Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Nov 1.
Published in final edited form as: Dev Biol. 2014 Aug 27;395(1):131–143. doi: 10.1016/j.ydbio.2014.08.021

Co-regulation of invected and engrailed by a complex array of regulatory sequences in Drosophila

Yuzhong Cheng 1, Alayne L Brunner 1, Stefanie Kremer 1, Sarah K DeVido 1, Catherine M Stefaniuk 1, Judith A Kassis 1,*
PMCID: PMC4189978  NIHMSID: NIHMS624571  PMID: 25172431

Abstract

invected (inv) and engrailed (en) form a gene complex that extends about 115kb. These two genes encode highly related homeodomain proteins that are co-regulated in a complex manner throughout development. Our dissection of inv/en regulatory DNA shows that most enhancers are spread throughout a 62kb region. We used two types of constructs to analyze the function of this DNA: P-element based reporter constructs with small pieces of DNA fused to the en promoter driving lacZ expression and large constructs with HA-tagged en and inv inserted in the genome with the phiC31 system. In addition, we generated deletions of inv and en DNA in situ and assayed their effects on inv/en expression. Our results support and extend our knowledge of inv/en regulation. First, inv and en share regulatory DNA, most of which is flanking the en transcription unit. In support of this, a 79-kb HA-en transgene can rescue inv en double mutants to viable, fertile adults. In contrast, an 84-kb HA-inv transgene lacks most of the enhancers for inv/en expression. Second, there are multiple enhancers for inv/en stripes in embryos; some of these may be redundant but others play discrete roles at different stages of embryonic development. Finally, no small reporter construct gave expression in the posterior compartment of imaginal discs, a hallmark of inv/en expression. Robust expression of HA-en in the posterior compartment of imaginal discs is evident from the 79-kb HA-en transgene, while a 45-kb HA-en transgene gives weaker, variable imaginal disc expression. We suggest that the activity of the imaginal disc enhancer(s) is dependent on the chromatin structure of the inv/en domain.

Keywords: enhancers, gene expression, gene regulation, segmentation

Introduction

The engrailed (en) gene encodes a homeodomain-containing protein that is a key developmental regulator in Drosophila. Mutations in en were first discovered in 1926 (Eker,1929). en1 flies have defects in wings and the scutellum, showing a role for en in the development of the adult cuticle. Further work on en1 led to the understanding that en is important for development of the posterior compartment of imaginal discs (Garcia-Bellido and Santamaria, 1972; Lawrence and Morata, 1975). When null mutations in en were made, its role in embryonic segmentation became evident (Kornberg,1981). en is classified as a segment-polarity gene, necessary for the development of every segment (Nüsslein-Volhard and Wieschaus, 1980). When en DNA was cloned, breakpoint mutations disrupting its function mapped over a 70kb region (Kuner et al., 1985). The developmental analysis of En protein distribution showed that in embryos En is expressed in stripes, portions of the head, discrete cells of the central and peripheral nervous systems, posterior spiracles, fat body, and hindgut (DiNardo et al., 1985). The invected (inv) gene encodes a closely related homeodomain protein (Coleman et al., 1987). inv is adjacent to en in the genome, shares regulatory DNA with en, and is functionally redundant with en (Gustavson et al., 1996).

En is a short-lived protein whose expression is continually regulated during development. Although En stripes are present in the posterior part of each segment throughout embryogenesis, they are regulated differently as development proceeds (DiNardo et al., 1988; Heemskerk et al., 1991). First activated by pair-rule genes, en is then regulated by cross talk with adjacent wingless (wg)-expressing cells, followed by a wg independent auto-regulatory period, and then late regulation. This model of en regulation was driven in part by the discovery of en DNA fragments that drove en-lacZ reporter genes in en-like stripes for only discrete time periods, or in only a subset of the stripes (DiNardo et al., 1988). Two DNA fragments that drive en expression in even-numbered stripes regulated by the pair-rule proteins have been identified (DiNardo et al., 1988; Kassis, 1990). In addition, an 8kb DNA fragment that drives en-like stripes at most stages of embryonic development has been known for a long time (Hama, Ali, and Kornberg 1990); however, the cis-regulatory modules for most of the expression patterns of inv and en have not been located.

We are interested in how Polycomb group (PcG) proteins regulate inv/en expression (DeVido et al., 2008). With a goal of making a simple reporter construct to study how PcG proteins regulate en expression in imaginal discs, we dissected the regulatory DNA of inv/en. During the course of these studies we generated 30 P-element based reporter constructs, two large HA-en transgenes, one large HA-inv transgene and a number of new inv and en deletion alleles. As suggested by the early studies (Kuner et al., 1985), inv/en regulatory DNA extends over a large region, about 62kb. Many discrete embryonic enhancers were found, but no single enhancer was located for expression in the posterior compartment of imaginal discs. Here we describe the incredible array of regulatory elements devoted to regulating the expression of inv/en.

Materials and Methods

Construction of P-based reporter constructs

P[en] has a unique SphI site located 396bp upstream of the en transcription start site. Constructs I–K, R, T, V, X, and Z were made by cutting SphI fragments out of lambda DNA en clones (Kuner et al., 1985), and subcloning into SphI cut P[en]. Constructs A, B, D, F, and Q were also isolated from lambda DNA en clones, cut out by EcoRI and cloned into an EcoRI site in P[en] located adjacent to the SphI site (an EcoRI site located downstream of the lacZ gene was filled in and destroyed prior to these clonings); fragment N was cloned as an EcoRI-SphI fragment into P[en]. Other constructs were made by amplifying genomic DNA by PCR and cloning into P[en] cut with either SphI, EcoRI, or both or into P[en] that had been modified to contain the restriction sites SpeI and NotI at the SphI site. Construct H has FRT and loxP sites flanking the PSE/PREs (construct P[en2] in DeVido et al., 2008). All DNA fragments were cloned into P[en] in the 5’ to 3’ with respect to the en transcription unit except fragment R. Cloning procedures for constructs with multiple fragments (ie HIJK) are available on request. Injections were done by Genetic Services, Inc.

Construction of phi-C31 reporter constructs

The HA-en45, HA-en79 and HA-inv84 constructs were made using a gap-repair protocol described in http://www.pacmanfly.org/protocols.html, and a recombineering protocol described by Warming et al. (2005) with modifications whenever needed. To facilitate selection of transformants, a 289bp eye and testis enhancer of the white gene (Qian et al., 1992; X:2691681..2691970) was added upstream of the promoter of the mini-white gene in the attB-p[acman]-ApR vector (Venken et al., 2009) by recombineering. Detailed information about these procedures is available upon request. HA-inv84 was inserted into the landing site VK33; HA-en45 and HA-en79 were inserted into both the attP3 and attP40 landing sites (Genetic Services and Bestgene, Inc.) Similar results were obtained in both landing sites.

Generation of en and inv deletions in situ

inv deletions were generated using FRT-mediated recombination from four PBac insertions in inv (obtained from the Harvard Exelixis stock center). Deletions were generated using standard procedures (Parks et al., 2004) and verified by PCR. enΔ17 was generated by imprecise excision of a P-element inserted at −412bp (Devido et al., 2008). No P-element sequences remain in this mutant. enΔ5.9 and enΔ6.4 were generated by imprecise excision of P(en3aboth)-en inserted 6kb upstream of the en transcription start site (DeVido et al., 2008; Kwon et al., 2009). In both of these mutants the mini-white gene (transcribed in the opposite orientation to en) remains at the deletion endpoint.

Antibody stainings

An antibody against Inv amino acids 1–154 (Enzymax) was made in guinea pigs (Covance, used at 1:500 dilution). Other primary antibodies used were: rabbit anti-β-galactosidase (1:15,000, Cappel); mouse anti-β-galactosidase (1:500, Invitrogen); mouse anti-HA (1:100, Covance); rabbit anti-En (1:500, Santa Cruz Biotechnology, Inc.) and rabbit anti-En (1:200; DiNardo et al., 1988). For the embryos in Fig. 1B, rabbit anti-β-galactosidase was used followed by immunoperoxidase staining with the Elite ABC kit (Vector Labs). For fluorescent stainings, Alexa Fluor secondary antibodies were used (Invitrogen) and embryos or discs were mounted in Vectashield with DAPI (Vector Labs).

Fig. 1. Enhancers of en and inv.

Fig. 1

Open in a new tab

(A) P-element vector (P[en]) used to assay the function of en regulatory DNA. It contains the en promoter, 396bp of upstream sequences, and an untranslated leader fusion between the en transcript and the Adh-lacZ reporter gene. inv/en DNA fragments were added to this vector at the location of the triangle. (B) The extent of each fragment cloned into P[en] is shown as a black line with a letter (exact coordinates are listed in Table 1) above the inv/en genomic DNA map (indicated by a long black line with hatch marks at 10kb intervals; numbers are coordinates on chromosome 2R, genome release v5). Expression pattern in embryos or the wing imaginal disc (wd) are shown above or below the genomic DNA with arrows pointing to the fragment(s) that generate the pattern. All embryos are anterior left. Embryos are lateral view except the following: D,W-dorsal; H, V-ventrolateral; J, K-ventral; P (upper)-dorsolateral. Embryo stages shown are as follows: Stage 13-C, E, P (lower). Stage 14-E, J, V. Stage 10-G, M (lower). Stage 11-H, I, M (upper), O, P (upper), Q, S, T. Stage 12- R. Stage 15-W. Stages are from Campos-Ortega and Hartenstein (1997). Fragments C, D, and I gave expression in a line at the anterior-posterior boundary in imaginal discs. Fragment O gave nearly ubiquitous expression in all discs. Fragment S gave expression in the anterior compartment of wing imaginal discs. Six fragments gave no reproducible expression in either embryos or imaginal discs (fragment A, B, F, L, Y, and Z). Fragment U gave expression in the anterior compartment in embryos (not shown).

Genetic crosses

To examine expression of constructs in wg and inv en double mutants, constructs inserted on chromosome 3 were crossed to wg1–8/CyO or Df(2R)enX31/CyO (hereafter referred to as enX31). Construct/+; enX31/+ or wg1–8/+ virgins and males were crossed to each other and embryos were collected, fixed, and stained for β-galactosidase and En antigens. Abnormal or absent En staining identified the mutants.

enB86 (an En protein null, Gustavson et al., 1996), enE, and enX31 were recombined onto a chromosome with the HA-en45 or HA-en79 inserted at the attP40 site. The presence of the en mutation was verified by PCR, and the transgene was detected by the presence of the w+ marker. HA-en45@attP40 enB86 and HA-en79@attP40 enB86 are homozygous viable. The enE chromosome used for recombination contained the dominant markers Sp Bl and Pin and the resulting recombinants are HA-en45 or HA-en79@attP40 enE Pin and are not homozygous viable. To assess rescue of inv en double mutants these flies were crossed to HA-en45- or HA-en79@attP40 enX31 L flies.

The identity and exact location of the en1 insertion element was determined by long PCR with primers that flank the en1 insertion site and sequencing the ends of the product. en1 is caused by the insertion of a blood retrotransposon at nucleotide 2R:7428462, approximately 13kb upstream of the en transcription start site (at nucleotide 2R:7415388).

Results

We tested the enhancer activities of DNA fragments using a P-element based vector that included the mini-white reporter gene, the en promoter, transcription start site, and untranslated leader fused to the Adh-lacZ reporter gene (Fig. 1A). We used the en promoter because some en enhancers exhibit promoter specificity (Kassis, 1990; Kwon et al., 2009). Because flies with deletions of inv DNA are viable (Gustavson et al., 1996 and see below), we reasoned that most regulatory DNA must be contained in the 70kb region extending from the 3’ end of the inv gene to the 3’ end of tou (the gene just upstream of en, see Fig. 1B). We made 30 constructs in P[en] (Table 1 and Fig. 1B). Multiple lines were generated for each construct, and 2–4 lines of each construct were stained with antibodies against β-galactosidase in embryos and for β-galactosidase activity in imaginal discs. The expression patterns of these constructs in embryos and wing imaginal discs are shown in Fig. 1B and listed in Table 1. We found evidence for at least 20 embryonic enhancers including those that drive expression in ectodermal stripes (discussed below), central and peripheral nervous systems, hindgut, head, and posterior spiracles (Table 1). We describe the activity of the stripe enhancers in more detail below.

Table 1.

Coordinates of DNA fragments in transgenes

Fragment Lab Name aCoordinates Size bExpression
A e614 7391250–7394957 3.7kb NE
B e606 7394958–7400993 6.0kb NE
C SLK1 7400904–7402968 2.1kb CL, DR
D e702 7402883–7406102 3.2kb PS
E SLK2 7405109–7409453 4.3kb HG, CL
F e710 7409191–7411923 2.7kb NE
G SD7 7411822–7414989 3.2kb ES, 15S, FB, H
H SK11A 7415785–7423711 7.9kb ES, OS, 15S, H, PS
I e8-2 7423711–7427003 3.3kb ES
J e8–29 7427004–7429280 2.3kb ML
K e8-5 7429281–7432610 3.3kb ML
L SLK3 7430970–7434527 3.6kb NE
M SLK5 7433766–7437251 3.5kb OS, 15S, H
N e10-8A 7432853–7440311 7.5kb OS, 15S, H
O SLK4 7435274–7439183 3.9kb 15S, H
P e10-8b 7439183–7446032 6.8kb LS, CNS (NB), PNS
Q y811 7441285–7448926 7.6kb CNS (NB)
R 7P 7446390–7453923 7.5kb CNS, H
S SD6 7448808–7455495 6.7kb 15S, H
T 7Z 7453924–7455476 1.5kb 15S, H
U enSN1 7454592–7457006 2.4kb cAS
V 4E 7455477–7459927 4.5kb CNS, PNS
W enSN2 7458471–7461442 3.0kb PS
X 4D 7459928–7463168 3.2kb PS
Y enNot 7462027–7465200 3.2kb NE
Z e12–20 7463192–7467362 4.2kb NE
IJ SD2 7423711–7429280 5.6kb ML
IJK SD3 7423711–7432610 8.9kb ES, ML
HIJ SD13 7415785–7429280 13.5kb 15S, H, PS
HIJK SD12 7415785–7432610 16.8kb S, H, PS
HA-en79 same 7386838–7466000 79.2kb See text
HA-en45 same 7404008–7448931 45.0kb See text
HA-inv84 same 7326251–7409745 83.5kb See text
Open in a new tab
a

Genome release v5.

b

In embryos. NE-no expression; CL-clypeolabrum; DR-dorsal ridge; PS-posterior spiracles; HG-hindgut; ES-even numbered stripes; 15S-15 stripes; FB-fat body; H-head expression; ML-midline cells; OS-odd numbered stripes; CNS-central nervous system; NB-neuroblasts; PNS-peripheral nervous system; LS-late abdominal stripes. For details see text.

c

AS-stripes in anterior compartment. β-gal stripes in all abdominal segments were formed, but they did not overlap with Inv/en stripes.

We were surprised that no construct mimicked the expression of inv/en in the posterior compartments of imaginal discs, although 3 constructs gave strong expression in a line at the anterior/posterior boundary in discs (Constructs C, D, and I). Unexpectedly, construct S had expression in the anterior compartment in discs, opposite to where En is expressed (Fig. 1B). Further, construct 0 gave nearly ubiquitous expression in discs (Fig. 1B). We reasoned that an imaginal disc enhancer might have been divided in the constructs. Therefore, we made a construct that contained 16.8kb of upstream DNA (fragments H, I, J, and K combined, Table 1) and additional constructs to ensure there was at least a 500bp overlap between each individual fragment upstream of en. Still we did not find a single fragment of DNA that recapitulated inv/en expression in the posterior compartment of imaginal discs.

Stripe enhancers of the inv/en genes. En is a short-lived protein whose expression is continually regulated during embryonic development (DiNardo et al., 1988); β-galactosidase is a long-lived protein in Drosophila that persists throughout embryonic development. In order to determine the temporal aspects of the stripe enhancers, we assayed the expression of lacZ RNA (which has a short half-life, Kassis, 1990) (Fig. 2). Construct H, that contains 8.3kb of sequences immediately upstream of the en transcription start site (Fragment H and 400bp present in the vector) gives stripes at most stages of development (Fig. 2). Like en, the expression is first turned on at the beginning of gastrulation with the even numbered stripes turned on more strongly at first. Also like en, lacZ RNA stripe intensities in Construct H embryos soon even out; then, unexpectedly, at about 7h of development (stage 11) and unlike en, the ventral and lateral lacZ stripes transiently fade. This stripe fading was also seen with a different reporter construct that contained a partially overlapping DNA fragment (DiNardo et al., 1988). The full stripes are restored by about 7.5h of development, and like en stripes, lacZ stripes from construct H are maintained throughout the rest of embryogenesis. This same pattern of lacZ expression is seen from a construct with 16.8kb of upstream sequences (including fragments H, I, J, and K, Table 1), with an even more pronounced loss of ventral staining at 7h (stage 11) of development (data not shown). lacZ stripes expressed from construct G (Fig. 2), that contains the DNA from the en transcription unit, are also activated early in development, but fade completely by about 7h of development (fat body expression persists). Constructs M and T have interesting expression patterns (Fig. 2). Both are expressed in all stripes at about 5h of development, and then begin to fade at 7h of development, the same time that construct H stripes begin to fade. However, construct M stripes fade dorsally first, then laterally, retaining ventral expression slightly longer (until about 7.5h of development, late stage 11). Construct T stripes also fade during this time period, but in this case they fade laterally first and then disappear completely at about 7.5h of development. We conclude that during late stage 11, En stripes are regulated by three different enhancers, with each responsive to different dorsal-ventral signals. After stage 11, lacZ stripes continued to be expressed throughout embryogenesis from construct H. Construct P also drives striped expression late in embryonic development (Fig. 1B). In this case β-galactosidase is expressed in neuroblasts of the CNS and PNS beginning at about stage 9, and then is expressed in ectodermal stripes only in the abdominal segments, beginning at about stage 12 (Fig. 1B).

Fig. 2. lacZ RNA expression shows the dynamic nature of inv/en stripes.

Fig. 2

Open in a new tab

lacZ RNA expression patterns from construct G, construct H, construct M, and construct T are shown. All embryos are oriented anterior left. The number in the right corner of each picture indicates the stage (from Campos-Ortega and Hartenstein,1997) and the letters indicate the view (L-lateral; V-ventral, VL-ventrolateral; DL-dorsolateral). lacZ RNA is gone from construct M and T soon after the last stage shown. Construct G-black arrow points to fat body expression. Construct H-black arrow points to a stripe that is fading ventrolaterally; white arrow points to salivary gland staining (an artifact of staining with this probe). Construct M-black arrow points to the dorsal region where stripes have faded. Construct T-black arrow points to the lateral region of the embryo where stripes have faded.

Previous studies demonstrated that En stripes are first turned on by the pair-rule proteins, then activated by wg signaling, then by wg-independent auto-regulation, followed by a period of late expression regulated by unknown activators and the Polycomb group of repressors (DiNardo et al., 1988; Heemskerk et al., 1991; Moazed and O’Farrell, 1992). Our data are consistent with this model with multiple enhancers present for many of the steps in this regulatory cascade (Fig. 3A). We discuss this in more detail below.

Fig. 3. Stripe enhancers for inv/en.

Fig. 3

Fig. 3

Open in a new tab

(A) A model of inv/en stripe regulation during embryogenesis (adapted from Heemskerk et al., 1991). A time line is shown on top (hours after egg laying at 25°). Regulatory inputs for inv/en expression are shown on the left. The boxes show the time of the regulatory input with shading indicating uncertainty in when that input begins. Only the smallest constructs with the listed regulatory inputs are shown. (B) Embryos were double-labeled with anti-En and anti-β-galactosidase using immunofluorescence. Abnormal or absent En protein allowed identification of the mutant embryos; only the anti-β-galactosidase staining is shown. The construct is written on the top; the genotype is written on the side (wt-wildtype). All embryos are anterior left. (H) stage 10 embryos; (M) stage 11 embryos, dorsal (top), dorsal-lateral (middle) and lateral (bottom) views; (T) stage 11 embryos; lateral (top) and dorsal (middle) and ventrolateral (bottom) views; arrow points to β-galactosidase antigen in the posterior region expressed in T (due to a position effect). This expression allowed the identification of the embryo with the transgene. (C) The en genomic region (as in Fig. 1) with ChIP-chip data of the pair-rule proteins Ftz, H, Prd, Run, and Slp1 (MacArthur et al., 2009) and of the ChIP-chip peaks of dTcf binding in embryos (red lines; Junion et al., 2012). Fragments of DNA giving striped expression are also shown. The en transcription unit is shown with exons as think black lines and introns as light black lines. The direction of en transcription is right to left. Dotted lines indicate the extent of deletions in the en mutants shown; the 3’ end of the enE deletion is beyond the genetic coordinates of this figure.

en is expressed in 15 stripes in embryos. Different sets of pair-rule proteins control the expression of the even- and odd-numbered en stripes; thus there are different enhancers for even and odd stripes (DiNardo and O’Farrell, 1987; DiNardo et al., 1988). We found there are 3 enhancers for both the even- and odd-numbered stripes (summarized in Fig. 3A). One fragment of DNA, fragment I, gives expression only in the even numbered stripes (Fig. 1B); these stripes are transiently expressed (data not shown), consistent with an enhancer regulated only by pair-rule proteins. Constructs G and H contain enhancers for both the even and odd stripes (Fig. 2). lacZ RNA is detected in both the even stripes (strongly) and in the odd stripes (weakly) at the beginning of gastrulation in construct G and H embryos (Fig. 2). Further, when construct G or H are present in either a wg or inv en double mutant background (enX31), all stripes are formed, consistent with activation of both the even- and odd-numbered stripes by the pair-rule proteins (data not shown and Fig. 3B).

An additional enhancer for early expression in odd-numbered stripes is present in construct M (and the overlapping construct N). Odd numbered stripes are detected at the beginning of gastrulation in construct M embryos (Fig. 2), with stripes in every segment prominent by stage 10. This later expression is due to the presence of a second enhancer (that is also present in construct O). This second enhancer drives expression in every segment and seems to respond either directly or indirectly to the presence of both Wg and Inv/En. To illustrate this, Fig. 3B shows construct M in wg and inv en double mutant backgrounds. Odd stripes, which are activated by the pair-rule proteins, are evident because of the long-lived β-galactosidase protein. In contrast, the even-numbered stripes are only very weakly present, suggesting they need to be activated either directly or indirectly by wg signaling and inv/en. When construct O was put into either a wg or an inv/en double mutant background, only very light stripes were seen (data not shown), suggesting that the second enhancer has been separated from the pair-rule enhancer in construct O. Finally, both wg and inv/en are required for the formation of β-galactosidase stripes from construct T as ectodermal β-galactosidase stripes are completely missing in construct T; wg embryos and nearly missing in construct T; enX31 embryos (Fig. 3B).

The location of pair-rule protein binding sites correlates remarkably well with the locations of the enhancers that respond to the pair-rule proteins (Fig. 3C data from MacArthur et al., 2009). Of the pair-rule proteins shown in Fig. 3C, Ftz (Fushi tarazu) is known to activate even-numbered En stripes, and Prd (Paired) activates odd-numbered En stripes (DiNardo and O’Farrell, 1987). H (Hairy), Run (Runt), and Slp1 (Sloppy-paired 1) are all repressors of En expression (Mannoukian and Krause,1993; Cadigan et al., 1994; Jaynes and Fujioka, 2004). There are Ftz and Prd binding sites within fragment G, consistent with it driving expression in every segment at the beginning of gastrulation. It was already known that the en intron gave expression in even- but not odd-numbered stripes early in development (Kassis, 1990). This is consistent with the location of the Ftz binding sites within the first en intron, while the Prd sites are located both within the first exon and the second exon and intron. Fragment H contains clustered binding sites for all the pair-rule proteins shown, consistent with its early expression in all stripes. Fragment I contains strong Ftz binding sites, but only weak Prd binding sites, and is only expressed in the even-numbered segments. Finally, fragment M contains strong Slp1, Run, and Prd binding sites and is only activated in the odd numbered stripes early. These pair-rule protein binding sites are not present in fragment 0, consistent with the lack of early expression from construct 0.

Similarly, the location of dTcf binding sites correlates well with the enhancers that respond to wg signaling (Fig. 3C; data from Junion et al., 2012). dTcf is the DNA binding protein activated by the wg signaling pathway. Fragments G, H, M, 0, and T all have dTcf binding sites, and there is only one dTcf binding site that does not fall within these fragments. Finally, we would like to point out that the repressors H and Slp1 are bound at all the stripe enhancers that function after the pair-rule stage suggesting that these proteins may regulate inv/en stripes throughout embryogenesis.

Insight into inv/en regulation through deletions in the endogenous inv/en locus.

Expression of inv and/or en in mutants with deletions of en regulatory DNA was informative and confirm and extend the transgene analysis. enΔ5.9 removes sequences from −400bp to −6.3kb upstream of the en transcription start site (Fig. 3C; Table 2). We could not detect any alterations in en expression in enΔ5.9 homozygous embryos. In particular, although pair-rule protein binding sites located within fragment H are deleted in this mutant, the En stripes look normal (Fig. 5A). This suggests these sites behave redundantly within the inv/en locus. Consistent with this, enΔ5.9 homozygotes and enΔ5.9/enX31 die after hatching as first instar larvae with normal looking cuticles (data not shown). Since construct H contains the only enhancer that gives expression in the late thoracic stripes and 5.9kb of it have been deleted in the enΔ5.9 mutant, we propose that the late enhancer is located between 6.3 and 8kb upstream of the en transcription start site. We do not know why enΔ5.9 larvae die but suspect there are either subtle changes in the embryonic expression pattern, or that larval En expression is disrupted. The enΔ6.4 deletion includes 6.3kb of upstream sequences and extends 100bp into the en transcription unit. As expected, en expression is not detected in these embryos, but inv is transiently expressed in 15 stripes (data not shown). inv is encoded in a large transcription unit and cannot substitute for En in early development, most likely because the Inv protein cannot be expressed early enough (Gustavson et al., 1996). Without early expression of En, Wg stripes fade, leading to the loss of Inv/En expression (DiNardo et al., 1988; Martinez-Arias et al., 1988).

Table 2.

Coordinates of inv/en deletions

Mutant Deletion Coordinates Size
enX31 7332587–7536107 203.5kb
enE 7383679–7425016 41.3kb
enΔ17 7412589–7429642 17.0kb
enΔ6.4 7415267–7421628 6.4kb
enΔ5.9 7415689–7421628 5.9kb
invΔ24 7357207–7381178 24.0kb
invΔ32 7357207–7389011 31.8kb
invΔ45 7357207–7402154 44.9kb
Open in a new tab

Fig. 5. Map showing the extent of the inv/en deletions and the large constructs.

Fig. 5

Open in a new tab

Genomic DNA is denoted by the line with hatch marks every 20kb with the coordinates on chromosome 2R shown. Arrows show the extent of transcription units with the names listed below. Dash lines indicate the DNA deleted in the mutants, and the names of the DNA fragments that are deleted are indicated (from Fig. 1B). enCX1 results from a breakpoint in the en first intron (Gustavson et al., 1996), and thus removes the DNA shown by the dashed line. The location of the insertion of the blood retrotransposon in en1 is indicated. The extent of each large transgene is shown (black lines); the location of the N-terminal HA tag is indicated by a triangle. The exact extents of the deletions and transgenes are listed in Table 2.

Expression of Inv in enΔ17 embryos is interesting. enΔ17 deletes 14.25kb of upstream en sequences and 2.8kb of the en transcription unit, including the first and second exons and the first intron. This deletion removes all enhancers for the even-numbered stripes, and ends in the middle of a Prd binding region within the second intron of en (Fig. 3). Consistent with this, Inv is expressed in the odd-numbered stripes only, easily detected during germ band elongation (Fig. 4B). This presence of stripes in only the odd-numbered segments strongly suggests that the pair-rule proteins are directly activating expression from the inv promoter, via the odd stripe enhancers in fragments G and M, located approximately 60 and 80kb away from the Inv promoter. Later in development, expression of Inv in the head and nervous system is evident, and expression in the odd-numbered stripes weakly persists in a patchy manner (Fig. 4C). The weak persistence of the late ectodermal stripes is surprising for two reasons. First, Inv expression cannot rescue the loss of Inv/En stripes seen in an En protein null mutant (see above). We suggest that the deletion in enΔ17 leads to earlier activation of the inv promoter and partial rescue of the late Inv stripes. Second, the only enhancer we found for late thoracic stripes (in construct H) has been entirely deleted in enΔ17. We don’t have a good explanation for this. It’s possible that some of the other enhancers actually work together in the context of the endogenous locus, leading to late expression not seen in the isolated fragments of DNA.

Fig. 4. En and Inv expression in en mutant embryos.

Fig. 4

Fig. 4

Open in a new tab

All embryos are anterior left. (A) In enΔ5.9 embryos, En expression looks normal. A stage 12 embryo (posterior dorsal view) showing normal stripes, PNS, hindgut and posterior spiracles. Embryos were collected from enΔ5.9/CyO wg-lacZ flies; embryos with no β-galactosidase antigen are homozygous enΔ5.9 (B) Wildtype (wt), encr17 and enE embryos stained with anti-En (rabbit); anti-Inv (guinea pig); and DAPI, as indicated. All embryos are approximately stage 11, anterior left. wt and enΔ17 embryos are lateral views. There is no En staining in enΔ17 and Inv is only present in the odd-numbered stripes. enE: A truncated Inv protein (Inv*) is expressed from enE, and is located mainly in the cytoplasm. The left panel is a ventral view, the middle panel is a dorsal view of the same embryo, showing all 15 Inv* stripes are made. A lateral view of the head region is shown in the right panel where a white arrow points to an adjacent wildtype embryo that has the En antigen. (C) Embryos of indicated genotypes, stage 14, lateral view. Inv (enΔ17) or Inv* (enE)is expressed in the head and nervous system in both mutant genotypes. In enΔ17, some cells in the odd-numbered stripes continue to express Inv in the ectoderm.

The enE deletion extends from 9.6kb upstream of the en transcription start site to 31.7kb downstream, deleting about half of the inv transcription unit including the homeodomain (Fig. 3C and 4). An anti-Inv antibody was made against the N-terminal region of Inv; it detects a truncated Inv protein in enE mutants. Inv is normally a nuclear protein, but the truncated Inv protein (hereafter called Inv*) is largely cytoplasmic. Inv* is expressed in fifteen stripes in germband elongated enE embryos (Fig. 5B); later in development these stripes disappear and Inv* is detected in the nervous system and head of late enE homozygous embryos (Fig. 5C), similar to what is seen for En expression in an en null mutant (Heemskerk et al., 1991). enE deletes fragment H and part of fragment I, the deletion ending in the middle of the pair-rule protein binding sites, but apparently leaving a functional enhancer for even-numbered stripes, otherwise only the odd-numbered stripes would be present. Later in development Inv* stripes fail to be maintained because of the lack of functional Inv or En.

Three deletions were made in the region of the inv gene (24kb, 32kb, and 45kb, Fig. 5, Table 2). invΔ24 and invΔ32 are homozygous viable, fertile, and look phenotypically normal. This is consistent with a previous report that the inv gene is dispensable for viability (Gustavson et al., 1996). In contrast, most invΔ45 and invΔ45/enE (enE is null for both en and inv, see Fig. 5) animals die as pharate adults; rare escapers have downturned wings and extra veins between veins 4 and 5, suggesting a subtle defect in en expression during wing development. The invΔ45 deletion extends from 6kb upstream of the inv transcription unit to 7kb downstream of it and removes fragments A, B, and most of C (814bp of it remain). Fragment C contains an enhancer for expression in the clypeolabrum but embryonic En expression in invΔ45 is normal suggesting that either this enhancer has not been removed or there is enhancer redundancy (with fragment E) for this tissue. Expression of En in imaginal discs of invΔ45 also looks normal (see below).

Enhancers for expression in the posterior compartment of imaginal discs are located upstream of en. Multiple DNA fragments activated lacZ expression in a stripe at the anterior/posterior boundary in imaginal discs (Fig. 1B), but no fragment caused expression in the posterior compartment of imaginal discs. We therefore turned to inv/en mutants to guide our search for this enhancer(s). Genetic evidence strongly suggests that the imaginal disc enhancers for expression in the posterior compartment are located upstream of the en transcription unit. enCX1 has a breakpoint within the first exon of En (Fig. 4) and produces a truncated En protein that is partially cytoplasmic and thus can be distinguished from the wildtype nuclear En protein (Gustavson et al., 1996). Double labeling enCX1/+ imaginal discs with an antibody that detects the N-terminal region of En and an antibody that detects both the En and Inv homeodomains shows that the truncated enCX1 protein is expressed in the posterior compartment of imaginal discs (Fig. 6A). This suggests that the imaginal disc enhancers are located upstream of the breakpoint in enCX1. Consistent with this, the invΔ45 deletion, which deletes from 6kb upstream of inv through 7kb downstream, does not disrupt the imaginal disc expression of en (Fig. 6B). Additional data suggests that the enhancers for expression in the posterior compartment of the imaginal discs are located greater than 14kb upstream of en. Inv is expressed in the posterior compartment when expressed from an enΔ17 chromosome, (rescued by a HA-en79 transgene, Fig. 6C), indicating that the imaginal disc enhancer(s) are maintained on the enΔ17 chromosome.

Fig. 6. Imaginal disc enhancer(s) are located more than 15kb upstream of the en transcription unit.

Fig. 6

Open in a new tab

(A) Wing pouch from an enCX1/+ 3rd instar imaginal disc stained for En (N-terminal region of En) and the homeodomain (HD), which recognizes both the Inv and En homeodomains (C-terminus). enCX1 produces a truncated En protein that does not contain the homeodomain and is localized to both the nucleus and the cytoplasm. A wild-type copy of En is also expressed from the wildtype chromosome. (B) Wing disc from an invΔ45 3rd instar larvae. En is expressed in the posterior compartment; Inv is not present. (C) Wing pouch from an HA-en79 enX31/enΔ17 3rd instar imaginal disc. Inv is expressed from the enΔ17 chromosome in the posterior compartment. This shows that the wing imaginal disc enhancer is still present on the enΔ17 chromosome.

The original en allele, en1 disrupts expression of En in the posterior compartment of wing discs (Brower, 1986). We determined that en1 is caused by an insertion of a blood retrotransposon 13kb upstream of the en promoter (see Materials and Methods). Because the en imaginal disc enhancers are located upstream of this, we hypothesize that this insertion disrupts communication between the wing imaginal disc enhancers and the inv and en promoters leading to the en1 loss-of-function phenotype.

Expression of Inv and En from large transgenes. Three large transgenes support and extend the regulatory DNA dissection and mutant analysis. The Inv protein was N-terminally HA-tagged in an 84kb construct that was integrated into the genome in the attP site at 65B2 (VK33)(Fig. 5; Table 1). HA-Inv was not expressed in stripes in embryos (Fig. 7A). HA-Inv embryonic expression was limited to the hindgut, clypeolabrum, and some cells in the nervous system (Fig. 7A, B). Curiously, although fragment D, which gives expression in the posterior spiracles, is contained within the HA-Inv transgene, HA-Inv was not detected in the posterior spiracles (Fig. 7A). In imaginal discs, HA-Inv was detected in a line at the A/P boundary in all discs; a leg disc is shown (Fig. 7C). HA-inv84 includes fragments C & D which both give expression in a line at the A/P boundary. These data show that this 84kb fragment does not contain an enhancer for expression in the posterior compartment of imaginal discs.

Fig. 7. HA-inv84 expression.

Fig. 7

Open in a new tab

HA-inv84@VK33 is expressed in embryos and larvae with wild type inv/en. (A) Stage 13 embryo, anterior left, dorsal up stained with anti-HA (red) and anti-En (green). HA-inv is seen in the hindgut (white arrow), but not the posterior spiracles (white oval). (B) Stage 12 embryo, ventral view, anterior left, head region only. HA is seen in the clypeolabrum, white * is background staining. (C) Leg disc. HA-inv is expressed in a line at the anterior-posterior boundary. Some HA-expression is also seen in the anterior compartment (white *); this expression is variable.

Two N-terminally tagged En constructs were made. HA-en79 contains 50kb of upstream sequences (including the 3’end of tou) and extends into the middle of the inv transcription unit (Fig. 5; Table 1). Based on our analysis, we would expect that it contains all of the inv/en regulatory DNA for expression in embryos. Consistent with this, the embryonic expression pattern of HA-en79 mimics en (Fig. 8A). Note that we have not examined the central nervous system expression in detail and are not sure of the accuracy of HA-en expression in this tissue. Not predicted by our dissection experiments, expression of HA-en79 in imaginal discs also mimics wild-type en (Fig. 9). Fig. 9B shows the expression of HA-en79 and the truncated Inv* protein in a wing disc from a HA-en79@attP40 enX31/HA-en79@attP40 enE larvae, that contains Inv* but no endogenous En. Expression of the truncated Inv protein from the enE chromosome is quite weak and broader than expected, perhaps reflecting a longer half-life and altered distribution of the non-nuclear protein. Some en79@attP40 enX31/HA-en79@attP40 enE flies survive, are fertile, and look phenotypically normal except they hold their wings out and are missing some bristles on the scutellum. In contrast, HA-en45 cannot rescue inv en double mutants. Embryonic HA-en45 stripes mimic inv/en stripes, but HA-en45 is not expressed in the peripheral nervous system (Fig. 8B) and the number of cells in the central nervous system that express HA-en45 is reduced compared with HA-en79 and wild type En (data not shown). HA-en45 does not contain fragment V that contains enhancers for the PNS and CNS (Fig. 1B). It is perhaps noteworthy that HA-en45 lacks fragment T, a stripe enhancer, yet has normal stripes throughout embryonic development. This suggests a redundancy in the embryonic stripe enhancers.

Fig. 8. HA-en expression from HA-en45 and HA-en79 in embryos.

Fig. 8

Open in a new tab

Stage 12 embryos (anterior left, dorsal up) showing HA-en expression (red) and Inv expression (green). White arrows points to the location of a single cell in the PNS that expresses wildtype Inv/En and is also expressed from the HA-en79 line (A), but not the HA-en45 line (B).

Fig. 9. HA-en expression from HA-en45 and HA-en79 in imaginal discs.

Fig. 9

Open in a new tab

(A) En (red) and Inv (green) expression in a 3rd instar wild type wing disc. (B) HA-en and Inv* (here detecting the truncated Inv* protein made from the enE chromosome) from a 3rd instar HA-en79@attP40 enE/HA-en79@attP40 enX31 wing disc. (C) HA-en and Inv in a 3rd instar wing disc from HA-en45 enB86 homozygote.

HA-en45 is able to rescue en mutants as long as at least one copy of inv is present. enB86 is an en null mutation with a 53bp deletion within the en coding region (Gustavson et al, 1996); the rest of the inv/en domain is intact. HA-en45@attP40 enB86 flies survive as homozygotes, and are fertile with no phenotypic defects. The expression of HA-en in HA-en45@attP40 enB86 wing discs is quite interesting. Though somewhat variable, we consistently see a big gap in HA-en staining in the middle of the wing pouch (Fig. 9C). Despite this, these wings look phenotypically normal, suggesting two things: 1) Inv expression is sufficient for normal wing development in these animals and (2) the abnormal HA-en expression does not disrupt wing development. HA-en45 is also expressed in legs discs in these animals, in a variegated manner (data not shown). These data show that an enhancer(s) for expression in the posterior compartment is contained within the HA-en45 transgene, but that it is not robust in this context.

Discussion

en and inv expression is controlled by a set of shared regulatory elements. The majority of these elements are contained within a 79-kb HA-en transgene that can rescue inv en double mutants to viable, fertile adults. Our analysis shows that many embryonic enhancers are located within discrete DNA fragments that can act autonomously to give specific subsets of inv/en expression. In contrast, we were not able to find a single small DNA fragment that recapitulated inv/en expression in the posterior compartment of imaginal discs.

Multiple enhancers exist for many aspects of en expression (Fig. 1, Table 1, Fig. 3A). These include multiple stripe enhancers, multiple enhancers for expression at the A/P boundary in imaginal discs, and multiple enhancers for expression in the clypeolabrum. Such duplication of enhancers is commonly seen in Drosophila developmental genes, and contributes to the robustness of expression (Yao et al., 2008; Perry et al., 2010; Frankel et al., 2010; Barolo, 2012; Fujioka and Jaynes, 2012). Thus for en and inv, there is redundancy both at the protein and the regulatory DNA level.

The role of inv

Our data are consistent with a previous report that inv function is dispensable for viability in the laboratory (Gustavson et al., 1996). Further, we show that most of the inv locus can be deleted without a loss of viability, showing that no vital regulatory DNA is present within inv. It was previously suggested that Inv and En play different roles in the patterning of the Drosophila wing (Simmonds et al., 1995). Our data are not consistent with this. First, invΔ24 and invΔ32 flies are phenotypically normal suggesting that En can fully substitute for Inv in wing development. Second, Inv can substitute for En in wing development in flies of the genotype HA-en45@attP40 enB86. In these flies, en protein is provided by the HA-en45 transgene, while Inv is expressed from the inv enB86 genomic region; HA-en is only expressed in the subset of the wing imaginal disc, while Inv expression is normal and the wings of these flies are phenotypically normal (Fig. 9 and data not shown). Note that HA-en45@attP40 cannot rescue flies that are inv en double mutants. This suggests that Inv can fully replace En function in many aspects of development.

Another curiosity of en genetics can also be explained by the presence of Inv. Breakpoint mutations in the en region could be categorized as “lethal” or “V;non-lethal” based on their ability to complement the lethality of en point mutants. Lethal breakpoint mutations were mapped to a 50 kb region from the 3’end of inv to about 20kb upstream of the en transcription start site; non-lethal mutations mapped to a region beginning about 30kb upstream of en to 50kb upstream (near the 3’end of tou) (Kuner et al., 1985). However, our HA-transgene rescue experiments show that regulatory DNA located more than 30kb upstream of en is required for rescue of inv en double mutants. We suggest that the complementation seen when crossing “non-lethal” en breakpoint mutants with en point mutants occurs because (1) en is expressed early in stripes from the breakpoint chromosome (inv expression is not sufficient at this stage) and (2) inv is expressed later in development, driven by regulatory DNA on the en point mutant chromosome, rescuing the lethality of the breakpoint chromosomes (which is unable to provide full function at this stage due to missing regulatory DNA). The inv gene is conserved (Peel et al., 2006) and mayplay an important role in natural populations of Drosophila that is not evident in the laboratory.

Where is the enhancer for expression of the inv/en in the posterior compartment of imaginal discs?

We expected, but did not find, a small fragment of DNA that drove expression in the posterior compartment of imaginal discs. Dissection of the brinker locus yielded multiple, discrete imaginal disc enhancers that drive lacZ in a pattern similar to brinker (Yao et al., 2008). In contrast, within the Ubx gene, no discrete DNA fragment was found that causes expression in only the haltere and third leg imaginal discs, where Ubx is expressed. Instead there is a DNA fragment that gives reporter gene expression in all discs (the imaginal disc enhancer, IDE). When the IDE is combined with an Ubx embryonic enhancer and a Polycomb response element (PRE), the reporter construct mimics Ubx expression (Müller and Bienz,1991; Castelli-Gair et al., 1992; Chan et al., 1994; Christen and Bienz,1994). Inv and En are Polycomb-regulated genes (Moazed and O’Farrell, 1990) and are within a H3K27me3 (histone H3, lysine 27 tri-methyl) domain that extends from the 3’ end of E(Pc) to the 3’ end of tou (Kharchenko et al., 2009). Similar to Ubx, the spatially restricted imaginal disc pattern of inv/en could result from a general disc enhancer(s) that, when combined with an early stripe enhancer(s) (to provide patterning information), and a PRE (to provide memory), results in expression of inv/en in the posterior compartment of imaginal discs. In our analysis, fragment 0 gave both striped expression and nearly ubiquitous expression of lacZ in all the imaginal discs. We reasoned that addition of a PRE to this fragment might lead to lacZ expression in the posterior compartment of imaginal discs, but this did not occur (data not shown). Fragment S also contains an imaginal disc enhancer but curiously, it gives expression in the anterior compartment of discs, the opposite of where inv/en are expressed (Fig. 1B). In a survey of cis-regulatory activity of 6,300 genomic fragments in imaginal discs, a 3.8kb DNA fragment (fragment 94D09) that overlaps with fragment S also caused expression in the anterior compartment in imaginal discs (Jory et al., 2012). In that survey, no fragment of DNA from the inv/en region gave expression in the posterior compartment of imaginal discs. Our analysis shows that the imaginal disc enhancer(s) is within a 36kb region—from 14 to 50kb upstream of en. Further, our data show that in order to get expression in the posterior compartment of imaginal discs, the imaginal disc enhancers needs to be in the context of other en regulatory DNA.

A few curiosities and final thoughts

Two results are very puzzling to us. The first is the lack of expression of HA-inv in posterior spiracles of embryos that have the HA-inv84@VK33 transgene (Fig. 7). Inv, like En, is normally expressed in posterior spiracles (PS). We suggest that the PS enhancer is able to act over a short distance to activate reporter gene expression when it is cloned next to the en promoter in the P[en] reporter construct. In contrast, when in the context HA-inv84, it is not able to activate the inv promoter located about 45kb away. We would have liked to see if this was due to a position effect, ie is the DNA surrounding the VK33 insertion site inhibiting the activity of the PS enhancer. However, we were not able to test this hypothesis because, despite numerous tries by two companies, no transgenic flies were ever recovered with HA-inv84 inserted at attP3 or attP40. Similarly, we were not able to obtain transgenic flies with HA-en79 inserted at VK33. While these results might suggest that something in these constructs prevents their integration into particular attP sites, recovery of transgenic flies for all three of the large phi-C31 constructs took numerous injections, and it is possible that with more injections we would eventually obtain additional lines.

Another surprising result is the expression of Inv in stripes of the thoracic segments late in embryonic development in enΔ17 embryos. enΔ17 removes the late enhancer present in fragment H, the only enhancer that drives stripes in the thoracic segments late in embryonic development. We suggest that the late thoracic stripes in enΔ17 are due to epigenetic memory. enΔ17 did not delete fragments M an odd-stripe early enhancer and 0 and T, that contain response elements for wg signaling and wg-independent autoregulation. In enΔ17 embryos we suggest that thoracic stripes are established by the odd-stripe enhancer, then driven by enhancers in fragment 0 and T, and finally maintained in an active state via the trithorax group of genes. It is interesting that even in the brinker locus, which is not Polycomb-regulated, putting multiple enhancers together can modify their function (Yao et al., 2008). We suggest that a similar case occurs at en and other large loci, i.e., you can take it apart, and find active components, but in the intact locus, there is cross talk between enhancers, and the transcriptional output is due to a combination of inputs from all the enhancers. In addition, at inv/en there is regulation by the Polycomb and trithorax group genes. Given this complexity, it is perhaps remarkable that so many of the embryonic enhancers are present in discrete DNA fragments.

Highlights.

  • en and inv share regulatory DNA that extends over a 62kb region.

  • Embryonic enhancers are contained within relatively small, discrete fragments.

  • No small fragment of DNA could recapitulate inv/en expression in imaginal discs.

  • A 79-kb HA-tagged en transgene can rescue inv en double mutants.

  • There is redundancy in both regulatory DNA and protein function in inv and en.

