Local action of long-range repressors in the Drosophila embryo - PubMed (original) (raw)

Local action of long-range repressors in the Drosophila embryo

Y Nibu et al. EMBO J. 2001.

Abstract

Previous studies have identified two corepressors in the early Drosophila embryo: Groucho and dCTBP: Both proteins are recruited to the DNA template by interacting with short peptide motifs conserved in a variety of sequence-specific transcriptional repressors. Once bound to DNA, Groucho appears to mediate long-range repression, while dCtBP directs short-range repression. The short-range Krüppel repressor was converted into a long-range repressor by replacing the dCtBP interaction motif (PxDLSxH) with a Groucho motif (WRPW). The resulting chimeric repressor causes a different mutant phenotype from that of the native Krüppel protein when misexpressed in transgenic embryos. The different patterning activities can be explained on the basis of long-range silencing within the hairy 5' regulatory region. The analysis of a variety of synthetic transgenes provides evidence that Groucho-dependent long-range repressors do not always cause the dominant silencing of linked enhancers within a complex cis-regulatory region. We suggest a "hot chromatin" model, whereby repressors require activators to bind DNA.

PubMed Disclaimer

Figures

None

Fig. 1. Summary of expression vectors. Krüppel and snail coding sequences were placed under the control of heterologous twist and eve enhancers, respectively. The twist enhancer (‘twi’ in the diagrams) corresponds to two tandem copies of a modified 250 bp PE enhancer sequence that contains optimal Dorsal operator sites and Twist bHLH-binding sites. Two tandem copies of a modified eve stripe 2 enhancer (‘st2’ in D) were used to misexpress Snail. An FRT–stop–FRT cassette was inserted between the promoter and coding region to circumvent lethality resulting from the misexpression of these regulatory genes. (A) The wild-type Krüppel coding region was placed under the control of the modified twist enhancer. The encoded protein is composed of 502 amino acids and contains a dCtBP corepressor interaction motif, PEDLSMH, at position 464. (B) Same as (A) except that the Krüppel coding region was mutagenized to disrupt the dCtBP motif. The first three amino acids were converted into alanines. This protein is unable to bind dCtBP in vitro and has only limited repressor function in vivo (Nibu et al., 1998b). (C) Same as (B) except that a C-terminal portion of the Hairy repressor was attached in-frame to the 3′ end of the Krüppel coding region. The Hairy sequence contains the weak dCtBP-binding motif, PLSLVIK, and also contains the strong Groucho corepressor-binding motif, WRPW. (D) The snail–hairy fusion gene was placed under the control of the eve stripe 2 enhancer. Both dCtBP motifs in the snail coding sequence were left intact. The same portion of the Hairy protein as used in (C) was attached to the 3′ end of the snail coding region.

None

Fig. 2. Misexpression of Krüppel alters the hairy expression pattern. Embryos were collected from the indicated transgenic strains and hybridized with either a digoxigenin-labeled Krüppel antisense RNA probe (A–C) or a hairy probe (D–F). The different Krüppel coding sequences exhibit similar levels of expression in ventral regions of transgenic embryos. Note that these ectopic patterns are considerably weaker than the endogenous Krüppel pattern, which is seen as a broad band in central regions. Activated forms of the transgenes were obtained from transgenic males carrying the FLP recombinase under the control of the testis-specific tubulin promoter. (A and D) Krüppel and hairy expression patterns, respectively, in transgenic embryos carrying the mutant Krüppel coding region lacking the dCtBP repression domain (Figure 1B). The hairy expression pattern is essentially normal, although the spacing between stripes is sometimes irregular, and there may be a slight reduction in the ventral expression of stripe 7 (red arrowhead in D). The repression of hairy stripe 1 in ventral regions is also observed in wild-type embryos (not shown). (B and E) Krüppel and hairy expression patterns, respectively, in transgenic embryos carrying the wild-type Krüppel coding sequence. There is consistent ventral repression of stripes 2, 6 and 7 (red arrowheads). Stripe 5 sometimes exhibits an expanded pattern, possibly due to the repression of the endogenous Giant repressor (not shown). (C and F) Krüppel and hairy expression patterns, respectively, in transgenic embryos carrying the Krüppel–hairy fusion gene. There is a consistent repression of stripe 5, in addition to stripes 2, 6 and 7, as seen with the wild-type Krüppel transgene (red arrowheads in F; compare with E). Stripe 5 is not repressed by the wild-type Krüppel transgene.

None

Fig. 3. A Snail–Hairy fusion protein mediates local repression. Wild-type and transgenic embryos were hybridized with either a digoxigenin-labeled rhomboid or snail RNA probe. Embryos are oriented with anterior to the left and dorsal up. (A) snail expression pattern in a wild-type embryo. Staining is restricted to ventral regions that will invaginate to form the mesoderm. (B) snail expression pattern in a transgenic embryo carrying the snail–hairy fusion gene under the control of the eve stripe 2 enhancer (see Figure 1). Staining is detected in both the ventral mesoderm and an ectopic stripe (arrowhead). (C) rhomboid expression pattern in a wild-type embryo. In this lateral view, staining is detected in both ventral regions (the lateral neurogenic stripes) and in the dorsal ectoderm. (D) rhomboid expression pattern in a transgenic embryo carrying the st.2–snail–hairy fusion gene. The ectopic stripe of snail–hairy expression creates a gap in the neurogenic stripes (arrowhead). However, staining in the dorsal ectoderm is not affected (arrow).

None

Fig. 4. Local action of the Hairy repressor. Transgenic embryos express the indicated lacZ reporter genes (see diagrams below embryos). Embryos were stained with a digoxigenin-labeled lacZ antisense RNA probe and are oriented with anterior to the left. (A) Lateral view of an embryo expressing the NEE-2×PE (twist) transgene. A composite lacZ staining pattern is observed, including lateral stripes in the neurogenic ectoderm (arrowhead ‘NEE’) and expression in the ventral mesoderm (arrowhead ‘twist’). The binding of the short-range Snail repressor to the NEE (indicated by ‘S’ in the diagram) does not interfere with the expression of the twist enhancer. (B) Same as (A) except that the 5′ NEE was modified to include two synthetic Hairy repressor sites (‘h‘, see diagram). The Hairy repressor mediates periodic repression of the NEE pattern (arrowhead ‘NEE’). The binding of Hairy to the modified NEE also causes periodic repression of the mesoderm pattern (arrowhead ‘twist’). (C) lacZ staining pattern obtained with a transgene that contains the modified h-NEE-h enhancer placed 5′ of the race enhancer. As seen in (B), Hairy represses the modified NEE (arrowhead ‘NEE’), but fails to repress the race staining pattern in the dorsal ectoderm (arrowhead ‘race’). (D) lacZ staining pattern obtained with a transgene that contains tandem copies of a modified race enhancer placed 5′ of the twist enhancer sequences. The h-race-h enhancer exhibits periodic stripes of repression, presumably due to the binding of the Hairy repressor (arrowhead ‘race’). However, the binding of Hairy to h-race-h does not influence the staining pattern directed by the linked twist enhancer (arrowhead ‘twist’).

None

Fig. 5. Summary of the hot chromatin model. The diagrams in (B–D) present the hypothetical on/off states of several enhancers located in different positions along the dorsoventral axis of the early embryo. (A) Summary of the hairy 5′ _cis_-regulatory region. hairy stripes 2, 5, 6 and 7 are regulated by four different enhancers in the 5′-flanking region. The stripe 5 and stripe 6 enhancers are separated by >2 kb. Previous studies (Langeland et al., 1994) have shown that the stripe 6 enhancer contains a series of high-affinity Krüppel-binding sites. Consequently, low levels of the Krüppel repressor gradient (see diagram on left) are sufficient to bind these operator sites and repress stripe 6 expression. The stripe 5 enhancer is active in regions containing low and intermediate levels of Krüppel since the enhancer contains low-affinity sites that are occupied only by high con centrations of the Krüppel protein. The repression of hairy stripe 5by the Krüppel–Hairy fusion protein (Figure 2) suggests that the binding of this repressor to the stripe 6 enhancer can work over a long range to repress the stripe 5 enhancer. (B) Summary of the synthetic h-NEE-h–2×PE (twi) transgene. The Snail repressor normally binds to the NEE and represses its activity in the ventral mesoderm. Snail recruits the dCtBP corepressor and works locally, within the limits of the NEE, and does not interfere with the activity of the linked twi (2×PE) enhancer. In contrast, the Hairy repressor recruits Groucho, and the Hairy–Groucho complex not only represses the modified NEE, but can also work over a distance to repress the linked twi enhancer. (C) Summary of the h-NEE-h–2×race transgene. Hairy fails to repress the race enhancer in dorsal regions of transgenic embryos. We propose that the h-NEE-h enhancer might be inaccessible for the binding of Hairy in dorsal regions where the race enhancer is active. The NEE is activated by the maternal Dorsal gradient. There is no Dorsal activator present in the dorsal ectoderm and, consequently, the h-NEE-h and NEE enhancers might not be available for binding Hairy (or Snail–Hairy) in these regions where the race and rhomboid amnioserosa enhancers are active. (D) Summary of the h-race-h–twist transgene. The race enhancer is probably activated by Mad and Medea, and other transcription factors that are restricted to the dorsal ectoderm (reviewed by Podos and Ferguson, 1999). The h-race-h enhancer might be condensed in ventral regions due to the absence of these activators. As a result, Hairy is unable to bind the modified race enhancer in ventral regions where the twist enhancer is expressed.

Similar articles

Cited by

References

    1. Arnosti D.N., Barolo,S., Levine,M. and Small,S. (1996) The eve stripe 2 enhancer employs multiple modes of transcriptional synergy. Development, 122, 205–214. - PubMed
    1. Ashe H. and Levine,M. (1999) Local inhibition and long-range enhancement of Dpp signal transduction by Sog. Nature, 398, 427–431. - PubMed
    1. Barolo S. and Levine,M. (1997) hairy mediates dominant repression in the Drosophila embryo. EMBO J., 16, 2883–2891. - PMC - PubMed
    1. Bier E., Jan,L.Y. and Jan,Y.N. (1990) rhomboid, a gene required for dorsoventral axis establishment and peripheral nervous system development in Drosophila melanogaster. Genes Dev., 4, 190–203. - PubMed
    1. Fisher A.L., Ohsako,S. and Caudy,M. (1996) The WRPW motif of the hairy-related basic helix–loop–helix repressor proteins acts as a 4-amino-acid transcription repression and protein–protein interaction domain. Mol. Cell. Biol., 16, 2670–2677. - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources