Mechanism and Bicoid-dependent control of hairy stripe 7expression in the posterior region of the Drosophila embryo
1997; Springer Nature; Volume: 16; Issue: 14 Linguagem: Inglês
10.1093/emboj/16.14.4403
ISSN1460-2075
Autores Tópico(s)Plant Molecular Biology Research
ResumoArticle15 July 1997free access Mechanism and Bicoid-dependent control of hairy stripe 7 expression in the posterior region of the Drosophila embryo Anna La Rosée Anna La Rosée Abteilung Molekulare Entwicklungsbiologie, Max-Planck-Institut für biophysikalische Chemie, Am Fassberg, D-37077 Göttingen, Germany Search for more papers by this author Thomas Häder Thomas Häder Abteilung Molekulare Entwicklungsbiologie, Max-Planck-Institut für biophysikalische Chemie, Am Fassberg, D-37077 Göttingen, Germany Search for more papers by this author Heike Taubert Heike Taubert Abteilung Molekulare Entwicklungsbiologie, Max-Planck-Institut für biophysikalische Chemie, Am Fassberg, D-37077 Göttingen, Germany Search for more papers by this author Rolando Rivera-Pomar Rolando Rivera-Pomar Abteilung Molekulare Entwicklungsbiologie, Max-Planck-Institut für biophysikalische Chemie, Am Fassberg, D-37077 Göttingen, Germany Search for more papers by this author Herbert Jäckle Corresponding Author Herbert Jäckle Abteilung Molekulare Entwicklungsbiologie, Max-Planck-Institut für biophysikalische Chemie, Am Fassberg, D-37077 Göttingen, Germany Search for more papers by this author Anna La Rosée Anna La Rosée Abteilung Molekulare Entwicklungsbiologie, Max-Planck-Institut für biophysikalische Chemie, Am Fassberg, D-37077 Göttingen, Germany Search for more papers by this author Thomas Häder Thomas Häder Abteilung Molekulare Entwicklungsbiologie, Max-Planck-Institut für biophysikalische Chemie, Am Fassberg, D-37077 Göttingen, Germany Search for more papers by this author Heike Taubert Heike Taubert Abteilung Molekulare Entwicklungsbiologie, Max-Planck-Institut für biophysikalische Chemie, Am Fassberg, D-37077 Göttingen, Germany Search for more papers by this author Rolando Rivera-Pomar Rolando Rivera-Pomar Abteilung Molekulare Entwicklungsbiologie, Max-Planck-Institut für biophysikalische Chemie, Am Fassberg, D-37077 Göttingen, Germany Search for more papers by this author Herbert Jäckle Corresponding Author Herbert Jäckle Abteilung Molekulare Entwicklungsbiologie, Max-Planck-Institut für biophysikalische Chemie, Am Fassberg, D-37077 Göttingen, Germany Search for more papers by this author Author Information Anna La Rosée1, Thomas Häder1, Heike Taubert1, Rolando Rivera-Pomar1 and Herbert Jäckle 1 1Abteilung Molekulare Entwicklungsbiologie, Max-Planck-Institut für biophysikalische Chemie, Am Fassberg, D-37077 Göttingen, Germany The EMBO Journal (1997)16:4403-4411https://doi.org/10.1093/emboj/16.14.4403 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Pair-rule gene hairy (h) expression in seven evenly spaced stripes, along the longitudinal axis of the Drosophila blastoderm embryo, is mediated by a modular array of separate stripe enhancer elements. The minimal enhancer element, which generates reporter gene expression in place of the most posterior h stripe 7 (h7-element), contains a dense array of binding sites for factors providing the trans-acting control of h stripe 7 expression as revealed by genetic analyses. The h7-element mediates position-dependent gene expression by sensing region-specific combinations and concentrations of both the maternal homeodomain transcriptional activators, Caudal and Bicoid, and of transcriptional repressors encoded by locally expressed zygotic gap genes. Caudal and Bicoid, which form complementing concentration gradients along the longitudinal axis of the embryo, function as redundant activators, indicating that the anterior determinant Bicoid is able to activate gene expression in the most posterior region of the embryo. The spatial limits of the h stripe 7 domain are brought about by the local activities of repressors which prevent activation. The results suggest that the gradients of Bicoid and Caudal combine their activities to activate segmentation genes along the entire axis of the embryo. Introduction The spatial organization of the Drosophila body is indicated by the expression of pair-rule genes in a series of seven sharp transverse stripes along the anterior–posterior axis of the blastoderm embryo (Ingham, 1988; Pankratz and Jäckle, 1990). Dissection of the genetic requirements leading to the proper formation and positioning of the stripe expression domains revealed that pair-rule genes, such as even-skipped (eve) and hairy (h), require multiple regulatory inputs of transcription factors encoded by the maternal anterior determinant bicoid (bcd) and the zygotic gap genes hunchback (hb), Krüppel (Kr), knirps (kni) and giant (gt) (for review, see Pankratz and Jäckle, 1993). The homeodomain protein Bicoid forms an anterior–posterior concentration gradient which arises from pre-localized maternal mRNA in the anterior pole of the egg (Berleth et al., 1988; Driever and Nüsslein-Volhard, 1988). Bicoid and its posterior counterpart, the homeodomain protein encoded by caudal (cad), initiate the zygotic expression of the gap genes in spatially restricted adjacent domains along the pre-blastoderm embryo (for review, see Rivera-Pomar and Jäckle, 1996). This leads to a region-specific scenario whereby different combinations and concentrations of transcription factors provide positional information which is decoded by the cis-acting control regions of the subordinate pair-rule genes. Recent studies have identified the specific trans-acting requirement for the generation of eve, h, fushi tarazu (ftz) and runt (run) stripe expression along the longitudinal axis of the blastoderm embryo (for review, see Pankratz and Jäckle, 1993). ftz expression depends on a small cis-acting element which directs the expression of all stripes (Hiromi et al., 1985; Hiromi and Gehring, 1987) and run expression on a large cis-acting region that cannot be separated into individual stripe elements (Klingler et al., 1996). In contrast, the eve and h genes contain large upstream control regions comprising autonomous regulatory modules (Howard et al., 1988; Goto et al., 1989; Pankratz et al., 1990; Small et al., 1991, 1992). Each module acts as a 'stripe element' to decode the positional information provided by the local combinations of transcription factors and to mediate gene expression in a specific stripe domain. To date, the best studied example of pair-rule stripe expression is eve stripe 2, a transverse stripe which covers a band of three to four peripheral nuclei in the anterior third of the syncytial blastoderm (Frasch et al., 1987; Small et al., 1991, 1992). eve stripe 2 expression depends on two activators, Bicoid and Hunchback (Small et al., 1991). The expression borders of eve stripe 2 are drawn by repression from the basic leucine zipper protein Giant and the zinc finger protein Krüppel expressed at each side of the eve stripe 2 domain (Small et al., 1991; Stanojevic et al., 1991). The binding sites for these four DNA-binding proteins are clustered and partially overlapping within the small regulatory element that mediates eve stripe 2 expression (Small et al., 1992). Disruption of Bicoid- or Hunchback-binding sites causes reduced stripe expression, while the disruption of Giant or Krüppel sites results in an expansion of the stripe expression domain (Small et al., 1991, 1992; Arnosti et al., 1996a). These studies provided a relatively simple model of the mechanism of how crude transcription factor gradients generate a single stripe expression domain in the anterior region of the embryo: binding of Hunchback and Bicoid causes activation, and the binding of repressing factors may exclude activator binding (Small et al., 1991, 1992; Stanojevic et al., 1991) or prevent activator function by quenching (Arnosti et al., 1996a). Bicoid and Hunchback expression overlap in the anterior half of the embryo only. Thus, the formation of pair-rule stripes in more central and in the posterior region of the embryo is likely to involve different activators. Earlier studies on the regulation of h stripe 5 and 6 expression have suggested that the posterior gap gene kni, in combination with as yet unidentified components, are necessary for their activation and that other gap gene products such as Krüppel, Giant and posteriorly expressed Hunchback provide the flanking repressor activities that establish the sharp limits of the expression domains (Howard and Struhl, 1990; Pankratz et al., 1990; Riddihough and Ish-Horowicz, 1991). As seen with the eve stripe 2 element, the binding sites for Knirps and Krüppel were found in overlapping clusters, suggesting that, despite the different players, the mode of activation and its spatial restriction by repression are similar to eve stripe 2 (Langeland et al., 1994). Here we show the trans- and cis-acting requirement for the formation and positioning of the most posterior h stripe 7. Our results indicate that h stripe 7 activation involves several factors including Caudal and the anterior morphogen Bicoid. The spatial limit of the expression domain is set by repression from gap gene products expressed in the flanking domains. Gene activation depends on the number of activator-binding sites present on the enhancer element. Deletions of such sites resulted in a h stripe 7 sub-element which is not sufficient for gene activation in the wild-type embryo, but which conducts gene expression in embryos deficient for repressors. Addition of in vitro binding sites for Caudal or Bicoid to this element restored the activation, indicating that the h stripe 7-element integrates positional information by decoding local combinations and concentrations of activators and repressors. The results also demonstrate that the anterior morphogen Bicoid, in addition to Caudal, is capable of activating gene expression in the most posterior region of the embryo, suggesting that the two opposing transcription factor gradients of maternal origin provide the basis for segmentation gene activation throughout the embryo. Results The hairy stripe 7-element Previous studies revealed a 1.5 kb DNA fragment of the h upstream region which conducts lacZ reporter gene expression in place of the endogenous h stripe 7 of the transgenic blastoderm embryo (Howard and Struhl, 1990; Pankratz et al., 1990; Riddihough and Ish-Horowicz, 1991). This stripe is the most posterior h stripe expressed. In order to monitor the trans-acting requirement for the activation and the spatial regulation of this stripe, we determined the minimal sequence requirement that is able to conduct reporter gene expression in the domain of h stripe 7 in wild-type embryos. Figure 1 summarizes various deletion constructs and the reporter gene expression patterns conducted in transgenic embryos. The results indicate that in wild-type embryos, h stripe 7 expression depends on a minimal 932 bp fragment which we refer to as the 'h7-element' (Figure 2A). In accordance with earlier observations of Riddihough and Ish-Horowicz (1991), the expression domain mediated by the h7-element is shifted by about one cell posteriorly relatively to endogenous h stripe 7 expression. Shorter DNA fragments containing different portions of the h7-element resulted in either weak and irregular reporter gene expression patterns or, as in the case of the 597 bp h7NF-element, failed to mediate reporter gene activation in the wild-type embryo (Figure 1). Figure 1.Summary of h–lacZ fusion gene constructs (left side) and corresponding blastoderm expression patterns (right side). The numbers 1–7 (top row) refer to cis-acting stripe elements of the hairy upstream region (Pankratz et al., 1990). Note the minimal h7-element (932 bp; 'h7', bold) which mediates h stripe 7 expression in wild-type embryos (see Figure 2). Restriction enzyme sites are A, AvaI; B, BalI; Ba, BanII; C, ClaI; F, AflII; H, HincII; K, KpnI; N, NlaIV; Na, NaeI; R, EcoRI. Download figure Download PowerPoint Figure 2.β-Gal reporter gene expression of the h7–lacZ fusion gene in maternal effect and gap mutant embryos. Gene abbreviations refer to embryos which are deficient for the respective gene activity. (A) h7–lacZ transgene-bearing wild-type embryo showing β-gal expression in the position of h stripe 7. (B) In the absence of bcd activity, the expression domain is duplicated. The posterior domain appears irregular and is shifted anteriorly. (C) In the absence of tor activity, the expression domain is shifted posteriorly and expands to the posterior pole. (D) In the absence of maternal cad activity, expression is slightly reduced. The anterior expression domain is due to the balancer chromosome, which carries a hb–lacZ marker gene construct to identify the genotype of the embryos (see Materials and methods). (E) In the absence of zygotic cad and bcd activity, the expression domain is duplicated and strongly reduced. (F) In the absence of nos activity, the h7-mediated stripe domain expands anteriorly. (G) In the absence of maternal and zygotic cad activity, expression is strongly reduced. (H) In the absence of cad (maternal and zygotic) and bcd activity, the expression domain is reduced and shifted anteriorly. (I) kni mutant embryo showing that the expression domain is expanded anteriorly. (J) Kr mutant embryo showing two expression domains. Note that the posterior expression is reduced. (K) In tll mutant embryos expression is absent. (L) In kni, tll double mutant embryos, the expression domain expands into the posterior pole region. Download figure Download PowerPoint In order to unravel the genetic control of h stripe 7 expression, we placed the h7-element-containing lacZ reporter gene ('h7–lacZ') into embryos which lack the activity of either maternal or zygotic components required for normal segmentation in the embryo. In wild-type embryos, h expression is preceded by the expression of gap genes, the first zygotically expressed members of the segmentation gene cascade which are activated in response to the maternal organizer system. This system includes the transcriptional activators Bicoid and Caudal, the transcription factor Hunchback and the unknown transcriptional regulator at the receiving end of the torso-dependent Ras/Raf signal transduction pathway (for reviews, see St Johnston and Nüsslein-Volhard, 1992; Rivera-Pomar and Jäckle, 1996). The hairy stripe 7-element mediates maternal gene activities Activation of h7–lacZ expression is dependent on the anterior and posterior, but not on the terminal maternal organizer systems. In embryos lacking terminal torso activity, h7–lacZ is activated at the posterior pole exclusively (Figure 2C). In embryos lacking bcd as the key component of the anterior organizer system, the h7–lacZ expression domain is duplicated. The normal posterior expression domain appears irregular and is shifted anteriorly (Figure 2B). In the absence of nanos activity, a key component of the posterior system, h7–lacZ expression expands anteriorly (Figure 2F). Furthermore, in the absence of cad, h7–lacZ expression is strongly affected (Figure 2G). cad is expressed both maternally and zygotically (Macdonald and Struhl, 1986; Mlodzik and Gehring, 1987). In the absence of zygotic cad activity, h7–lacZ expression appears normal (data not shown). In the absence of maternal cad activity, and in the absence of both maternal and zygotic cad activities, h7–lacZ expression is decreased (Figure 2D and G). This suggests that maternal rather than zygotic cad activity is required for the normal expression level of h7–lacZ. In embryos lacking bcd and zygotic cad activities, h7–lacZ expression is decreased (Figure 2E). This observation suggests that bcd participates in the activation of h7–lacZ expression. Embryos which lack bcd and both maternal and zygotic cad activities show weak h7–lacZ expression (Figure 2H). This indicates that a third component, Kr (see below), participates in the activation of h7–lacZ expression in the posterior region of the embryo. Note that such embryos lack the anterior expression domain seen in Bicoid-deficient embryos (Figure 2B), suggesting that their anterior expression domain depends on ectopic cad activity in the anterior region due to the lack of bicoid-dependent translational repression of caudal mRNA (reviewed in Rivera-Pomar and Jäckle, 1996). The hairy stripe 7-element mediates gap gene activities The spatial limit of h7–lacZ expression is significantly altered in the gap mutants tll, kni and Kr, but not in embryos lacking either hb, gt or the activity of the terminal gap gene huckebein (hkb). In kni mutant embryos, h7–lacZ expression expands anteriorly (Figure 2I). Thus, kni is required to repress h7–lacZ expression and thereby establishes the anterior border of the expression domain. In tll mutant embryos, h7–lacZ expression is absent (Figure 2K), implying that tll activity is necessary for the activation of the transgene. However, h7–lacZ expression is observed in embryos lacking torso activity (see above) which fail to express tll (Klingler et al., 1988). This suggests that tll does not act as an activator but provides activation indirectly by derepression. In fact, the repressing activity of kni expands posteriorly in embryos deficient for tll (Pankratz et al., 1989). In order to determine whether posteriorly expanded kni activity is the cause of repressed h7–lacZ expression in tll mutant embryos, we monitored h7–lacZ expression in embryos lacking both tll and kni activities. Figure 2L shows that h7–lacZ expression in kni, tll double mutant embryos expands into the posterior pole region. tll is therefore not only required to prevent kni-dependent repression of h7–lacZ activity but serves also to prevent terminal h7–lacZ expression in wild-type embryos. In summary, these results suggest that the anterior border of the h7–lacZ expression domain is established by repression in response to kni, while the posterior border of h7–lacZ expression is set in response to the repressor activity of tll. A surprising result was obtained with Kr mutants. Kr is expressed in the central region of the embryo, anteriorly adjacent to the kni expression domain. In the absence of Kr activity, two h7–lacZ expression domains were observed (Figure 2J). h7–lacZ expression in the normal stripe 7 position was reduced, while a second and stronger activity of h7–lacZ expression appeared in the Kr expression domain found in wild-type embryos, covering the area of h stripes 3 and 4. The expression of these stripes was shown to be dependent on cis-acting sequences 5′ to the h7-element, which give rise to only a single stripe in Kr mutant embryos (Hartmann et al., 1994). The finding of a corresponding stripe of h7–lacZ expression suggests that the h7-element hosts a second element which is silent in the presence of Kr activity. The results also suggest that Krüppel functions as a repressor in the context of the h stripe 3,4 sub-element, while it is likely to represent the third activator required for h7–lacZ expression in embryos which lack bcd and cad activities (see above). In vitro binding of trans-acting factors The genetic analysis of h7 expression revealed the maternal and gap gene requirement for the activation and proper spatial expression of the h7–lacZ gene. The six genes identified encode transcription factors. Caudal and Bicoid are homeodomain proteins, Krüppel and Hunchback are C2H2 zinc finger proteins, and Knirps and Tailless are orphan receptor-type zinc finger proteins (for review, see Jäckle and Sauer, 1993). We next asked whether the h7-element contains in vitro binding sites for these proteins. As summarized in Figure 3, DNase footprinting detected multiple sites for each factor which are scattered throughout the element. Many of the putative activator-binding sites (Caudal, Bicoid and Krüppel) are overlapping with putative repressor-binding sites for Knirps and Tailless, suggesting that part of the suppression could be provided through competition of activator and repressor binding, as observed in eve stripe 2 and the Kr control elements (Hoch et al., 1992; Small et al., 1992). Some of these putative repressors, however, bind to individual sites which are close to non-overlapping activator sites. This suggests that repression mediated by the h7-element also involves quenching of activator activity by close-by repressors. The finding that the trans-acting factors required for the control of h7–lacZ expression bind to the h7-element in vitro is consistent with the argument that their control is of a direct nature. Figure 3.Schematic representation of in vitro binding sites within the 932 bp h7-element. (A) Physical map of the h7-element, showing diagnostic restriction sites: A, AvaI; B, BalI; C, ClaI; F, AflII; N, NlaIV. Regions protected by Bicoid (BCD), Caudal (CAD), Krüppel (KR), Knirps (KNI), Hunchback (HB) and Tailless (TLL) are shown as boxes. (B) Sequence of the h7-element. For orientation, note the diagnostic restriction sites N, C, B and F (see A) and the five TLL-binding sites (boxed). Download figure Download PowerPoint Mechanism of hairy stripe 7-element control The 597 bp h7NF-element does not mediate reporter gene expression in wild-type embryos (Figure 4A). An anterior expression domain shown in Figure 4A is mediated by vector DNA. It appeared only when the transgenic embryos were overstained after in situ hybridization, and serves as an internal staining control (see also Rivera-Pomar et al., 1995). The lack of h7NF-dependent gene activation in transgenic wild-type embryos suggests that the h7NF-element contains either too few or lacks the essential activator-binding sites. Alternatively, the combination of activator and repressor sites present on the h7NF-element may be unbalanced such that bound repressors outnumber and thus overrule regulation by the activators. In order to distinguish between these possibilities, we monitored the h7NF-dependent reporter gene expression in kni, tll double mutant embryos. Figure 4B shows the expression of the h7NF-lacZ transgene in such embryos, indicating that the lack of the two repressors derepresses h7NF-mediated activation. Figure 4.h–lacZ fusion gene constructs and corresponding β-gal expression patterns. CAD-binding sites are indicated by open boxes and BCD-binding sites by filled boxes. For orientation of the binding sites, see Materials and methods. Gene abbreviations refer to embryos which are deficient for the respective gene activity. (A) The h7NF-element (see Figure 1) fails to mediate expression in wild-type embryos. The anterior expression is due to vector sequences and serves as an internal control demonstrating that even after a prolonged staining, no h7NF-mediated β-gal expression can be detected. (B) In kni, tll double mutant embryos, the h7NF-element mediates posterior expression, indicating that kni- and tll-dependent repression is lifted. The anterior expression domain (see A) is due to vector sequences. (C) Four CAD in vitro binding sites (see Materials and methods) fail to mediate expression. (D) Four CAD in vitro binding sites added to the h7NF-element rescue gene activation. (E) Four BCD in vitro binding sites (see Materials and methods) mediate expression in the anterior-most region of the embryo (see also Hoch et al., 1991). (G) Six BCD in vitro binding sites mediate gene activation in an anterior-to-posterior gradient (Rivera-Pomar et al., 1995). Four BCD (F) and six BCD (H) in vitro binding sites added to the h7NF-element cause gene activation in the most posterior region of the embryo. (I) The h7FA-element mediates a very weak and irregular h stripe 7-like expression. (J) The addition of six BCD in vitro binding sites to the h7FA-element restores a normal h stripe 7 domain. Download figure Download PowerPoint We next asked whether additional activator-binding sites on the h7NF-element override repression. For this, we added four Caudal-binding sites to the h7NF-element. While control constructs driven by the four Caudal sites alone were not expressed (Figure 4C), the addition of the Caudal-binding sites to the h7NF-element rescues gene activation (Figure 4D). Thus, the four added Caudal sites must act in concert with the activator-binding sites already present on the h7NF-element. These results establish that the lack of h7NF-dependent activation can be overcome by either reducing repressor activities or by increasing the number of activator-binding sites. This suggests a mechanism for gene activation that depends on a balanced ratio of enhancer-bound repressors and activators. Bicoid activates gene expression in the posterior region directly Genetic analysis of h7–lacZ reporter gene expression led to the surprising result of bcd-dependent gene activation in the most posterior h stripe 7 expression domain of the blastoderm embryo. In view of this result, we added four or six Bicoid-binding sites to the h7NF-element. In isolation, the four Bicoid-binding sites cause gene activation in the most anterior region of the embryo (Figure 4E), while the four Bicoid-binding sites in combination with the h7NF-element mediate gene activation in the most posterior region of the wild-type embryo (Figure 4F). Similarly, six Bicoid-binding sites mediate gene activation in an anterior-to-posterior gradient (Figure 4G), while their combination with the h7NF-element causes gene activation in the most posterior region of the embryo (Figure 4H). We note that the six Bicoid-binding sites contain a single Caudal-binding site. However, no h7NF-mediated gene expression was observed in embryos lacking bcd activity, irrespective of whether or not the four or six Bicoid-binding sites were added (data not shown). This indicates that gene activation mediated by the modified h7NF-element is dependent on Bicoid and that the single Caudal-binding site did not allow for cad-dependent gene activation. In the case of six Bicoid-binding sites, anterior and posterior expression domains are observed, but no expression is found in the central region of the embryo. This indicates that the repressors present on the h7NF-element function also in the context of the six added Bicoid sites. However, the activation of posterior gene expression by Bicoid is reduced as compared with the level of gene activation mediated by corresponding Caudal sites. To demonstrate further that Bicoid is able to regulate stripe expression in the posterior region of the embryo, we fused the six Bicoid-binding sites to a 280 bp fragment of the h7-element, termed the h7FA-element (see Figure 1). This element mediates only very weak and irregular h stripe 7-like expression (Figure 4I). Transgenic embryos containing the fusion gene construct show that the addition of the Bicoid-binding sites can restore a normal h stripe 7 domain (Figure 4J). These results establish that the anterior determinant Bicoid functions as a transcriptional activator of pair-rule gene expression in the posterior region of the embryo. Discussion We have presented evidence that the generation of the most posterior stripe expression domain of the pair-rule gene h involves at least three different activators and spatially restricted repressors which antagonize their activities. We demonstrate that the ability of non-functional cis-acting elements to mediate gene expression can be restored either by deleting repressors from the embryo or by adding activator-binding sites to the control region. The data suggest that the region-specific transcription in stripes depends on the combination and number of enhancer-bound activators and repressors which determine whether the cis-acting element mediates gene activation or fails to provide this activity. Regulation of h7 stripe expression In the absence of either cad or bcd activity, h7 expression is still activated. Even if both activities are deleted from the embryo, activation occurs. Thus, a third activator, likely to be Krüppel, must function in such embryos. The assignment of Krüppel as an activator in the context of the h7-element is consistent with the observation that h7–lacZ expression in Kr mutant embryos is considerably reduced, while a second expression domain, likely to be part of the h stripe 3,4-element, becomes de-repressed in Kr-deficient embryos. This would imply that Krüppel acts both as a repressor and as an activator within the h7-element, depending on its concentration. At high concentrations of Krüppel, in the central region of the embryo, it prevents activation of stripe 3,4 expression while at low concentrations, such as in the posterior region of the embryo, Krüppel can act as an activator. This conclusion is consistent with tissue culture studies showing that Krüppel is able to provide both activities at different concentrations (Sauer and Jäckle, 1991). Whether activation is provided by Krüppel monomers and repression by Krüppel dimers, as has been shown with both tissue culture and in vitro studies (Sauer and Jäckle, 1993), remains to be shown. The generation of the sharp h7 expression borders depends on at least two repressors, one on each side of the expression domain. The posterior border is set in response to tll activity under the control of the terminal maternal organizer system. The anterior border of the expression domain is due to repression in response to kni activity. In embryos lacking both bcd and cad activities, where kni activity is absent (Rivera-Pomar et al., 1995), an anterior border of the h7–lacZ expression is established in a position slightly more anterior than in wild-type. This indicates that a repressor other than Knirps is involved in setting this anterior border. This repressor could be maternal hb activity, consistent with the finding of multiple Hunchback-binding sites within the h7-element. Repression
Referência(s)