Acknowledgements

We thank Tom Kornberg for en lambda phage, en and inv mutant flies, and encouragement; Pat O’Farrell for en lambda phage and anti-En antibody; the Harvard Exelixis and Bloomington stock centers for fly stocks; and Miki Fujioka, Jim Jaynes, and Payal Ray and two helpful reviewers for comments on this manuscript. This work was supported by the Intramural Research Program of the Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Barolo S. Shadow enhancers: Frequently asked questions about distributed cis-regulatory information and enhancer redundancy. Bioessays. 2012;34:135–141. doi: 10.1002/bies.201100121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Brower DL. engrailed gene expression in Drosophila imaginal discs. EMBO J. 1986;5:2649–2656. doi: 10.1002/j.1460-2075.1986.tb04547.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Cadigan KM, Grossniklaus U, Gehring WJ. Localized expression of Sloppy Paired protein maintains the polarity of Drosophila parasegments. Genes Dev. 1994;8:899–913. doi: 10.1101/gad.8.8.899. [DOI] [PubMed] [Google Scholar]
  4. Campos-Ortega JA, Hartenstein V. The embryonic development of Drosophila melanogaster. Second Edition. Germany: Springer-Verlag Berlin Heidelberg; 1997. [Google Scholar]
  5. Castelli-Gair J, Müller J, Bienz M. Function of an Ultrabithorax minigene in imaginal cells. Development. 1992;114:877–886. doi: 10.1242/dev.114.4.877. [DOI] [PubMed] [Google Scholar]
  6. Chan C-S, Rastelli L, Pirrotta V. A Polycomb response element in the Ubx gene that determines an epigenetically inherited state of repression. EMBO J. 1994;13:2553–2564. doi: 10.1002/j.1460-2075.1994.tb06545.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Christen B, Bienz M. Imaginal disc silencers from Ultrabithorax: evidence for Polycomb response elements. Mech. Dev. 1994;48:255–266. doi: 10.1016/0925-4773(94)90064-7. [DOI] [PubMed] [Google Scholar]
  8. Coleman KG, Poole SJ, Weir MP, Soeller WC, Kornberg T. The invected gene of Drosophila: sequence analysis and expression studies reveal a close kinship to the engrailed gene. Genes Dev. 1987;1:19–28. doi: 10.1101/gad.1.1.19. [DOI] [PubMed] [Google Scholar]
  9. DeVido SK, Kwon D, Brown JL, Kassis JA. The role of Polycomb-group response elements in regulation of engrailed transcription in Drosophila. Development. 2008;135:669–676. doi: 10.1242/dev.014779. [DOI] [PubMed] [Google Scholar]
  10. DiNardo S, Kuner JM, Theis J, O’Farrell PH. Development of embryonic pattern in D. melanogaster as revealed by accumulation of the nuclear Engrailed protein. Cell. 1985;43:59–69. doi: 10.1016/0092-8674(85)90012-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. DiNardo S, O’Farrell PH. Establishment and refinement of segmental pattern in the Drosophila embryo-spatial control of engrailed expression by pair-rule genes. Genes Dev. 1987;1:1212–1225. doi: 10.1101/gad.1.10.1212. [DOI] [PubMed] [Google Scholar]
  12. DiNardo S, Sher E, Heemskerk-Jongens J, Kassis JA, O’Farrell PH. 2-tiered regulation of spatially patterned engrailed gene-expression during Drosophila embryogenesis. Nature. 1988;332:604–609. doi: 10.1038/332604a0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Ecker R. The recessive mutant engrailed in Drosophila melanogaster. Hereditas. 1929;12:217–222. [Google Scholar]
  14. Frankel N, Davis GK, Vargas D, Wang S, Payre F, Stern DL. Phenotypic robustness conferred by apparently redundant transcriptional enhancers. Nature. 2010;466:490–493. doi: 10.1038/nature09158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Fujioka M, Jaynes JB. Regulation of a duplicated locus: Drosophila sloppy paired is replete with functionally overlapping enhancers. Dev. Biol. 2012;362:309–319. doi: 10.1016/j.ydbio.2011.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Garcia-Bellido A, Santamaria P. Developmental analysis of the wing disc in the mutant engrailed of Drosophila melanogaster. Genetics. 1972;72:87–104. doi: 10.1093/genetics/72.1.87. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gustavson E, Goldsborough AS, Ali Z, Kornberg TB. The Drosophila engrailed and invected genes: Partners in regulation, expression, and function. Genetics. 1996;142:893–906. doi: 10.1093/genetics/142.3.893. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hama C, Ali Z, Kornberg TB. Region-specific recombination and expression are directed by portions of the Drosophila-engrailed promoter. Genes Dev. 1990;4:1079–1093. doi: 10.1101/gad.4.7.1079. [DOI] [PubMed] [Google Scholar]
  19. Heemskerk J, DiNardo S, Kostriken R, O’Farrell PH. Multiple modes of engrailed regulation in the progression towards cell fate determination. Nature. 1991;352:404–410. doi: 10.1038/352404a0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Jaynes JB, Fujioka M. Drawing lines in the sand: even skipped et al. and parasegment boundaries. Dev. Biol. 2004;269:609–622. doi: 10.1016/j.ydbio.2004.03.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Jory A, Estella C, Giorgianni MW, Slattery M, Laverty TR, Rubin GM, Mann RS. A survey of 6,300 genomic fragments for cis-regulatory activity in the imaginal discs of Drosophila melanogaster. Cell Reports. 2012;2:1014–1024. doi: 10.1016/j.celrep.2012.09.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Junion G, Spivakov M, Girardot C, Braun M, Gustafson EH, Birney E, Furlong EEM. A transcription factor collective defines cardiac cell fate and reflects lineage history. Cell. 2012;148:473–486. doi: 10.1016/j.cell.2012.01.030. [DOI] [PubMed] [Google Scholar]
  23. Kassis JA. Spatial and temporal control elements of the Drosophila engrailed gene. Genes Dev. 1990;4:433–443. doi: 10.1101/gad.4.3.433. [DOI] [PubMed] [Google Scholar]
  24. Kharchenko PV, Alekseyendo A-A, Schwartz YB, Minoda A, Riddle NC, et al. Comprehensive analysis of the chromatin landscape in Drosophila melanogaster. Nature. 2009;471:480–485. doi: 10.1038/nature09725. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kornberg T. engrailed: A gene controlling compartment and segment formation in Drosophila. Proc. Natl. Acad. Sci. USA. 1981;78:1095–1099. doi: 10.1073/pnas.78.2.1095. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kuner JM, Nakanishi M, Ali Z, Drees B, Gustavson E, Theis J, Kauvar L, Kornberg T, O’Farrell PH. Molecular cloning of engrailed: A gene involved in the development of pattern in Drosophila melanogaster. Cell. 1985;42:309–316. doi: 10.1016/s0092-8674(85)80126-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kwon D, Mucci D, Langlais KK, Americo JL, DeVido SK, Cheng Y, Kassis JA. Enhancer-promoter communication at the Drosophila engrailed locus. Development. 2009;136:3067–3075. doi: 10.1242/dev.036426. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Lawrence PA, Morata G. Control of compartment development by the engrailed gene in Drosophila. Nature. 1975;255:614–617. doi: 10.1038/255614a0. [DOI] [PubMed] [Google Scholar]
  29. MacArthur S, Li X-Y, Li J, Brown JB, Chu HC, et al. Developmental roles of 21 Drosophila transcription factors are determined by quantitative differences in binding to an overlapping set of thousands of genomic regions. Genome Biology. 2009;10:R80. doi: 10.1186/gb-2009-10-7-r80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Manoukian AS, Krause HM. Control of segmental asymmetry in Drosophila embryos. Development. 1993;118:785–796. doi: 10.1242/dev.118.3.785. [DOI] [PubMed] [Google Scholar]
  31. Martinez-Arias A, Baker NE, Ingham PW. Role of segment polarity genes in the definition and maintenance of cell states in the Drosophila embryo. Development. 1988;103:157–170. doi: 10.1242/dev.103.1.157. [DOI] [PubMed] [Google Scholar]
  32. Moazed D, O’Farrell PH. Maintenance of the engrailed expression pattern by Polycomb group genes in Drosophila. Development. 1992;116:805–810. doi: 10.1242/dev.116.3.805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Müller J, Bienz M. Long range repression conferring boundaries of Ultrabithorax expression in the Drosophila embryo. EMBO J. 1991;10:3147–3155. doi: 10.1002/j.1460-2075.1991.tb04876.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Nüsslein-Volhard C, Wieschaus E. Mutations affecting segment number and polarity in Drosophila. Nature. 1980;287:795–801. doi: 10.1038/287795a0. [DOI] [PubMed] [Google Scholar]
  35. Parks AL, Cook KR, Belvin M, Dompe NA, Fawcett R, et al. Systematic generation of high-resolution deletion coverage of the Drosophila melanogaster genome. Nature Genetics. 2004;36:288–292. doi: 10.1038/ng1312. [DOI] [PubMed] [Google Scholar]
  36. Peel AD, Telford MJ, Akam M. The evolution of hexapod engrailed-family genes: evidence for conservation and concerted evolution. Proc. Biol. Sci. 2006;273:1733–1742. doi: 10.1098/rspb.2006.3497. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Perry MW, Boettiger AN, Bothma JP, Levine M. Shadow enhancers foster robustness of Drosophila gastrulation. Curr. Biol. 2010;20:1562–1567. doi: 10.1016/j.cub.2010.07.043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Qian S, Varjavand B, Pirrotta V. Molecular analysis of the zeste-white interaction reveals a promoter-proximal element essential for distant enhancer-promoter communication. Genetics. 1992;131:79–90. doi: 10.1093/genetics/131.1.79. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Venken KJT, Carlson JW, Schulze KL, Pan H, He Y, Spokony R, Wan KH, Koriabine M, de Jong PJ, White KP, Bellen HJ, Hoskins RA. Versatile P[acman] BAC libraries for transgenesis studies in Drosophila melanogaster. Nature Methods. 2009;6:431–434. doi: 10.1038/nmeth.1331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Warming S, Constantino N, Court DL, Jenkins NA, Copeland NG. Simple and highly efficient BAC recombineering using galK selection. Nucl. Acids Res. 2005;33:e36. doi: 10.1093/nar/gni035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Yao L-C, Phin S, Cho J, Rushlow C, Arora K, Warrior R. Multiple modular promoter elements drive graded brinker expression in response to the Dpp morphogen gradient. Development. 2008;135:2183–2192. doi: 10.1242/dev.015826. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES