Eve, and generates two unknown Eve-negative neurons. Three transcription factors that positively regulate eve expression are detected in GMC4-2b: Pros, Ftz, and Hkb. The pattern of Pdm1 expression is too complex to score at the time GMC4-2b is born. The negative regulator Klu is detected in GMC4-2b but not GMC4-2a. It is concluded that GMC4-2b expresses at least three of the four positively acting transcription factors that are required to activate eve (Pros, Ftz, Hkb), and at least one negative regulator of eve expression (Klu). The absence of eve expression is likely due to the presence of Klu, rather than the absence of a positive regulator, because klu mutants can activate eve transcription in GMC4-2b (McDonald, 2003).

The sequential expression of Pros, Hkb, Ftz, Pdm1, and Eve in GMC4-2a raises the possibility that these four transcription factors act in a linear pathway to regulate eve expression. If so, then a mutant in an early-acting gene should lead to loss of expression of all later-acting genes in the pathway. Alternatively, the four transcription factors could all act directly to activate eve transcription, with expression of eve occurring only after all transcription factors are present. In this case, mutants in one gene should have no effect on any other gene except eve. To distinguish between these two models, pros, hkb, ftz, and pdm1 mutants were examined for expression of all four transcription factors and eve. Pdm1 is detected in GMC4-2a in all mutant genotypes: Ftz is detected in GMC4-2a in all mutant genotypes: pros, hkb, and pdm1, and Hkb is detected in GMC4-2a in all mutant genotypes. Finally, Pros is observed in GMC4-2a in all mutant genotypes, as expected because Pros is transcribed and translated in neuroblasts and is asymmetrically partitioned into each GMC. Taken together, these data support the model that all four transcription factors act directly to activate eve transcription, with expression of eve occurring only after all transcription factors are present (McDonald, 2003).

To test the model that Pros, Hkb, Ftz, and Pdm1 transcription factors directly regulate eve expression, the eve cis-regulatory DNA that confers regulated expression in the NB4-2 lineage was identified. Eve is expressed in a subset of neurons in the embryonic CNS, including the aCC/pCC neurons derived from NB1-1, the U1-5 neurons derived from NB7-1, the EL neurons derived from NB 3-3, and the RP2/sib neurons derived from NB4-2. An eve cis-regulatory element [R79R92; from ~7.9 and ~9.2 kilobase pair (kb) on the eve genomic map] has been defined that accurately directs lacZ expression to the Eve+ cells within two NB lineages: GMC4-2a and its RP2 progeny and GMC1-1a and its aCC/pCC progeny. The properties of this element are examined in this study in detail. When the R79R92 eve element was truncated to ~7.9 to ~8.6 kb (R79N86), lacZ expression in RP2 and aCC was normal, whereas expression in the pCC neuron was reduced. Truncation of the eve element to ~7.9 to ~8.4 kb (R79S84) almost completely abolished expression of lacZ in pCC, although occasionally expression in pCC was observed at low levels, whereas expression in RP2 and aCC remained high. Further truncation of the left end point to ~8.0 kb (S80S84) resulted in a reduction of expression in both aCC and RP2. Addition of the region ~8.4 to ~8.6 kb to this fragment (S80N86) increased the level of expression. However, because the region ~8.4 to ~9.2 kb (S84R92) did not show any ability to activate lacZ, the region ~8.4 to ~8.6 kb is apparently insufficient on its own to direct expression, and thus serves an auxiliary function. The removal of ~8.2 to ~8.4 kb from P80N86 abolished expression (SNdeltaSC). Together with the fact that each of the fragments ~7.9 to ~8.2 kb (S79C82) and ~8.2 to ~9.2 kb (C82R92) failed to activate lacZ, this indicates that both of the regions ~7.9 to ~8.2 kb and ~8.2 to ~8.4 kb are necessary to direct expression, and that neither alone is sufficient. Consistent with this, two tandem copies of ~8.2 to ~8.4 kb failed to activate lacZ (C82S84x2), suggesting that the two regions may provide qualitatively different activities. In summary, the critical eve cis-regulatory element for the GMC4-2a and RP2 lies in a 0.5 kb fragment of genomic DNA between ~7.9 and ~8.4 kb (McDonald, 2003).

Do the genes that activate or repress eve expression in the NB4-2 lineage work through the minimal 500 bp RP2/aCC eve enhancer? Expression of R79S84-lacZ was assayed in pros, ftz, hkb, pdm1, and klu mutant embryos, and whether it was regulated identically to the endogenous eve gene was tested. ftz, pdm1, and hkb mutant embryos show loss of R79S84-lacZ in the RP2 neuron but not the aCC neuron, identical to the pattern of endogenous eve expression in these mutants. pros mutants show loss of eve-lacZ in both RP2 and aCC, identical to the pattern of endogenous eve expression in pros mutants. In embryos lacking klu, R79S84-lacZ is expressed in two cells at the RP2 position, whereas expression in aCC is normal; this matches the pattern of endogenous eve expression in klu mutant embryos. It is concluded that the R79S84 minimal eve cis-regulatory element precisely reproduces the pattern of endogenous eve expression within the NB4-2 lineage, and that transcription factors regulating eve in GMC4-2a can act through this enhancer to activate or repress eve expression (McDonald, 2003).

Expression of eve is not detected in GMC4-2b in wild-type embryos, but mutations in the klu gene result in ectopic expression of eve in GMC4-2b. Klu contains four predicted zinc fingers, one of which is highly homologous to the WT1 zinc finger domain. The consensus binding site for the WT1 zinc finger transcription factor is a ten nucleotide sequence, 5'-(C/G/T)CGTGGG( A/T)(G/T)(T/G)-3', with variable nucleotides shown in parentheses. It was reasoned that if Klu directly binds to the eve enhancer to repress expression in GMC4-2b, one or more WT1 consensus binding sites should be found in the minimal eve enhancer R79S84. Three conserved putative Klu-binding sites were found in the R79S84 sequence: site 1, GGGTGGGGAG at nucleotides ~8066 to ~8075; site 2, GCGTGGGTGA at nucleotides ~8090 to ~8099; and site 3, TCGCCCACCA at ~8262 to ~8271. Based on the fact that altering the C2, G3, G5, G6, and G7 to T or T4 to A in the WT1-consensus binding site abolished WT1 binding, nucleotide substitutions were made in the three putative Klu-binding sites. In sites 1 and 2, As were substituted for T4, G6, and G7. In site 3, which is a reversed binding site, Ts were substituted for C4, C6, and A7. These substitutions were made at all three sites; transgenic lines were constructed expressing the mutant enhancer driving lacZ (eveK123-lacZ), and the pattern of lacZ expression was examined in the CNS of wild-type embryos and embryos misexpressing Klu protein in the NB4-2 lineage (McDonald, 2003).

In wild-type embryos, the eveK123-lacZ transgene is expressed in the aCC and RP2 neurons, similar to the wild-type (R79S84) eve-lacZ transgene. However, in one or two hemisegments per embryo, an extra cell expressing eveK123-lacZ adjacent to the RP2 neuron was observed. This phenotype is very similar to wild-type (R79S84) eve-lacZ expression in klu mutant embryos, although slightly less penetrant. It is concluded that the eveK123-lacZ transgene mimics the klu mutant phenotype, and it is proposed that Klu represses eve expression via direct binding to one or more of these sites (McDonald, 2003).

To further test this hypothesis, gain of function experiments were used to test whether ectopic Klu in GMC4-2a can repress eve-lacZ expression via these sites. Expression of a wild-type (R79S84) eve-lacZ transgene was compared with a transgene containing three mutated Klu consensus binding sites (eveK123-lacZ) in embryos where Scabrous-Gal4 (Sca-Gal4) drives ectopic expression of UAS-klu in all neuroblast lineages. The wild-type (R79S84) eve-lacZ expression is partially repressed by ectopic Klu expression, but the eveK123-lacZ transgene with mutated Klu sites is repressed to a lesser extent. This difference in repression is only observed when the levels of transgene expression are lowered by raising the embryos at 18°C; when the transgenes are more strongly expressed (by raising the embryos at 23°C) no detectable repression was observed. Taken together, Klu loss of function and misexpression studies indicate that Klu acts partly, but not completely, through three predicted Klu-binding sites to repress eve expression in the NB4-2 lineage (McDonald, 2003).

In summary, hkb, ftz, pdm1, and pros are independently required to activate eve expression in GMC4-2a. This suggests that the eve enhancer is capable of integrating the input of all four of these transcription factors to activate transcription. Hb and Ind are also necessary for eve expression in GMC4-2a, but it is not known if they act directly on the eve element or via one of the four transcription factors described in this study. Putative binding sites were found for each of the positively acting transcription factors within the minimal eve element, but mutation of these sites had no effect on expression of the eve-lacZ transgene in embryos (M. Fujioka, J.A. McDonald, and C.Q. Doe, unpublished results reported in McDonald, 2003). It remains to be determined whether Pros, Hkb, Ftz, or Pdm1 activate eve transcription via direct binding to the minimal eve element, or indirectly by activating or facilitating the binding of other transcriptional activators (McDonald, 2003).

Based on functional dissection of the RP2/aCC/pCC eve element, it seems to be composed of three parts. The regions ~7.9 to ~8.2 kb and ~8.2 to ~8.4 kb are each necessary to direct the expression pattern (together they comprise the minimal element for expression in RP2 and aCC), while the region ~8.4 to ~8.6 kb enhances the level of expression. Expression in the pCC neuron is further enhanced by the region extending to ~9.2 kb. The two regions within the minimal element seem to be regulated by different factors, because two copies of ~8.2 to ~8.4 kb (increasing the number of activator binding sites within this region by twofold) could not substitute for the function of the region ~7.9 to ~8.2 kb. This is consistent with the fact that at least four factors are independently required to activate eve in RP2 neurons. How does Klu repress eve expression in GMC4-2b? Negative regulation of eve expression by Klu is due to direct binding to the eve minimal element. (1) It is shown that klu mutants exhibit similar derepression of the eve minimal element transgene and the endogenous eve gene in the NB4-2 lineage; (2) three consensus binding sites are detected for Klu in the eve minimal element (comparison of Drosophila virilis and Drosophila melanogaster shows that the three identified sites are highly conserved); (3) mutation of these sites results in ectopic expression of eve-lacZ in the NB4-2 lineage in wild-type, and (4) mutation of these sites impairs repression of eve-lacZ by ectopic Klu in the NB4-2 lineage. The predicted Klu binding sites (K123) are probably only a subset of relevant Klu binding sites, however, because mutation of the sites gives only partially penetrant phenotypes (McDonald, 2003).

Surprisingly, it was not possible to separate the GMC4-2a/ RP2 element from the GMC1-1a/aCC/pCC element. In both NB 1-1 and NB 4-2 lineages, eve is expressed in the first-born GMC and its neuronal progeny. Both first-born GMCs share expression of several transcription factors, including Pros and Ftz. However, many other transcription factors are differentially expressed, such as the GMC1-1a specific expression of Vnd and Odd-skipped, and the GMC4-2a specific expression of Hkb, Pdm1, and Ind. It is possible that one or more commonly expressed transcription factors are required for expression of eve in both GMC1-1a and GMC4-2a, such as Pros, and this is why the elements cannot be subdivided (McDonald, 2003).

The evolution of eve stripe 2 enhancer is non-clock-like, with important functional differences between closely related species and functional convergence between distantly related species

Lack of knowledge about how regulatory regions evolve in relation to their structure-function may limit the utility of comparative sequence analysis in deciphering cis-regulatory sequences. To address this, reverse genetics was applied to carry out a functional genetic complementation analysis of a eukaryotic cis-regulatory module -- the even-skipped stripe 2 enhancer -- from four Drosophila species. The evolution of this enhancer is non-clock-like, with important functional differences between closely related species and functional convergence between distantly related species. Functional divergence is attributable to differences in activation levels rather than spatiotemporal control of gene expression. These findings have implications for understanding enhancer structure-function, mechanisms of speciation and computational identification of regulatory modules (Ludwig, 2005).

Initially, a fly line, EveDeltaS2E, was created in which the native eve S2E was deleted. Then attempts were made to complement, that is, rescue this lethal mutation with the introduction of a transgene, denoted S2E-EVE, containing an eve S2E from one of the four species (D. melanogaster, D. yakuba, D. erecta, or D. pseudoobscura) linked to a functional eve promoter and coding region. This allowed comparison of both viabilities and developmental consequences among lines differing only in the evolutionary source of their S2E. By genetically manipulating rescue-transgene copy number, effects of Eve abundance on viability and development could also be investigated (Ludwig, 2005).

The eve S2E deficiency mutant was created by removing a 480-bp fragment corresponding to the minimal stripe 2 element (MSE;) from a 15-kb cloned copy of the eve locus. A transgene containing the complete fragment is capable of rescuing eve null mutant flies to fertile adulthood. EveDeltaS2E is functionally a null allele for stripe 2, as evidenced by the expression of the segment polarity gene, engrailed (en). Establishment of en 14-stripe pattern is a complex process that includes involvement by eve early stripes. Eve stripe 2 corresponds to parasegment 3, which is bordered by en stripes 3 and 4. It was hypothesized that these en stripes might be developmental indicators of early eve stripe 2 expression. Indeed EveDeltaS2E embryos lacking a functional S2E produce a short parasegment 3 and vestigial en stripe 4. This defect alone is almost certainly a lethal condition (Ludwig, 2005).

Transgenes containing precisely orthologous S2Es from each of the four species linked to the D. melanogaster eve promoter and coding region were introduced onto the third chromosome. The fragment chosen for investigation was 692 bp in length in D. melanogaster. It contains the central MSE, and every other previously identified TF-binding site in the S2E region. Notably, this fragment contains completely conserved sequences at its 5' and 3' ends in all four species, thus ensuring that precisely orthologous fragments could be compared. As expected, all four S2E-Eve transgenes express a single early eve stripe in the expected spatial location (Ludwig, 2005).

The D. melanogaster S2E rescue transgene, and its considerably diverged D. pseudoobscura ortholog, each restore complete eve stripe 2 biological activity when placed in a genetic background lacking a native S2E. The DNA fragment investigated, therefore, entails both the biological and evolutionary units of enhancer function. This fragment was chosed based on its extensive prior characterization, including genetic, reverse genetic, and footprinting analyses. In particular, transcription factor footprinting data appear to have nicely delineated the functional enhancer (Ludwig, 2005).

Experiments with S2Es of these two species demonstrated that both intact enhancers, but not the chimeras between them, drive the correct spatiotemporal pattern of reporter gene expression (Ludwig, 2000). The rescue experiments reported here extend this finding by showing that the two orthologs are in fact biologically indistinguishable. These new results reinforce the contention that the phenotypic character -- early stripe 2 expression -- must be under stabilizing selection. The character itself remains unchanged over evolutionary time despite substitutions in nearly all the TF-binding sites, the gain and loss of some of them, and considerable change in the spacing between sites. This suggests that unlike proteins, where functional conservation usually means selective constraint on important amino acids (such as the active site of an enzyme), enhancers have a more flexible architecture that allows modification, and perhaps even turnover, of their 'active' sites. Dissimilarities in the structure-function of enhancers and proteins result in different emergent 'rules' of molecular evolution (Ludwig, 2005).

The fact that the S2E fragment from D. erecta is essentially unresponsive to the D. melanogaster morphogen-gradient environment, but the precisely orthologous segment from D. melanogaster (and D. pseudoobscura) responds properly, proves that this fragment must contain evolved differences of functional significance between the species. The lack of biological activity of the D. erecta transgene in D. melanogaster should perhaps come as no surprise, however: Its lower sensitivity to activation may represent the ancestral state of the enhancer. What is surprising is the rapidity with which these functional differences evolve (Ludwig, 2005).

Phylogenetic footprinting of distantly related species can readily identify strongly conserved motifs but runs the risk of not detecting enhancers that have retained their function but have evolved structurally. To overcome this, a technique called phylogenetic shadowing -- the comparison of noncoding sequences among closely related species -- has recently emerged (Boffelli, 2003). The results show that there is no necessary relationship between enhancer phylogenetic (or sequence) relatedness and functional similarity. Closely related species cannot be assumed to be more functionally conserved than distantly related species in enhancer structure-function (Ludwig, 2005).

D. erecta produces a native early eve stripe 2. Why then does the S2E fragment from this species not produce a robust early stripe when placed in D. melanogaster? The first possibility is that the fragment investigated no longer contains a functional enhancer and has been replaced by an equivalent enhancer somewhere else in the eve locus. This possibility can easily be ruled out: the overall architecture of the eve locus, including all of its 5' and 3' enhancers, is well conserved, and there is no new cluster of the appropriate TF-binding sites that could act as a S2E. Another unlikely possibility is that the locus has been duplicated, and the fragment investigated has become functionally inert (i.e., equivalent to a pseudogene). There is no indication of a duplicated eve locus in the D. erecta genome, and all features of the eve locus (including its S2E) are intact and do not indicate any degeneration (Ludwig, 2005).

This leads to the conclusion that the D. erecta fragment used in these experiments contains the S2E. Three additional possibilities can be considered. The first is that this fragment is no longer the complete biological unit, that is, novel binding sites have evolved in this species distal or proximal to this fragment, which have become assimilated into the active enhancer by a process that can be called accretion. patser, a binding-site prediction program (Hertz, 1999) identifies a single potential bicoid-binding site 135 bp upstream of Block-A, the distal end of the D. erecta S2E transgene. This potential site contains an unconventional bicoid-binding motif, TCAATCCC. The next closest potential binding site is another 350 bp further upstream and also has an unconventional binding-site sequence (ACAATCGG). So, although the possibility cannot be ruled out that these are biologically active sites that contribute to S2E activity, they are relatively distant from the recognized S2E (and other bicoid sites), and their sequences do not have the consensus core motif (TAATC). Future experiments will allow formal testing of whether this enhancer has physically expanded. If so, this would be the first documented case for accretion, the adaptive expansion of an enhancer (Ludwig, 2005).

The second possibility is coevolution between the D. erecta S2E and its promoter region, such that it is not capable of driving transcription properly from a D. melanogaster promoter. This is viewed as unlikely for several reasons. (1) Prior to designing these experiments, this issue was investigated with the core promoters and S2Es of D. melanogaster and D. virilis (which is an outgroup to the species studied). No difference could be detected in spatial or temporal expression of each S2E with either promoter (Ludwig, unpublished data). (2) The core promoter regions of D. melanogaster and D. erecta are highly conserved, including complete preservation of both the TATAA and the GAGA site. Indeed there are only four nucleotide differences (and no indels) between the species in a 150-bp stretch containing these sites. (3) One might expect most functional changes in the core promoter to be pleiotropic, given the presence of more than a dozen other separable enhancers in the eve locus, and therefore to be selected against (Ludwig, 2005).

The final possibility is that the D. erecta S2E fragment does contain the entire biological enhancer, but that the trans-acting environment -- the morphogen gradients to which the enhancer responds -- differ between the species, causing the enhancers to have evolved to accommodate the differences. In other words, the sensitivity, or set point, of the binary (on-off) switch function has coevolved with the trans-acting environment in order for the S2E to maintain the appropriate response to evolved activation inputs. This hypothesis implicates in particular: (1) evolutionary shifts in the bicoid and/or hunchback activator gradients. There has been a lineage-specific addition of the functionally required bcd-3 binding site in D. melanogaster that is not present in any of the other species. (2) There is also a lineage-specific loss of the hunchback-1 (hb-1) site in D. erecta (which may be present in its sister taxon, D. yakuba). It is proposed that the lack of sites for these activators, and the presence of a species-specific six-base-pair insertion in the overlapping hb-2/kr-2 binding sites reduces the ability of the D. erecta enhancer to respond to the activator gradients of D. melanogaster (Ludwig, 2005).

This hypothesis predicts stronger activator gradients in D. erecta than in D. melanogaster. Although this possibility has not been investigated directly, it is noted that native eve blastoderm stripes do not reside in the same physical locations in embryos of the two species, but rather are displaced posteriorly in D. erecta compared to D. melanogaster. A similar effect can be mimicked in D. melanogaster with the addition of extra copies of bicoid gene, which shifts the morphogen gradient posteriorly (Ludwig, 2005).

The regulation of development is often modeled as a logic circuit, with cis-regulatory sequences functioning as switches controlling information flow. The long-term functional preservation of both the spatiotemporal and the activation strengths of the D. melanogaster and D. pseudoobscura S2Es speaks to the general conservation of this genetic network in fruit fly development. The results also provide an indication that the stoichiometry of the regulatory components could matter critically for normal development, at odds with theoretical predictions. Epistatic changes accompanying interspecific inviability and sterility may therefore arise more readily as a consequence of quantitative shifts in gene expression than as a result of alterations in the topology of the developmental circuits (Ludwig, 2005).

Prospero, targeting eve and ftz, acts as a binary switch between self-renewal and differentiation in Drosophila neural stem cells

Stem cells have the remarkable ability to give rise to both self-renewing and differentiating daughter cells. Drosophila neural stem cells segregate cell-fate determinants from the self-renewing cell to the differentiating daughter at each division. This study shows that one such determinant, the homeodomain transcription factor Prospero, regulates the choice between stem cell self-renewal and differentiation. The in vivo targets of Prospero have been identified throughout the entire genome. Prospero represses genes required for self-renewal, such as stem cell fate genes and cell-cycle genes. Surprisingly, Prospero is also required to activate genes for terminal differentiation. In the absence of Prospero, differentiating daughters revert to a stem cell-like fate: they express markers of self-renewal, exhibit increased proliferation, and fail to differentiate. These results define a blueprint for the transition from stem cell self-renewal to terminal differentiation (Choksi, 2006).

To identify sites within the Drosophila genome to which Prospero binds, use was made of an in vivo binding-site profiling technique, DamID. DamID is an established method of determining the binding sites of DNA- or chromatin-associated proteins. Target sites identified by DamID have been shown to match targets identified by chromatin immunoprecipitation (ChIP). DamID enables binding sites to be tagged in vivo and later identified on DNA microarrays. In brief, the DNA or chromatin-binding protein of interest is fused to an Escherichia coli adenine methyltransferase (Dam), and the fusion protein is expressed in vivo. The DNA-binding protein targets the fusion protein to its native binding sites, and the Dam methylates local adenine residues in the sequence GATC. The sequences near the protein-DNA interaction site are thereby marked with a unique methylation tag, over approximately 2-5 kilobase pairs (kb) from the binding site. The tagged sequences can be isolated after digestion with a methylation-sensitive restriction enzyme, such as DpnI (Choksi, 2006).

Dam was fused to the N terminus of Prospero, and transgenic flies were generated. The fusion protein is expressed from the uninduced minimal Hsp70 promoter of the UAS vector, pUAST, as high levels of expression of Dam can result in extensive nonspecific methylation and cell death. As a control for nonspecific Dam activity, animals expressing Dam alone were generated. To assess the sites to which Prospero binds during neurogenesis, genomic DNA was extracted from stage 10-11 embryos, approximately 4-7 hr after egg laying (AEL), expressing either the Dam-Prospero fusion protein or the Dam protein alone. The DNA was digested with DpnI and amplified by PCR. DNA from Dam-Prospero embryos was labeled with Cy3, and control DNA with Cy5. The samples were then cohybridized to genomic microarrays. Microarrays were designed that tile the entire euchromatic Drosophila genome. A 60 base oligonucleotide was printed for approximately every 300 bp of genomic DNA, resulting in roughly 375,000 probes on a single array (Choksi, 2006).

Log-transformed ratios from four biological replicates (two standard dye configurations plus two swapped dye configurations) were normalized and averaged. Regions of the genome with a greater than 1.4-fold log ratio (corresponding to approximately a 2.6-fold enrichment) of Dam-Prospero to the control over a minimum of four adjacent genomic probes were identified as in vivo Prospero binding sites. Using these parameters, a total of 1,602 in vivo Prospero binding sites were identified in the Drosophila genome. This work demonstrates that it is possible to map in vivo binding sites across the whole genome of a multicellular organism (Choksi, 2006).

Prospero is known to regulate the differentiation of photoreceptors in the adult eye, and recently sites have been characterized to which Prospero can bind upstream of two Rhodopsin genes, Rh5 and Rh6. A variant of the Prospero consensus sequence is found four times upstream of Rh5 and four times upstream of Rh6. Prospero was shown to bind this sequence in vitro, by band shift assay, and also by a 1-hybrid interaction assay in yeast. In addition, deletion analysis demonstrated that the consensus sequence is required for the Pros-DNA interaction both in vivo and in vitro. It was found that 67% of in vivo binding sites contain at least one Prospero binding motif. Combining in vivo binding-site data with searches for the Prospero consensus sequence reveals 1,066 distinct sites within the Drosophila genome to which Prospero binds during embryogenesis (Choksi, 2006).

A total of 730 genes have one or more of the 1,066 Prospero binding sites located within 1 kb of their transcription unit. Statistical analyses to determine GO annotation enrichment on the members of the gene list that had some associated annotation (519) was performed by using a web-based set of tools, GOToolbox. Using Biological Process (GO: 0008150) as the broadest classification, a list was generated of overrepresented classes of genes (Choksi, 2006).

The three most significant classes of genes enriched in the list of putative Prospero targets are Cell Fate Commitment, Nervous System Development, and Regulation of Transcription. Utilizing GO annotation, it was found that nearly 41% of all annotated neuroblast fate genes (11 of 27) are located near Prospero binding sites and that approximately 9% of known cell-cycle genes are near Prospero binding sites. These include the neuroblast genes achaete (ac), scute (sc), asense (ase), aPKC, and mira and the cell-cycle regulators stg and CycE. In addition, it was found that the Drosophila homolog of the mammalian B lymphoma Mo-MLV insertion region 1 (Bmi-1) gene, Posterior sex combs, is located near a Prospero binding site. Bmi-1 is a transcription factor that has been shown to regulate the self-renewal of vertebrate hematopoetic stem cells. It is concluded that Prospero is likely to regulate neuroblast identity and self-renewal genes as well as cell-cycle genes directly, repressing their expression in the GMC (Choksi, 2006).

Prospero enters the nucleus of GMCs, and its expression is maintained in glial cells but not in neurons . Therefore the list of targets was searched for genes annotated as glial development genes. Prospero binds near 45% of genes involved in gliogenesis. Among the glial genes, it was found that the master regulator of glial development, glial cells missing (gcm), and gilgamesh (gish), a gene involved in glial cell migration, are both near Prospero binding sites and are likely directly activated by Prospero in glia (Choksi, 2006).

In summary, Prospero binds near, and is likely to regulate directly, genes required for the self-renewing neural stem cell fate such as cell-cycle genes. It was also found that Prospero binds near most of the temporal cascade genes: hb, Kruppel (Kr), nubbin (nub/pdm1), and grainyhead (grh) and to genes required for glial cell fate. The in vivo binding-site mapping experiments are supportive of a role for Prospero in regulating the fate of Drosophila neural precursors by directly controlling their mitotic potential and capacity to self-renew (Choksi, 2006).

The Drosophila ventral nerve cord develops in layers, in a manner analogous to the mammalian cortex. The deepest (most dorsal) layer of the VNC comprises the mature neurons, while the superficial layer (most ventral) is made up of the mitotically active, self-renewing neuroblasts. Neuroblast cell-fate genes and cell-cycle genes are normally expressed only in the most ventral cells, while Prospero is found in the nucleus of the more dorsally lying GMCs. If in GMCs, Prospero normally acts to repress neuroblast cell-fate genes and cell-cycle genes, then in a prospero mutant, expression of those genes should expand dorsally. Conversely, ectopically expressed Prospero should repress gene expression in the neuroblast layer.

The neuroblast genes mira, ase, and insc and the cell cycle genes CycE and stg show little or no expression in differentiated cells of wild-type stage 14 nerve cords. Expression of these neuroblast-specific genes was examined in the differentiated cells layer of prospero embryos and it was found that they are derepressed throughout the nerve cord of mutant embryos. mira, ase, insc, CycE, and stg are all ectopically expressed deep into the normally differentiated cell layer of the VNC. To check whether Prospero is sufficient to repress these genes, Prospero was expressed with the sca-GAL4 driver, forcing Prospero into the nucleus of neuroblasts. Prospero expression is sufficient to repress mira, ase, insc, CycE, and stg in the undifferentiated cell layer of the VNC. These data, combined with the Prospero binding-site data, demonstrate that Prospero is both necessary and sufficient to directly repress neuroblast genes and cell-cycle genes in differentiated cells. This direct repression of gene expression is one mechanism by which Prospero initiates the differentiation of neural stem cells (Choksi, 2006).

Having shown that Prospero directly represses genes required for neural stem cell fate, it was asked whether Prospero also directly activates GMC-specific genes. Alternatively, Prospero might regulate a second tier of transcription factors, which are themselves responsible for the GMC fate. Of the few previously characterized GMC genes, it was found that Prospero binds to eve and fushi-tarazu (ftz). In the list of targets several more GMC genes were expected to be found, but not genes involved in neuronal differentiation, since Prospero is not expressed in neurons. Surprisingly, however, it was foudn 18.8% of neuronal differentiation genes are located near Prospero binding sites (Choksi, 2006).

To determine Prospero's role in regulating these neuronal differentiation genes, in situ hybridization was carried out on prospero mutant embryos. Prospero was found to be necessary for the expression of a subset of differentiation genes, such as the adhesion molecules FasciclinI (FasI) and FasciclinII (FasII), which have roles in axon guidance and/or fasciculation. Netrin-B, a secreted protein that guides axon outgrowth, and Encore, a negative regulator of mitosis, also both require Prospero for proper expression. Therefore, in addition to directly repressing genes required for neural stem cell self-renewal, Prospero binds and activates genes that direct differentiation. These data suggest that Prospero is a binary switch between the neural stem cell fate and the terminally differentiated neuronal fate (Choksi, 2006).

To test to what extent Prospero regulates the genes to which it binds, genome-wide expression profiling was carried out on wild-type and prospero mutant embryos. While the DamID approach identifies Prospero targets in all tissues of the embryo, in this instance genes regulated by Prospero were assayed in the developing central nervous system. Small groups of neural stem cells and their progeny (on the order of 100 cells) were isolated from the ventral nerve cords of living late stage 12 embryos with a glass capillary. The cells were expelled into lysis buffer, and cDNA libraries generated by reverse transcription and PCR amplification. cDNA libraries prepared from neural cells from six wild-type and six prospero null mutant embryos were hybridized to full genome oligonucleotide microarrays, together with a common reference sample. Wild-type and prospero mutant cells were compared indirectly through the common reference (Choksi, 2006).

In the group of Prospero target genes that contain a Prospero consensus sequence within 1 kb of the transcription unit, 91 show reproducible differences in gene expression in prospero mutants. Seventy-nine percent of these genes (72) exhibit at least a 2-fold change in levels of expression. Many of the known genes involved in neuroblast fate determination and cell-cycle regulation (e.g., asense, deadpan, miranda, inscuteable, CyclinE, and string) show increased levels in a prospero mutant background, consistent with their being repressed by Prospero. Genes to which Prospero binds, but which do not contain an obvious consensus sequence, are also regulated by Prospero: CyclinA and Bazooka show elevated mRNA levels in the absence of Prospero, as does Staufen, which encodes a dsRNA binding protein that binds to both Miranda and to prospero mRNA (Choksi, 2006).

Expression of genes required for neuronal differentiation is decreased in the prospero mutant cells, consistent with Prospero being required for their transcription. These include zfh1 and Lim1, which specify neuronal subtypes, and FasI and FasII, which regulate axon fasciculation and path finding (Choksi, 2006).

The stem cell-like division of neuroblasts generates two daughters: a self-renewing neuroblast and a differentiating GMC. Prospero represses stem cell self-renewal genes and activates differentiation genes in the newly born GMC. In the absence of prospero, therefore, neuroblasts should give rise to two self-renewing neuroblast-like cells (Choksi, 2006).

The division pattern of individual neuroblasts was studied in prospero mutant embryos by labeling with the lipophilic dye, DiI. Individual cells were labeled at stage 6, and the embryos allowed to develop until stage 17. S1 or S2 neuroblasts were examined, as determined by their time of delamination. Wild-type neuroblasts generate between 2 and 32 cells, producing an average of 16.2 cells. Most of the clones exhibit extensive axonal outgrowth. In contrast, prospero mutant neuroblasts generate between 8 and 51 cells, producing an average of 31.8 cells. Moreover, prospero mutant neural clones exhibit few if any projections, and the cells are smaller in size. Thus, prospero mutant neuroblasts produce much larger clones of cells with no axonal projections, suggesting that neural cells in prospero mutants undergo extra divisions and fail to differentiate (Choksi, 2006).

Recently it was shown, in the larval brain, that clones of cells lacking Prospero or Brat undergo extensive cell division to generate undifferentiated tumors. Given that Prospero is nuclear in the GMC but not in neuroblasts, the expanded neuroblast clones in prospero mutant embryos might arise from the overproliferation of GMCs: the GMCs lacking Prospero may divide like neuroblasts in a self-renewing manner. It is also possible, however, that neuroblasts divide more frequently in prospero mutant embryos, giving rise to supernumerary GMCs that each divide only once. To distinguish between these two possibilities, the division pattern of individual GMCs was followed in prospero mutant embryos (Choksi, 2006).

S1 or S2 neuroblasts were labeled with DiI as before. After the first cell division of each neuroblast, the neuroblast was mechanically ablated, leaving its first-born GMC. All further labeled progeny derive, therefore, from the GMC. Embryos were allowed to develop until stage 17, at which time the number of cells generated by a single GMC was determined (Choksi, 2006).

To determine whether mutant GMCs are transformed to a stem cell-like state, stage 14 embryos were stained for the three neuroblast markers: Miranda (Mira), Worniu (Wor), and Deadpan (Dpn). In wild-type embryos at stage 14, the most dorsal layer of cells in the VNC consists mostly of differentiated neurons. As a result, few or none of the cells in this layer express markers of self-renewal. Mira-, Wor-, and Dpn- expressing cells are found on the midline only or in lateral neuroblasts of the differentiated cell layer of wild-type nerve cords. In contrast, a majority of cells in the differentiated cell layer of stage 14 prospero mutant embryos express all three markers: Mira is found cortically localized in most cells of the dorsal layer of prospero nerve cords; Wor is nuclear in most cells of mutant VNCs; Dpn is ectopically expressed throughout the nerve cord of prospero mutants (Choksi, 2006).

Expression of neuroblast markers in the ventral-most layer of the nerve cord (the neuroblast layer), to exclude the possibility that a general disorganization of cells within the VNC contributes to the increased number of Mira-, Wor-, and Dpn-positive cells in the dorsal layer. The number of neuroblasts in a prospero mutant embryo is normal in stage 14 embryos, as assayed by Wor, Dpn, and Mira expression. Thus, the increased expression of neuroblast markers in prospero mutants is the result of an increase in the total number of cells expressing these markers in the differentiated cell layer. It is concluded that prospero mutant neuroblasts divide to give two stem cell-like daughters. GMCs, which would normally terminate cell division and differentiate, are transformed into self-renewing neural stem cells that generate undifferentiated clones or tumors (Choksi, 2006).

Therefore, Prospero directly represses the transcription of many neuroblast genes and binds near most of the temporal cascade genes: hb, Kruppel (Kr), nubbin (nub/pdm1), and grainyhead (grh), which regulate the timing of cell-fate specification in neuroblast progeny. Prospero maintains hb expression in the GMC, and it has been suggested that this is through regulation of another gene, seven-up (svp). Prospero not only regulates svp expression directly but also maintains hb expression directly. In addition, Prospero maintains Kr expression and is likely to act in a similar fashion on other genes of the temporal cascade. Intriguingly, Prospero regulates several of the genes that direct asymmetric neuroblast division (baz, mira, insc, aPKC). aPKC has recently been shown to be involved in maintaining the self-renewing state of neuroblasts (Choksi, 2006).

Prospero initiates the expression of genes necessary for differentiation. This is particularly surprising since prospero is transcribed only in neuroblasts, not in GMCs or neurons. Prospero mRNA and protein are then segregated to the GMC. Prospero binds near eve and ftz, which have been shown to be downstream of Prospero, as well as to genes required for terminal neuronal differentiation, including the neural-cell-adhesion molecules FasI and FasII. Prospero protein is present in GMCs and not neurons, suggesting that Prospero initiates activation of neuronal genes in the GMC. The GMC may be a transition state between the neural stem cell and the differentiated neuron, providing a window during which Prospero functions to repress stem cell-specific genes and activate genes required for differentiation. There may be few, or no, genes exclusively expressed in GMCs (Choksi, 2006).

Prospero acts in a context-dependent manner, functioning alternately to repress or activate transcription. This implies that there are cofactors and/or chromatin remodeling factors that modulate Prospero's activity. In support of this, although Prospero is necessary and sufficient to repress neuroblast genes, forcing Prospero into the nuclei of neuroblasts is not sufficient to activate all of the differentiation genes to which it binds (Choksi, 2006).

Neuroblasts decrease in size with each division throughout embryogenesis. By the end of embryogenesis, they are similar in size to neurons. A subset of these embryonic neuroblasts becomes quiescent and is reactivated during larval life: they enlarge and resume stem cell divisions to generate the adult nervous system. Neuroblasts in prospero mutant embryos divide to produce two self-renewing daughters but still divide asymmetrically with respect to size, producing a large apical neuroblast and a smaller basal neuroblast-like cell. The daughter may be too small to undergo more than three additional rounds of division during embryogenesis. prospero mutant cells eventually stop dividing, and a small number occasionally differentiate. This suggests that there is an inherent size limitation on cell division. The segregation of Brat, or an additional cell fate determinant, to the daughter cell may also limit the potential of the prospero mutant cells to keep dividing (Choksi, 2006).

The Prox family of atypical homeodomain transcription factors has been implicated in initiating the differentiation of progenitor cells in contexts as varied as the vertebrate retina, forebrain, and lymphatic system. Prospero/Prox generally regulates the transition from a multipotent, mitotically active precursor to a differentiated, postmitotic cell. In most contexts, Prox1 acts in a similar fashion to Drosophila Prospero: to stop division and initiate differentiation (Choksi, 2006).

It is proposed that Prospero/Prox is a master regulator of the differentiation of progenitor cells. Many of the vertebrate homologs of the Drosophila Prospero targets identified in this study may also be targets of Prox1 in other developmental contexts. Prospero directly regulates several genes required for cell-cycle progression, and it is possible that Prox1 will regulate a similar set of cell-cycle genes during, for example, vertebrate retinal development. In addition, numerous Prospero target genes have been identified whose orthologs may be involved in the Prox-dependent differentiation of retina, lens, and forebrain precursors (Choksi, 2006).

Spatial specificity of mesodermal even-skipped expression relies on multiple repressor sites

Individual cardiac progenitors emerge at defined positions within each segment in the trunk mesoderm. Their specification depends on segmental information from the pre-patterned ectoderm, which provides positional information to the underlying cardiac mesoderm via inductive signals. This pattern is further reinforced by repressive interactions between transcription factors that are expressed in neighboring sets of cardiac progenitors. For example, even-skipped (eve) and ladybird early (lbe) gene products mark adjacent cardiac cell clusters within a segment, and their antagonistic interaction results in mutually exclusive expression domains. Lbe acts directly on the eve mesodermal enhancer (eme) to participate in restricting its expression anteriorly. It is hypothesized that additional repressive activities must regulate the precise pattern of eve expression in the cardiac mesoderm via this enhancer. In this study, two additional repressor motifs: 4 copies of an 'AT'-rich motif (M1a-d) and 2 copies of an 'GC'-rich motif (M2a,b), were identified which when mutated cause expansion of eme-dependent reporter gene expression. Potential negative regulators of eve and were examined and it was found that their overexpression is sufficient to repress eve as well as the eme enhancer via these sites. These data suggest that a combination of factors is likely to interact with multiple essential repressor sites to confer precise spatial specificity of eve expression in the cardiac mesoderm (Liu, 2008).

Although each of the identified repressor sites is necessary, none is individually sufficient for restricting the eme enhancer activity to the eve expression domain. Several additional homeodomain proteins, including Msh, C15 and Lim3, are capable of repressing mesodermal eve expression by interacting with specific sites within the enhancer element. While the repression of mesodermal eve expression by Msh, C15 and Lim3 is likely mediated by the AT-rich M1 sites and the Lb2 site, the repression of eve expression by Lbe requires both the AT-rich M1 and the Lb2 sites as well as the GC-rich M2a site. Therefore, each of the four repressor sites apparently is required on order to confer sensitivity to repression by Lbe. This raises the possibility that repression is the result of a complex in which the cooperation of all four repressor elements is required for successful repression (Liu, 2008).

A prominent feature of the Drosophila is its segmental polarity that includes distinct cardiac cell types that are precisely positioned within each segment. These cardiac progenitors are specified along the anterior-posterior axis during development and are marked by Lbe, Eve or Svp. As the embryo develops, a linear heart tube is formed and this metameric arrangement of cardiac cells types continues to be maintained. Within each hemi-segment, the anterior two pairs form the tinman-expressing 'working myocardium', while the posterior pair that expresses svp and the T-box transcriptional factor Doc form the ostia. Previous studies suggested that repressive interactions between cardiac factors expressed in non-overlapping subtypes of cardiac cells likely contribute to the diversification and maintenance of cellular identities. Svp and Tin have been shown to repress each other's expression during heart tube formation, and the current data suggest that antagonistic interactions between Lbe and Eve are also a part of this mutual repression network. In addition, the data show that eve expression within the cardiac mesoderm is negatively regulated by multiple repressor sites, thus further supporting the idea that transcriptional repression mechanisms play a prominent role in the generation of cellular diversity in the developing heart. Roles were demonstrated for two potential repressors, C15 and Lim3. Although they do not seem to be essential for patterning mesodermal eve expression, they are normally expressed in the cardiac mesoderm in the vicinity of the Eve cells and they do repress the eme enhancer via the identified repressor sites when ectopically expressed. Therefore, it is also possible that they function redundantly with other negative regulators yet to be identified (Liu, 2008).

Default repression is a common mechanism utilized by major signaling pathways, including Wnt, Shh and Notch pathways, to restrict target gene expression. In the absence of signaling, signal-regulated transcription factors function mainly as transcriptional repressors, thus preventing low levels of target gene expression that might be activated by weak, local activators ('default repression'). In response to signals, some transcription factors are then converted into transcriptional activators to promote target gene expression. Thus, transcriptional repression and activation can be mediated by the same binding sites. Default repression mechanisms may also contribute to the restricted mesodermal eve pattern. It has been reported that mesodermal eve expression is under the direct transcriptional control of Wg signaling. Mutating several putative binding sites for dTCF, the transcriptional mediator of Wg signaling, results in an expansion of low-level reporter gene expression within the cardiac mesoderm that is unaffected by reduced wg activity. Thus, dTCF may serve as a default signal to restrict mesodermal eve expression in the absence of wg signaling (Liu, 2008).

It has been shown that Hh signaling not only promotes eve and svp but also inhibits lbe expression in the dorsal mesoderm. One mechanism for Hh signaling may be via inhibition of Cubitus interruptus (Ci)-mediated repression. Interestingly, there is some similarity between the M2a sequence examined in this study (TGGGCCCT) and the consensus sequence for Ci (TGGGTGGTC). This raises the interesting possibility that M2a site may be a putative Ci binding site in eme. Thus, mutations of M2a site, which result in the anterior expansion of eme activity into Lbe expressing cells, may reflect a lack of repression by Ci. Alternatively, the M2a site may mediate transcriptional repression by Lbe or its potential cofactors. The latter hypothesis is more consistent with the observation that reporter gene expression is rendered insensitive to inhibition by Lbe overexpression when the M2a site is mutated in eme. As the M2a site does not resemble the Lbe consensus sequence, the idea is favored that another factor binds to the M2a site, which then cooperates with Lbe in repressing mesodermal eve expression. This interaction may be facilitated by the close proximity of the two sites (Liu, 2008).

In sum, the in vivo functional dissection of eme has revealed that each of two AT-rich sites, M1b or M1c and the previous studied Lb2 site, when mutated, causes reporter gene expansion that encompasses the entire cardiac mesoderm, overlapping with Tinman protein at late stage 12. In addition, the GC-rich site M2a is required for repression anterior to the Eve cluster. The absolute requirement of each repressor site for successful restriction of eve expression within the cardiac mesoderm is in striking contrast to the mechanism of incremental activation of this enhancer in the cardiac mesoderm by activators such as Tinman, dTCF, Mad, E-box and ETS sites. Repression through these repressor sites may require cooperation between the sites, perhaps via a repressor complex. Thus, eliminating the function of any of these sites will disrupt interactions with the complex causing de-repression within the 'activator'-dependent cardiac mesoderm (Liu, 2008).

Rearrangements of 2.5 kilobases of noncoding DNA from the Drosophila even-skipped locus define predictive rules of genomic cis-regulatory logic

Rearrangements of about 2.5 kilobases of regulatory DNA located 5' of the transcription start site of the locus generate large-scale changes in the expression of stripes 2, 3, and 7. The most radical effects are generated by juxtaposing the minimal stripe enhancers MSE2 and MSE3 for stripes 2 and 3 with and without small ‘spacer’ segments less than 360 bp in length. These fusion constructs were placed in a targeted transformation site and quantitative expression data were obtained for these transformants together with their controlling transcription factors at cellular resolution. These data demonstrated that the rearrangements can alter expression levels in stripe 2 and the 2-3 interstripe by a factor of more than 10. It was reasoned that this behavior would place tight constraints on possible rules of genomic -regulatory logic. To find these constraints, the new expression data together with previously obtained data on other constructs were confronted with a computational model. The model contained representations of thermodynamic protein-DNA interactions including steric interference and cooperative binding, short-range repression, direct repression, activation, and coactivation. The model was highly constrained by the training data, which it described within the limits of experimental error. The model, so constrained, was able to correctly predict expression patterns driven by enhancers for other genes; enhancers not included in the training set; stripe 2, 3, and 7 enhancers from various Drosophilid and Sepsid species; and long segments of regulatory DNA that contain multiple enhancers. The model further demonstrated that elevated expression driven by a fusion of MSE2 and MSE3 was a consequence of the recruitment of a portion of MSE3 to become a functional component of MSE2, demonstrating that -regulatory ‘elements’ are not elementary (Kim, 2003).

This work has gone beyond modeling individual experimentally identified enhancers, and has done so at a level of resolution comparable to that required for organismal survival. Although previous previous work with a version of this model not incorporating cooperativity or coactivation was comparably accurate and capable of representing stripe 7 expression driven by sequences outside of the 3/7 enhancer, the modeled DNA contained only one classical enhancer, S2E. In contrast, the expression data used in the present study not only involved two enhancers, but more importantly dealt with a situation in which the function of these enhancers was critically altered by juxtaposing them and thus altering their function. These rearrangements provided a powerful constraint on the possible rules of transcriptional control, as demonstrated by the prediction of expression patterns seen in this study. Finally, the model can be used as an analytic tool with which to understand how multiple transcriptional mechanisms operate simultaneously to produce observed patterns of expression (Kim, 2003).

Highly precise experimental data made this study possible, and their importance cannot be overemphasized. The inherent transcriptional machinery is exquisitely precise, and fundamental understanding of its functioning requires data at a cellular level of precision. The dataset has that level of precision because simultaneous staining of reporter-driven lacZ expression and native Eve protein was performed, allowing registration pf the reporter data with the full TF dataset. The intrinsic variability of gene expression prevents such registration by measurements of the position of reporter expression alone. This point illuminates a problem regarding the current unbalanced state of technology in genomics. Sequence can be obtained readily and cheaply. Yet, the inability to monitor gene expression at cellular resolution in a high throughput manner together with a lack of understanding of the code for regulatory logic has in general limited genomic level investigations of regulatory DNA to statistical association studies. The work reported in this study was made possible by a high resolution dataset created over many years. Although the data was quantitated using high throughput methods, staining and microscopy were carried out manually (Kim, 2013).

The quality of fit to the training data indicates that the model is reasonably complete for the stripe 2 and 3 eve enhancers at the developmental time assayed. Previous attempts to model both stripes simultaneously failed, most probably because of a failure to incorporate coactivation of Hb by Bcd and Cad. Further support for the current model is afforded by its predictive capability. In melanogaster, accurate predictions were obtained for expression driven by the stripe 5 and 4_6 enhancers. It was also possible to correctly predict the effects of site-directed mutations affecting only 2-6 base pairs. This result indicates that the model might ultimately have utility in predicting the effects of SNPs, a point with implications for both medicine and evolutionary biology (Kim, 2013).

With respect to stripes 2, 3, and 7 in non- melanogaster species there are no contradictions to available experimental results. This is a strong indication that major elements of the fundamental rules of transcription have been uncovered, as these diverged enhancers have considerable turnover in binding site composition among the Drosophilids and no homology except for short sequences involving overlapping binding sites in. In fact, enhancers from only 4 Drosophilid and 3 Sepsid species have been qualitatively assayed by transformation into melanogaster, so that a rich set of quantitative predictions is finished that can be examined in future experiments (Kim, 2013).

With respect to predictions of the expression of other Drosophila genes, good results were obtained for the h 3_4 and run 3_7 enhancers. The predicted run 1_7 enhancer pattern had better registration of stripe 7 with protein pattern than predicted for 3_7, with the strange result that the predicted pattern is in perfect alignment with run stripe 2 rather than stripe 1. This last prediction may be erroneous. Although no published co-staining data is available of the run 1_7 enhancer with native run protein or RNA, such data exists for a larger segment of DNA which drives run stripes 1, 3, and 5 and contains run 3_7. With respect to gap genes, this study has good agreement of predicted patterns for the hb pThb1 and Kr CD1 enhancers, but the agreement is poorer for other Kr and hb enhancers, kni, and gt. In the case of gt, the lack of expression in the native domain is a consequence of the presence of numerous Gt binding sites. There are indications that Gt has autoactivation activity. It is possible that Gt has a coactivator on its own promoter that was not included in this study (Kim, 2003).

Although enhancers are frequently referred to as cis-regulatory 'elements', they are not elementary or fundamental objects. They are not elementary because they do not have well-defined boundaries. This study has demonstrated the context-dependent border of MSE2 by showing that the increased level of stripe 2 expression in M32 was a consequence of the recruitment of 40% of MSE3 to become a functional component of MSE2. Moreover, MSE2 and S2E both drive stripe 2 and can rescue lethality, and MSE2 is not completely minimal in the sense that smaller regions of DNA within it can drive weak and variable stripe 2 expression. Enhancers are not functionally fundamental objects because most enhancers drive expression domains which are similar to but not identical with those driven by the intact locus. Complete fidelity requires additional sequences. With respect to eve stripe 3, this point has been evident for some time in mutant genotypes, although the additional sequences required are as yet unidentified. In the case of hb, the lack of fidelity is evident in wild type and complete fidelity is restored by a shadow enhancer. The real challenge in regulatory genomics is the prediction of expression from an entire locus (Kim, 2003).

The ability to model expression of the fusion constructs and to predict expression of stripes 2, 3, and 7 driven by 5' noncoding sequence and stripes 4, 5, and 6 by eve 3' noncoding sequence demonstrates that the applicability of the model is not limited to previously identified enhancers. These results support an idea advanced by Gray, Levine, and coworkers that short range repression is required for the independent action of multiple enhancers and theory indicates that eve stripes are generated by repression from gap genes. Because gap gene expression domains are wider than eve stripes, silencing from these genes would result in a repressed region comparable in size to that of a gap domain and could not produce the observed stripes (Kim, 2003).

Predictions of expression driven by large DNA segments are less clean than those of single enhancers in the sense that they required hand tuning of the threshold to prevent completely saturated expression domains comprising stripes 2-3 and 4-6 respectively. This saturation appears to involve a lack of balance between activators and repressors as the length of modeled DNA increases, but it is not possible at this time to distinguish between problems with the model and the training data. With respect to the model, this lack of balance may stem from the unlimited range of activators and the limited range of quenchers. In order to know whether this model property is biologically correct or incorrect, it is necessary to quantitatively determine how the amplitude of a given stripe changes as it is driven by larger DNA fragments. This point is not captured in training data because only the four fusion constructs, all of similar total length, were transformed to a targeted site. Shorter and longer DNA fragments were not targeted transformants and hence required a free parameter scaling the amplitude to account for position effect. The quantitative characterization of expression driven by fragments of varying size transformed to a common chromosomal site is an important experimental task for future work. It will also be important to generate rescue constructs containing both native and lacZ message in order to standardize between observed levels of native and reporter transcripts. It is believed that the results in this paper, while incomplete, demonstrate the feasibility of constructing a precise, quantitative, and predictive model of an entire locus that would also account for its enhancer structure (Kim, 2003).

In conclusion, this model demonstrates that short-range quenching and coactivation are essential mechanisms conferring independent action of enhancers in the large even-skipped regulatory DNA. No decisive evidence was found that the length scales over which these interactions occur are fundamentally different. Short range quenching had a length scale of 150 bp, set from published experiments. The length scale of coactivation of Hb by Bcd was almost exactly the same, despite it being allowed to vary in the fitting procedure. These mechanisms are clearly necessary for understanding the regulation of the entire eve locus, and establishing their sufficiency will be the subject of future work. In the case of both mechanisms it is expected that better knowledge of phenomenology would lead to superior understanding. For example, Arnosti's group has produced greatly improved data on short range repression that suggests periodic behavior in limitations exist not only for the data but also for the model. Alternatively, it might be more useful to reduce the number of parameters by constraining the range and functional form of all short range interactions to be identical. Such a choice would reflect a picture in which the scale of all short range interactions are set by the length of DNA associated with a single nucleosome (160-240 bp). Fixing this length scale based on structural considerations would connect the model presented in this paper with an important body of data (Kim, 2003).

Predictions of expression patterns from many Drosophilidae and Sepsidae strongly suggest that the fundamental rules of metazoan transcription are well conserved over the course of evolution. As a syncytium, the Drosophila blastoderm is very specialized as a developmental system but there is no reason to think that transcription in this system operates differently than in the rest of the metazoa. As yet there are two barriers that must be crossed to establish a general theory of eukaryotic transcriptional control. One is experimental -- training data require not only expression levels and regulatory sequence, but also the concentrations of TFs. Another is theoretical -- a framework is needed to understand long range interactions in the chromatin (Kim, 2003).

The Drosophila eve Insulator Homie Promotes eve Expression and Protects the Adjacent Gene from Repression by Polycomb Spreading

The Drosophila even skipped (eve)gene has a Polycomb-group response element (PRE) at one end, flanked by an insulator, an arrangement also seen in other genes. This study show that this insulator has three major functions. It blocks the spreading of the eve Pc-silenced region, preventing repression of the adjacent gene, TER94. It prevents activation of TER94 by eve regulatory DNA. It also facilitates normal eve expression. When the insulator Homie is deleted in the context of a large transgene that mimics both eve and TER94 regulation, TER94 is repressed. This repression depends on the eve PRE. Ubiquitous TER94 expression is 'replaced' by expression in an eve pattern when Homie is deleted, and this effect is reversed when the PRE is also removed. Repression of TER94 is attributable to spreading of the eve Pc-silenced domain into the TER94 locus, accompanied by an increase in histone H3 trimethylation at lysine 27. Other PREs can functionally replace the eve PRE, and other insulators can block PRE-dependent repression in this context. The full activity of the eve promoter is also dependent on Homie, and other insulators can promote normal eve enhancer-promoter communication. These data suggest that this is not due to preventing promoter competition, but is likely the result of the insulator organizing a chromosomal conformation favorable to normal enhancer-promoter interactions. Thus, insulator activities in a native context include enhancer blocking and enhancer-promoter facilitation, as well as preventing the spread of repressive chromatin (Fujioka, 2013).

3D chromatin interactions organize Yan chromatin occupancy and repression at the even-skipped locus

Long-range integration of transcriptional inputs is critical for gene expression, yet the mechanisms remain poorly understood. This study investigated the molecular determinants that confer fidelity to expression of the heart identity gene even-skipped (eve). Targeted deletion of regions bound by the repressor Yan defined two novel enhancers that contribute repressive inputs to stabilize tissue-specific output from a third enhancer. Deletion of any individual enhancer reduced Yan occupancy at the other elements, impacting eve expression, cell fate specification, and cardiac function. These long-range interactions may be stabilized by three-dimensional chromatin contacts that were detected between the elements. This work provides a new paradigm for chromatin-level integration of general repressive inputs with specific patterning information to achieve robust gene expression (Webber, 2013b).

In addition to the muscle/heart enhancer targeted by the Yan–Pnt switch (MHE), a genome-wide occupancy study of Yan mapped three other regions to the eve locus. The four Yan-bound domains are referred as D1, D2, D3/MHE, and D4. To explore the contribution of individual Yan-bound regions to mesodermal eve expression specific deletions were recombineered into a functional genomic bacterial artificial chromosome (BAC) construct encompassing the entire 16.4-kb eve locus carrying an in-frame YFP tag. As enhancer-blocking activity has previously been identified between the MHE and Ter94 gene, it was reasoned that the D4 Yan-bound element was more likely to contribute to Ter94 than to eve regulation, and it was excluded from the analysis. Transgenes were generated, and altered function was assessed by examining expression in the Eve-positive cardiogenic precursors. The simple prediction for deletion of a functional Yan-bound region is that loss of repression should increase expression of the target gene; however, if the deletion also removes binding sites for critical activators, then reduced expression is expected. Either scenario predicts reduced ability to complement an eve-null. Alternatively, if elements are functional yet fully redundant under optimal conditions, no change in expression or viability is predicted (Webber, 2013b).

The control eve-YFP transgene fully complemented an eve-null background, with 86% rescue to adulthood, and was expressed in a pattern identical to that of endogenous eve. Thus, in stage 11 embryos, Eve-YFP was prominent in the muscle and cardiogenic precursors as well as in clusters of neurons in the CNS. Deletion of the D1, D2, or D3 Yan-bound regions altered expression and reduced genetic rescue efficiency, indicating important contributions to regulation of eve. As predicted based on loss of key activating inputs, deletion of the D3/MHE region significantly reduced mesodermal expression, with EveΔD3-YFP expression detectable, on average, in only one cell per cluster. For the EveΔD1-YFP and EveΔD2-YFP transgenes, although expression in the mesoderm appeared qualitatively normal, quantification of the YFP signal revealed significant increases in both the mean intensity and embryo-to-embryo variation relative to the wild-type Eve-YFP control. The reduced ability of the deletion transgenes to rescue an eve-null suggests that these Yan-bound elements are not redundant. The deleted regions do not include known stripe enhancers, and cuticle preparations of dead embryos from the rescue experiments did not show segmentation defects. Thus, the reduced viability is unlikely attributed to axial patterning defects (Webber, 2013b).

Precise regulation of eve expression in the cardiogenic mesoderm is essential for acquisition of heart cell identity and ultimately for cardiac function. Consistent with this, the elevated expression of EveΔD1-YFP reflected in part an increase in the number of Eve-positive mesodermal cells specified, raising the possibility that compromised heart development might contribute to reduced fitness. To test this, pupal heart rates were measured. In agreement with a previous study, the heart rate of eve−/−; eveΔD3-YFP animals was significantly reduced relative to the eve−/−; eve-YFP control. eve−/−; eveΔD1-YFP larvae also had a reduced heart rate, suggesting that the ectopic cell fate specification detected compromises the ensuing developmental program. Deletion of the D2 region did not significantly change the number of Eve-positive mesodermal cells specified, and the heart rate measured in eve−/−; eveΔD2 pupae was comparable with that of the control (Webber, 2013b).

Adult heart function is commonly assessed with assays that measure the general vigor and activity of the fly. In both a flight assay and an assay measuring the negative geotaxis response before and after heat stress, eve−/−; eveΔD3-YFP and eve−/−; eveΔD1-YFP flies again performed poorly relative to eve−/−; eve-YFP control animals. eve−/−; eveΔD2-YFP animals exhibited a response intermediate to that of the eve−/−; eveΔD1-YFP and control eve−/−; eve-YFP animals in the geotaxis assay prior to heat stress. However, after stress, the activity of eve−/−; eveΔD2-YFP flies was significantly below that of the control. Together, these results confirm the physiological importance of the Yan-bound D1 region and support the hypothesis that it mediates repressive inputs important for stabilizing mesodermal eve expression within levels required to support normal cell fate acquisition during heart development. As the D2 deletion partially overlaps a neuronal enhancer, further analysis will be required to determine whether the reduced fitness and activity result from either compromised neural function, subtle cardiac defects, or a combination of the two (Webber, 2013b).

To obtain more direct evidence that compromised Yan-mediated repression contributes to increased expression and reduced fitness of the deletion mutants, the effectiveness was tested of twist-Gal4-driven overexpression of a constitutively active form of Yan, YanACT, in repressing expression of the deletion transgenes. If Yan-mediated repressive inputs act exclusively on the MHE, then YanACT should repress mesodermal expression of EveΔD1-YFP and EveΔD2-YFP but not EveΔD3-YFP. Alternatively, if Yan-mediated regulation across multiple genomic elements contributes to stable eve expression, then all deletion transgenes should be sensitive to YanACT. The results support the latter scenario, as YanACT fully repressed all three eve-YFP deletion transgenes (Webber, 2013b).

It was next asked whether heterozygosity for yan or pnt might, respectively, enhance or suppress the loss of robustness measured for the eve deletion constructs. Using either a null allele or a small deficiency, it was found that heterozygosity for yan increased the mean intensity and variation of wild-type Eve-YFP expression and further increased the intensity of EveΔD2-YFP expression. While yan heterozygosity did not significantly increase EveΔD1-YFP expression, the number of Eve-positive cells specified was increased. These changes also impact fitness, as yan heterozygosity further reduced the genetic rescue efficiency of the eve transgenes. Conversely, heterozygosity for pnt suppressed the increased intensity and variability of EveΔD1-YFP and EveΔD2-YFP expression. A reduced pnt dose also improved the negative geotaxis response following heat shock. Together, these results suggest that the D1 and D2 elements are required for robust regulation of eve expression in the muscle and cardiogenic precursors such that loss of Yan-mediated repression upon deletion of either element permits inappropriate Pnt-mediated activation of eve, presumably through the MHE, leading to increased and more variable expression. Such destabilization of the Yan–Pnt switch makes the system prone to aberrant cell fate specification and ultimately reduces fitness (Webber, 2013b).

To explore further the properties of the D1 and D2 Yan-bound regions, it was asked whether they define autonomous cis-regulatory elements or whether their activity requires the endogenous genomic context. Reporter constructs carrying sequences spanning the chromatin immunoprecipitation (ChIP)-defined Yan-bound regions were generated, and expression was examined in stage 11 embryos. In contrast to the control MHE reporter that expressed specifically in the expected mesodermal clusters, the D1 and D2 reporters showed weak ubiquitous expression. A slightly stronger signal was detected in the mesoderm for both reporters but in domains broader than the three cell clusters seen with the MHE reporter. Removing yan increased the intensity but not the specificity of the expression patterns of the D1 and D2 reporters, suggesting that the individual elements can be recognized and repressed by Yan outside of the context of the intact eve locus. The autonomous ability of each element to recruit Yan was confirmed by ChIP-PCR, although quantitative PCR (qPCR) analysis revealed that occupancy was reduced relative to that measured at the endogenous locus (Webber, 2013b).

Because the D1 and D2 Yan-bound elements do not appear to provide independent additive inputs capable of directing mesodermal eve expression, it was asked whether their primary role might be to integrate Yan-repressive inputs across the intact locus. It was hypothesized that long-range interactions between these regions might stabilize Yan binding, thereby providing a buffering mechanism that sets the precise MHE-driven eve expression pattern needed for accurate cell fate specification. If correct, then deletion of any individual Yan-bound region not only should abolish occupancy at the site of deletion, but might also perturb binding at the remaining elements. To test this, Yan occupancy was assessed by ChIP-qPCR in embryos homozygous for the eve deficiency and carrying the deletion transgenes. Supporting the model of coordinated stabilization of Yan occupancy, deletion of any individual Yan-bound domain reduced Yan occupancy by about twofold at the remaining two eve elements but did not alter enrichment at a distant locus. Thus, precise regulation of mesodermal eve expression relies on long-range coordination between cis-regulatory elements rather than additive inputs from autonomous enhancers (Webber, 2013b).

One possible mechanism is that the 3D chromatin environment facilitates interactions between the D1, D2, and D3 elements. Using proximity ligation and PCR analysis (referred to as chromatin conformation capture [3C]), a long-range interaction was detected between a D1–D2 fragment and the D3/MHE fragment. Sequencing confirmed the expected identity of the amplified ligated fragment. Control primers located just outside of the eve locus did not yield a 3C product with the D1–D2 fragment. Furthermore, the 3C interaction was not detected in adult heads, suggesting context-specific chromatin contacts. Finer-scale mapping in both wild-type and yan mutant embryos should provide further insight into how the 3D chromatin environment facilitates cooperative recruitment of Yan across the locus and how this in turn both stabilizes the Yan–Pnt switch in optimal conditions and buffers against genetic or environmental variation that might limit Yan (Webber, 2013b).

In conclusion, the results suggest a novel regulatory mechanism in which, rather than contributing patterning information, the D1 and D2 enhancers act to dampen the expression control mediated by the D3/MHE. More broadly, this raises the possibility that even when transcription factor-bound regions identified through ChIP sequencing (ChIP-seq) fail to give clear expression patterns as autonomous reporters, they may still contribute to gene regulation in vivo. In the case of Yan, it is speculated that its intrinsic polymerization ability allows it to exploit the 3D chromatin environment to organize long-range repressive complexes that coordinate information across multiple enhancers (Webber, 2013).

Many parallels can be drawn between Yan/Pnt regulation of the transition from uncommitted progenitor to specified cell fate within a developing tissue and the control of the pluripotent state in mammalian cells. Given the potential for natural, genetic, or environmental variation to perturb the timing, accuracy, or stability of such transitions, mechanisms to impart robustness to gene expression are almost certainly required and very likely conserved across species. Thus, the regulatory paradigm established in this study may be broadly relevant (Webber, 2013b).

Continued: even-skipped Transcriptional regulation part 2/3 | part 3/3


Interactive Fly, Drosophila even-skipped: Biological Overview | Evolutionary Homologs | Targets of activity | Post-transcriptional regulation | Protein interactions | Developmental Biology | Effects of Mutation | References

Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.

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