Download New functions of the Drosophila rhomboid gene

Development 120, 2329-2338 (1994)
Printed in Great Britain © The Company of Biologists Limited 1994
2329
New functions of the Drosophila rhomboid gene during embryonic and adult
development are revealed by a novel genetic method, enhancer piracy
Rebekka Noll†, Mark A. Sturtevant, Raghava R. Gollapudi and Ethan Bier*
Department of Biology and Center for Molecular Genetics, University of California, San Diego, La Jolla, CA 92093, USA
*Author for correspondence
†Present address: Department of Cellular and Developmental Biology, Harvard Medical School, Boston Massachusetts 02115, USA
SUMMARY
Localized expression of the Drosophila rhomboid (rho) gene
has been proposed to hyperactivate EGF-Receptor
signaling in specific cells during development of the embryo
and adult. In this report we use a novel transposon based
genetic method, enhancer piracy, to drive ectopic
expression of a rho cDNA transgene by endogenous
genomic enhancers. Many enhancer piracy transposon-rho
insertions cause dominant phenotypes, over half of which
cannot be duplicated by ubiquitous expression of rho.
Genetic interactions between various dominant enhancer
piracy alleles and mutations in the EGF-R/RAS signaling
pathway indicate that many of these novel phenotypes
result from ectopic activation of EGF-R signaling.
Patterned mis-expression of the rho cDNA transgene correlates in several cases with localized dominant enhancer
piracy phenotypes. Enhancer piracy lines reveal an unanticipated role for rho in imaginal disc formation and
provide the first evidence that mis-expression of rho is sufficient for converting entire intervein sectors into veins.
Enhancer piracy may prove to be a general strategy for
obtaining dominant alleles of a gene of interest in diverse
insects, worms, plants, and potentially in vertebrates such
as mice and fish.
INTRODUCTION
signaling as ubiquitous expression of rho from a heat shock
construct generates several dominant phenotypes consistent
with activating the EGF-R signaling pathway in inappropriate
cells. Various phenotypes have been recovered in heat shock
experiments depending on the stage at which heat induction is
administered (Freeman et al., 1992; Sturtevant et al., 1993;
Ruohola-Baker et al., 1993; J.W. O’Neill and E. Bier, unpublished data; M. Sturtevant and E. Bier, unpublished data – see
Table 1 in this study), confirming the expectation that localized
expression of rho is important for targeting EGF-R signaling
to appropriate cells at a variety of developmental stages.
One difficulty in analyzing the role of ectopic rho expression
during many developmental stages such as embryogenesis is
that rho is often expressed in many different cell types simultaneously. This complicates interpretation of resulting heat
shock phenotypes since mixtures of supernumerary and
missing cell types are expected. One solution to this problem
is to provide ectopic rho expression in a limited number of
cells. To this end we employed enhancer piracy, a novel
genetic approach similar to that used in the enhancer trapping
method. Enhancer trapping has proved to be a versatile and
powerful method for identifying genes based on the activity of
neighboring enhancer elements (O’Kane and Gehring, 1987;
Bier et al., 1989; Bellen et al., 1989). Enhancers functioning
at a distance can activate the expression of a reporter gene
carried on a transposon inserted at neighboring chromosomal
The Drosophila rhomboid (rho) gene is expressed in a complex
localized pattern during embryogenesis and imaginal disc
development and is required for the differentiation of many of
the primordia in which it is expressed (Mayer and NüssleinVolhard, 1988; Bier et al., 1990; Sturtevant et al., 1993).
Several lines of evidence suggest that one important function
of rho is to participate as an accessory molecule to promote
signaling through the EGF-Receptor (EGF-R)/RAS pathway.
The clearest evidence that rho functions to amplify EGF-R
signaling derives from the detailed similarity of embryonic loss
of function phenotypes for rho (Mayer and Nüsslein-Volhard,
1988; Bier at al., 1990), Egf-r (Clifford and Schüpbach, 1992;
Raz and Shilo, 1992, 1993), and spitz (Mayer and NüssleinVolhard, 1988) which encodes an EGF-like ligand (Rutledge
et al., 1992); from a wealth of self consistent genetic interactions between rho alleles and mutations in components of the
EGF-R signaling pathway (Sturtevant et al., 1993); and from
evidence that rho acts upstream of Egf-r in oogenesis
(Ruohola-Baker et al., 1993). An interesting feature of regulation of EGF-R signaling by rho is that the ligand and receptor
are ubiquitously expressed during embryogenesis and imaginal
disc development, while rho expression is highly localized and
is mostly confined to cells requiring the activity of EGF-R.
Localized expression of rho is critical for restricting EGF-R
Key words: rhomboid, Egf-r, Drosophila, enhancer piracy, enhancer
trapping, transposon, dominant allele, genetic interaction
2330 R. Noll and others
sites (Bier et al., 1989; Wilson et al., 1989). A remarkable
feature of enhancer trap screens is that over half of the sites of
random transposon insertion express the reporter gene in some
tissue-specific or spatially localized pattern (O’Kane and
Gehring, 1987; Bier et al., 1989; Bellen et al., 1989). We
reasoned that the same genetic strategy could be applied to
driving the localized expression of an active gene of interest in
place of the passive reporter gene. Thus, mobilization of an
enhancer trap-like vector carrying a cDNA of interest should
result in transposon insertions near endogenous genomic
enhancers which could drive localized expression of the
cDNA. Since ubiquitous expression of rho yields dominant
phenotypes at several stages of development (Sturtevant et al.,
1993) and because several of the original HS-rho transformants
have dominant constitutive extra wing vein phenotypes
(Sturtevant et al., 1993), we expected that a more systematic
attempt to generate dominant alleles might reveal new
functions of rho. Furthermore, such dominant phenotypes
could serve as valuable genetic tools for isolating second site
suppressors or enhancers of the dominant HS-rho phenotypes.
In this paper we use enhancer piracy to generate a variety of
dominant rho alleles. Over half of these enhancer piracy phenotypes cannot be mimicked by ubiquitous expression of rho,
although we recovered enhancer piracy insertion lines with
dominant phenotypes typical of all those observed following
various heat shock treatments of HS-rho flies. Some of these
new dominant rho alleles provide important insights into the
developmental function of rho. In several cases we determined
that enhancer piracy lines exhibiting localized defects express
the rho transgene in a pattern corresponding to the primordia
for those affected structures. We discuss the potential of
enhancer piracy as a versatile general method for obtaining
dominant alleles of Drosophila genes and the possibility of
employing similar strategies in other organisms.
MATERIALS AND METHODS
Fly stocks
All genetic markers and chromosome balancers used are described in
Lindsley and Grell (1968) and Lindsley and Zimm (1992). Crosses to
generate transpositions from the X chromosome to autosomes using
a genomic source of transposase (∆2-3) were performed as described
in Bier et al. (1989). Several strains of flies carrying X chromosome
insertions of the HS-rho P-element construct (Sturtevant et al., 1993)
were tested in pilot studies according to the scheme depicted in Fig.
1 to determine which line gave the optimal transposition frequency.
The line with the highest transposition frequency (60% of vial crosses
using 2 HS-rho; ∆2-3 Sb/+ males × 5 yw females gave at least one
jump) was used to generate the majority of lines generated for this
study. Other stocks were obtained from the Bloomington, Indiana
Drosophila Stock Center and the Bowling Green, Ohio Stock Center.
Mounting fly wings
Wings from adult flies were dissected in isopropanol and mounted in
Canadian Balsam mounting medium (Gary’s Magic Mount) following
the protocol of Lawrence et al. (in Roberts, 1986). Mounted wings
were photographed under Nomarski optics with a 4× lens on a
compound microscope. Alternatively, whole flies or portions of flies
were photographed through a dissection microscope.
In situ hybridization to whole-mount embryos or discs
In situ hybridization to whole-mount discs and embryos was
performed using digoxigenin-labeled RNA probes (BoehringerMannheim, 1093 657) according to the method Tautz and Pfeiffle
(1989), using 4 µg/ml Proteinase K instead of 40 µg/ml as required
for digoxigenin-labeled DNA probes. rho transgene expression was
visualized with a 850 bp probe specific to the 3′ transcribed SV40
sequences derived from the HS-rho vector (Sturtevant et al., 1993).
No signal is observed when using this 3′ trailer probe on wild-type
embryos.
Other molecular techniques
RNA probe synthesis was performed according to BoehringerMannheim protocols and other cloning techniques followed standard
procedures, as described by Maniatis et al. (1982).
RESULTS
Enhancer piracy is an efficient method for
recovering dominant rho phenotypes
We mobilized a P-element heat shock vector (hs-CaSpeR; see
Bang and Posakony, 1992) carrying the rho protein coding
region (HS-rho, see Sturtevant et al., 1993) from the X chromosome to autosomal sites using the genetic scheme depicted
in Fig. 1. We chose a heat shock vector for this screen because
many fly strains already exist carrying various cDNA inserts
in heat shock vectors. As the hs-CaSpeR vector provides high
levels of ubiquitous expression following heat shock induction
and functions efficiently for generating enhancer piracy lines
(see below), it is an excellent vector for both of these purposes
and is consequently ideal for mis-expressing a gene of interest.
Individual transformant flies were scored for dominant
yw
;
+
Sb
2-3 p[y ]
TM6Ubx
w p[HS-rho, w + ]
Sb
;
+
+
w p[HS-rho, w +]
;
w p[HS-rho, w + ]
X
+
2-3 p[y ]
X
+
yw
yw
;
+
+
(white eyes)
Transposition to an autosome
yw
orange
eyes
yw
;
;
p[HS-rho, w +]
+
+
+
;
Establish Stocks
;
+
+
p[HS-rho, w + ]
X
yw
yw
+
Test for dominant
sterility
2-3 = p-element transposase (defective for transposition)
p[HS-rho, w +] = p-element vector containing the rho coding region
under the control of the hsp70 promoter. The vector
also contains the dominant white (w+) marker.
Fig. 1. Genetic scheme for generating rhomboid enhancer piracy
transposon insertions. A single male transformant obtained from the
screen was immediately scored for phenotype and was then mated to
yw females to confirm that the dominant phenotype was heritable and
to establish permanent stocks for interesting phenotypes.
Heterozygous females from 113 lines were mated to yw males to
determine whether any insertions were dominant female sterile.
Enhancer piracy and rhomboid function 2331
Table 1. Enhancer piracy phenotypes due to HS-rho insertions
X
Y
Z
X
X+Y+Z
Observed
with heat
shock††
21
108
75
16
1
(−)†
(−)
(−)
9
1
(−)
(−)
(−)
14
3
8.5
43.5
30.2
6.5
0.4
8.5
43.5
30.2
15.7
2.0
NA
+
+/−
+/−
−
No. of lines (248 total)
Phenotype
Wing
None*
Weak‡
Moderate§
Strong¶
Serrated
% of total lines
Thorax
Haltereless
Shoulderless
Duplicated wing
2
0
0
6
6
2
0
6
0
0.8
0
0
3.2
4.8
0.8
−
−
−
Bristle
Extra orbitals
1
8
0
0.4
3.6
−
Eye
Rough eye
2
4
2
0.8
3.2
+
Leg
Malformed
0
2
2
0
1.6
−
Other
Female sterile**
1
0
0
0.9
0.9
+
X = >50% of individuals show phenotype.
Y = <50 % of individuals show phenotype (typically 20-40%).
Z = Primary transformant showed phenotype but either died or failed to mate.
*None = wild-type wing.
†(−) = Wing phenotypes of moderate or less than moderate strength were only scored if highly penetrant.
‡Weak = One or more of: very short extra vein segments, small deltas, and very short thickened vein segments.
§Moderate = One or more of: extended thickened veins, long extra vein segments, large and connected deltas, and small blisters.
¶Strong = One or more of: widespread thickened veins, large blisters, and shriveled wings.
**113 lines were tested for dominant female sterility.
††Phenotypes that have been observed with heat shock are marked with (+); phenotypes never been recovered with heat shock are marked with (−); and for
those marked with (+/−) some but not all of the phenotypes in this category were oberved with heat shock. For example, reproducibly localized blisters (moderate
wing - Fig. 2C) and extremely thick veins (strong wing - Fig. 2D) were not recovered in heat shock experiments.
NA = not applicable.
phenotypes at the time they were recovered and were then
mated to yw flies to determine whether the observed phenotypes
segregated with the HS-rho vector (marked with w+). The
degree of penetrance (i.e. greater or less than 50%) was also
determined for lines with phenotypes. The kinds of phenotypes
recovered from screening of 248 independent autosomal rho
enhancer piracy insertions and the relative frequency of these
phenotypes are summarized in Table 1. Many of these phenotypes have not been recovered in exhaustive heat shock
induction experiments (Table 1; J.W. O’Neill and E. Bier
unpublished data; M. Sturtevant and E. Bier, unpublished data).
Enhancer piracy phenotypes reveal novel
requirements for localized rho expression
Wing phenotypes, including a variety of strengths of excess
vein material or blisters, were recovered at high frequency
(Fig. 2B-D). This is consistent with previous data suggesting
that even low levels of ectopic rho expression during wing
development generate excess vein phenotypes (Sturtevant et
al., 1993). Some lines exhibited localized wing venation
defects, such as segments of thickened veins or small blisters
restricted to particular regions of the wing (Fig. 2C). The conversion of complete intervein sectors into vein tissue (between
arrows, but not between veins L3 and L4) in the line shown in
Fig. 2D reveals that all intervein cells in these sectors of the
wing can be induced to assume vein fates, a fact not illumi-
nated by heat shock experiments. An interesting category of
wing phenotype also not recovered by heat shock is a serrated
wing margin.
Another class of phenotypes only recovered in these
enhancer piracy experiments involves thoracic structures.
Tissues derived from haltere and wing discs are affected to
varying degrees in the enhancer piracy line shown in Fig. 2FH. The most common defect is a loss of one or both halteres
(Fig. 2F). Strongly affected individuals may also lack part of
the wing, resulting in the formation of a mirror symmetric
double anterior wing (Fig. 2G), or may lack virtually all structures derived from the wing disc which is manifested as a
shoulderless phenotype (Fig. 2H). Intriguingly, similar loss of
imaginal tissue phenotypes and mirror duplications have
recently been reported for various combinations of alleles of
the ventro-lateral group gene pointed (Scholz et al., 1993).
Various allelic combinations of sna and neighboring genes also
result in similar loss of halteres and defective thoracic phenotypes (Ashburner et al., 1990). The basis for the
haltereless/shoulderless phenotypes was investigated in two
independent enhancer piracy lines. As haltere discs were frequently missing from third instar larvae of these lines, we
examined the appearance of the wing and haltere discs during
embryogenesis. Whole-mount embryos collected from each of
these lines were hybridized with a snail RNA probe, which
labels developing haltere and wing discs in wild-type embryos
Fig. 2. Enhancer piracy lines with dominant phenotypes. (A) A wild-type wing. (B) A line
with typical strong extra wing vein phenotype. (C) A line with localized distal blister
(arrow). (D) A line with severely thickened distal veins (between arrows). (E) A wild-type
thorax viewed from above. Note the location of the two halteres posterior to the wings
(arrows). (F-H) A line with various degrees of missing haltere and wing imaginal
structures. (F) An individual with two missing halteres (arrows). (G) More strongly
affected individuals with symmetric duplication of anterior wing structures. Arrow marks
axis of symmetry. (H) An individual with the most severe shoulderless phenotype.
Arrowhead marks dorsal midline. (I) A wild-type eye viewed from the side. (J) A line with
rough eyes and ectopic orbital bristles (arrow). (K) A wild-type chorion. Note the location
of dorsal appendages (arrow) near the dorsal midline and displaced approximately 20% of
the length of the embryo in from the anterior tip. (L) A dominant female sterile line which
lays eggs having dorsalized chorions. Note the extreme anterior and ventrolateral location
of the dorsal appendages (arrow), consistent with a dramatic expansion of dorsal
positional values at the expense of ventral most structures. This panel is a composite of
two different focal planes so that the pattern of the eggshell can be seen as well as the
location of the dorsal appendages.
2332 R. Noll and others
Enhancer piracy and rhomboid function 2333
Fig. 3. Embryonic
phenotype of enhancer
piracy line missing haltere
and wing disc structures.
(A) Whole-mount in situ
hybridization of a snail
RNA digoxigenin-labeled
probe to a wild-type 12hour embryo (viewed from
the side with anterior to the
left). The haltere (h) and
wing (w) disc primordia are
labeled at this stage
(Alberga et al., 1991).
(B) snail expression in the
enhancer piracy line
missing haltere and wing
disc structures in the adult
(same line shown in Figs
2F-H). The haltere disc is
missing and the wing disc is reduced in size. (C) Two pairs of snail-expressing haltere and wing discs in a wild-type embryo viewed in
horizontal section. (D) An embryo from the same enhancer piracy line shown in B; two haltere discs and one wing disc are missing (viewed in
horizontal section).
(Alberga et al., 1991; Fig. 3A,C). Mutant embryos frequently
exhibit reduced numbers of sna-expressing haltere disc cells
and often entirely lack sna expression in at least one haltere
disc (Fig, 3B). sna expression was also affected in all or
portions of wing discs (Fig. 3D). In extreme cases, sna
expression in all haltere and wing disc primordia was eliminated (data not shown).
Although rough eyes, as illustrated by the enhancer piracy
line shown in Fig. 2J, have been observed following heat
induction of HS-rho lines during the early third larval instar
(Freeman et al., 1992), ectopic orbital bristles (arrow) have not
been recovered in heat shock experiments. A rare and poorly
penetrant phenotype observed in this screen was a distorted leg
similar to the ‘malformed’ phenotype described for mutants in
the broad locus (Gotwals and Fristrom, 1991). Finally, because
ectopic rho expression during oogenesis dorsalizes follicle
cells (Ruohola-Baker et al., 1993), we screened 113 lines for
dominant female sterility and found one highly penetrant
dominant female sterile line. Comparison of egg shells laid by
wild-type females (Fig. 2K) to those deposited by the dominant
female sterile line (Fig. 2L) revealed that the chorions are
indeed dorsalized.
Enhancer piracy alleles interact genetically with
mutations in the EGF-R/RAS signaling pathway
The basis for the various rho enhancer piracy phenotypes is
not known. However, based on our previous analysis of misexpressing rho during wing vein development (Sturtevant et
al., 1993), one explanation is that rho hyperactivates EGF-R
signaling in inappropriate cells and consequently disrupts
Table 2. Genetic interactions between enhancer piracy lines and mutants in components of the ventrolateral signaling
pathway
Test Mutant Genotype
Egf-rElp
+
Ras1e1b
+
±↓
0
±↓
0
↓
±↓
↑↑†
±↓
↑
0‡
26
94/361
19
7/37
20
2/10
0
0/54
52
29/56
0
0/29
0.5
2/392
18.5§,12.5§
35/191,10/80
0.4
1/248
26.5
102/384
0.2
1/462
25.1
120/479
46 (36)
39 (30)
15 (12)
70 (39)
20 (11)
11 (6)
52 (33)
38 (24)
11 (7)
57 (31)
32 (17)
11 (6)
28 (11)
38 (15)
35 (14)
78 (52)
19(13)
3 (2)
N=78
N=56
N=64
N=54
N=40
N=67
+
+
Df Egf-r
+
Extra orbital bristles*
all flies
Rough eye*
subtle
Serrated wing margin (%)
(no /total)
Female sterility (% hatched)
(no. hatched/total)
Enhancer piracy phenotype
Haltereless: 2 halteres: % (no.)
1 haltere: % (no.)
0 halteres: % (no.)
Total no. of flies
± indicates weak interaction; ↓ indicates suppression; ↑ indicates enhancement; 0 indicates no modification of the phenotype.
*Extra orbital and brisles and rough eye scored in same enhancer piracy line (see Fig. 2I,J).
†Eye is typically reduced to about half the size observed in Egf-rElp alone.
‡Eye has unmodified Star phenotype.
§Strong Egf-r alleles Egf-rIK35 (first entry) and Egf-r2L65 (second entry) were used in place of DfEgf-r.
Gap1B2
+
Starl(2)54
+
2334 R. Noll and others
normal development. It is also possible that some of the
dominant phenotypes observed result from interference with
developmental decisions that normally do not involve rho or
the EGF-R/RAS pathway. To distinguish between these possibilities, we crossed enhancer piracy lines with dominant phenotypes including extra orbital bristles, rough eyes, serrated
wings, or female sterility with flies carrying mutations in components of the EGF-R signaling
pathway and scored for changes in the
magnitude of the dominant phenotypes
in the trans-heterozygous progeny.
Results from these crosses, summarized
in Table 2, reveal that the various
enhancer piracy phenotypes are
strongly modified by mutations in the
EGF-R/RAS pathway in a pattern that
is consistent with rho acting together
with Star to promote signaling through
EGF-R in each of these cases. Thus,
heterozygosity for loss of function
mutations in Egf-r, Ras1, or Star tend to
suppress the various rho enhancer
piracy phenotypes, while the hyperactive Egf-rElp allele or a loss of function
Gap1 allele enhances these phenotypes.
As in the case of HS-rho lines having
extra vein phenotypes (Sturtevant et al.,
1993), Star is consistently the most
potent suppressor of dominant rho phenotypes.
As mentioned earlier, the dominant
female sterile enhancer piracy line lays
eggs with ventrally shifted dorsal
appendages (Figs 2J, 4A,B). We asked
whether the strong suppression of
female sterility observed with Egf-r,
Ras1, and Star (Table 2) was associated
with a reduction of the dorsalized
eggshell phenotype. The dorsalized
eggshell phenotype was partially suppressed by removing one active copy of
Egf-r (Fig. 4C,D) and eggs laid by
mothers trans-heterozygous for either
Ras1 (Fig. 4E,F) or for Star (Fig. 4G,H)
typically had dorsal appendages in
nearly wild-type locations (compare
with Fig. 2K). The location of the
dorsal appendages was not appreciably
altered by heterozygous Gap1 or by
Egf-rElp.
Expression of the rho transgene
in enhancer piracy lines
correlates with localized
phenotypes
To verify our expectation that the
localized phenotypes recovered in
several rho enhancer piracy lines were
due to spatially restricted expression of
the rho transgene, we hybridized
appropriate tissues with a probe
specific for the 3′ transcribed trailer provided by the HS vector
(Fig. 5). The pattern of rho transgene expression in wing discs
of the line having entire sectors of the wing converted to vein
(Fig. 2D) corresponds precisely to the affected regions of the
wing (Fig. 5B,C; between arrows, but not between veins L3
and L4). Consistent with the localized distal wing blister
phenotype of the line shown in Fig. 2C, ectopic expression of
Fig. 4. Suppression of the dorsalized eggshell phenotype by Ras1 and Star mutations. Eggs
were collected from mothers heterozygous for the dominant female sterile enhancer piracy
insertion alone (A,B) or in combination with a heterozygous test mutation (C-F). Arrows
indicate the anterior-posterior locations of the dorsal appendages relative to the anterior end of
the eggs, which are marked by a stars. (A,B) The eggs shown in these panels are typical
examples of dorsalized phenotypes observed in collections from this line. An example of
more strongly dorsalized eggs present in the same collection is shown in Fig. 2J. (C,D) Eggs
from female sterile mothers also heterozygous for Egf-r2L65. The dorsalized eggshell
phenotype is partially suppressed. (E,F) Eggs from female sterile mothers also heterozygous
for Ras1e1B. Note that the dorsal appendages are very close to the dorsal midline and at about
20% egg length as in wild type - see Fig. 2K for comparison. (G,H) Female sterile mothers
heterozygous for Stare(2)54 also lay eggs with nearly wild-type appearance.
Enhancer piracy and rhomboid function 2335
Fig. 5. Spatially restricted enhancer driven expression of the rho transgene in enhancer piracy lines with localized dominant phenotypes.
(A) Expression of endogenous rho gene during the late third instar in wing vein primordia. The future wing margin is labeled M and the
longitudinal vein primordia are numbered L1-L5. Arrows mark the primordia for the region between L2 and L5. (B) Expression of the rho
transgene during the late third instar in the enhancer piracy line with entire intervein sectors converted to veins (Fig. 2D). Continuous vein
sectors derive from the region between the arrows (i.e. between L2 and L5). There is a gap in transgene expression in the intervein sector
between primordia for L3 and L4 which provides an explanation for the fact that ectopic veins are excluded from this region in the transformant
line. (C) Expression of the rho transgene 7-8 hours after puparium formation (AP) in the same enhancer piracy line as that shown in B (see D
for the normal rho expression pattern in vein primordia at this stage). (D) Expression of endogenous rho gene at 7-8 hours AP in future wing
vein cells. (E) Expression of the rho transgene at 7-8 hours AP in the enhancer piracy line with a distal blister (Fig. 2C). (F) Lower
magnification view of E including rho transgene expression in the developing wing (w) haltere (h) and leg (l). (G) Localized expression of the
rho transgene in a strip of cells separating the eye (e) and antennal (a) primordia of the eye disc in the line with extra orbital bristles (Fig. 2J).
By this stage of the orderly array of developing ommatidial preclusters is irregular, and presumably underlies the final adult rough eye
phenotype. Ectopic expression of rho at earlier developmental stages in the eye disc, which is particularly intense in the morphogenetic furrow
(data not shown), may be responsible for this phenotype. (H,I) Expression of the rho transgene in the haltereless line shown in Fig. 2F-H. The
patches of cells in thoracic segments T2 and T3 are in the approximate locations where the wing and haltere disc cells develop. The
consequences of transgene expression in similar clusters of cells in gnathal segments G2 and G3, in T1, and in abdominal (A) segments are
unknown. cf, cephalic furrow.
the rho transgene in this line is restricted to a distal section of
the wing during the prepupal period (arrow in Fig. 5E). Other
discs also exhibit localized expression of the transgene at this
stage (Fig. 5F) including the distal tip of the haltere disc
(arrowhead) and stripes in the leg disc (stars). Restriction of
rho transgene expression to distal portions of both the wing
and haltere discs may reflect their common evolutionary
origin. The rho transgene in the transformant line shown in
Fig. 2J (which has rough eyes and extra orbital bristles) is
expressed in a strip of cells between the eye and antennal
domains of the eye-antennal disc (Fig. 5G). The extra bristles
observed in this transformant derive from this region of the
2336 R. Noll and others
disc (star). Earlier during eye disc development, the transgene
is expressed at high levels in the morphogenetic furrow (data
not shown). This early expression is likely to underlie the final
rough eye phenotype as heat induction experiments indicate
that mystery cells can be converted into photoreceptor
precursor cells by ectopic rho expression (Freeman et al.,
1992). Finally, we examined expression of the rho transgene
in embryos of the haltereless line shown in Fig. 2F-H. The
transgene is expressed in a segmentally repeated pattern of
small patches of cells at germband extension (Fig. 5H,I).
Groups of cells in thoracic segments T2 and T3 are at the
approximate locations from which wing and haltere discs subsequently form in wild-type embryos. In contrast, the endogenous rho gene is not expressed in wing or haltere disc
primordia during embryogenesis. The lethality of the
P-element insertion in the haltereless enhancer piracy line
shown in Fig. 2 F-G is not allelic to sna, eliminating the
remote possibility that embryos exhibiting a loss of sna
expression are homozygous for an insertion into the sna gene
itself. The isolation of a second independent viable haltereless
enhancer piracy line also supports the view that the dominant
loss of halteres results from ectopic rho expression. Thus, in
several cases, the pattern of localized rho transgene
expression in enhancer piracy lines correlates with the regions
that fail to differentiate normally, suggesting that the diverse
phenotypes recovered by this approach are indeed due to
localized mis-expression of rho.
DISCUSSION
The mechanism by which ectopic rho expression leads to the
variety of defects observed in the different enhancer piracy
lines remains to be determined. However, mutations in the
EGF-R/RAS signaling pathway modify several of these phenotypes, as would be expected if rho were functioning to
hyperactivate EGF-R signaling in each of these cases. These
results generalize previous findings (Rutledge et al., 1992;
Sturtevant et al., 1993; Ruohola-Baker et al., 1993) in which
rho was proposed to promote EGF-R signaling. The consistently strong suppression of dominant rho phenotypes by Star
also implicates Star as another generally required component
for facilitating EGF-R signaling. These dominant rho phenotypes, which are predictably modified by known mutations in
the EGF-R signaling pathway, are suitable for identifying new
loci interacting with rho (see below).
Dominant alleles are powerful genetic tools that can be used
for identifying additional genes functioning in a given
pathway. For example, in the case of rho, dominant extra wing
vein phenotypes are predictably modified in trans-heterozygous combinations with mutants in the EGF-R signaling
pathway (Sturtevant et al., 1993). These interactions are
opposite to interactions observed with loss of function rho
alleles. As mentioned above, this array of genetic interactions
strongly supports models in which rho functions to facilitate
or amplify EGF-R signaling. Several of the rho enhancer
piracy lines are currently being used to search for new second
site suppressors of dominant rho phenotypes. Comparing the
groups of interacting genes recovered in such screens based on
suppression of different dominant rho phenotypes (e.g. extra
wing veins versus dominant female sterility) should help dis-
tinguish genes acting at only one developmental stage from
those required in general for EGF-R signaling.
A striking feature of this screen was that many of the
dominant phenotypes recovered were not observed in a series
of heat induction experiments (Sturtevant et al., 1993;
Ruohola-Baker et al., 1993; J.W. O’Neill and E. Bier, unpublished data; M. Sturtevant and E. Bier, unpublished data).
These relatively rare dominant enhancer piracy phenotypes
might be difficult to recover by ubiquitous expression via heat
shock induction because ubiquitous expression at particular
stages is organismic lethal, or because the heat shock promoter
cannot provide high enough levels of target gene expression to
generate a phenotype. It is possible that some of the observed
phenotypes could be due to dominant inactivation of a gene
residing at the insertion site and not due to ectopic rho
expression. While we cannot rule out this possibility for each
individual case, it is highly unlikely that events of this kind
contribute significantly to the data in Table 1. First, dominant
phenotypes due to P-element insertion are exceedingly rare. In
screens of several thousand enhancer trap lines, only a small
handful of dominant visible phenotypes were recovered (Bier
et al., 1989). Furthermore, in nearly every case in this enhancer
piracy screen, we found several independent lines with a given
dominant phenotype. Since P-element insertion into the
‘hottest’ target sites is on the order of 1/400 and are more
typically in the range of 1/2,000 (Bier et al., 1989), it is very
unlikely that recovery of multiple examples of a given
dominant phenotype among 250 lines is due to independent
insertions into the same locus.
Two rare dominant alleles recovered in the enhancer piracy
screen provide insights into the developmental function of rho
that were not revealed from ubiquitous ectopic rho expression
(Sturtevant et al., 1993; M. Sturtevant and E. Bier, unpublished
data). The enhancer piracy line in which entire intervein sectors
are converted to vein demonstrates that rho is sufficient to
convert all intervein cells to the vein fate. A detailed analysis
of phenotypes resulting from staged heat inductions of an
inducible HS-rho line failed to reveal this potential since only
cells near forming veins or cells in particular stereotyped
intervein locations were highly susceptible to being transformed into a vein fate by ectopic rho expression (Sturtevant
et al., 1993; M. Sturtevant and E. Bier, unpublished data). As
the rho transgene is transiently expressed at high levels in
sharply localized patterns corresponding to the primordia of
affected wing sectors, this line underscores the value of
enhancer piracy in determining the range of consequences
resulting from tightly targeted mis-expression of rho. Thus,
although other gene(s) function in parallel with rho to mediate
the localized formation of veins (Sturtevant et al., 1993),
optimal mis-expression of rho is sufficient to convert all
intervein cells into veins. This suggests that when rho is
expressed in such a pattern, gene(s) mediating parallel vein
induction pathway(s) are either not required or are expressed
at functional levels in intervein territories as well as in vein
primordia.
The other novel class of phenotypes recovered is missing
halteres and or abnormal wing development (Fig. 2F-H). These
phenotypes derive from an early developmental defect during
embryogenesis since formation of haltere and wing disc
primordia is abnormal. It is interesting in this regard that
similar phenotypes have been observed in various combina-
Enhancer piracy and rhomboid function 2337
tions of pointed alleles (Scholz et al., 1993). It is unclear
whether these phenotypes result from loss or gain of pointed
function. Embryos lacking pointed function have defects in the
same structures requiring the activity of rho and other ventrolateral group genes (Mayer and Nüsslein-Volhard, 1988;
Klämbt, 1993). Thus, regulated activity of the ventrolateral
pathway may be required for the formation of wing and haltere
discs.
The fact that it is possible to recover enhancer piracy lines
with localized wing venation defects and that the rho transgene
is expressed in corresponding localized patterns in developing
wing primordia suggests that enhancer piracy can be used like
enhancer trapping to identify genes expressed in novel
localized patterns. For example, in the case of wing development, it should be efficient to search for enhancer elements
directing localized gene expression in wing discs using
enhancer piracy by screening for lines with localized ectopic
veins. This point is well illustrated by the pattern of rho
transgene expression (Fig. 5B,C) in wing discs of the enhancer
piracy shown in Fig. 2C. Further genetic analysis will be
required to determine whether this enhancer piracy insertion is
in or near a gene required for normal wing development.
There is good evidence that enhancer piracy will be useful
for generating dominant alleles of other genes of interest. In
independent studies using the same hs-CaSpeR vector
employed in this study, distinct position-dependent dominant
phenotypes have been observed for constructs carrying diverse
cDNAs for genes such as Suppressor of hairless Su(H)
(Schweisguth and Posakony, 1994), Hairless (H), Bearded
(Brd), and the HLH encoding genes, achaete (ac) and scute
(sc) (J.W. Posakony, personal communication). In the case of
HS-Su(H), dominant phenotypes behave genetically as would
be expected for hypomorphic alleles (Schweisguth and
Posakony, 1994). Other evidence, also suggests that similar
strategies for mis-expressing genes of interest should be
generally useful. For example, ectopic expression of an
activated RAS gene leads to an excess wing vein phenotype
very similar to that shown for rho in Fig. 2B (Brand and
Perrimon, 1993). Patterned mis-expression of the ubiquitously
expressed Serrate (Ser) gene also yields localized defects in a
subset of cells driving high levels of a Ser transgene (Speicher
et al., 1994). Finally, localized mis-expression of the
homeobox transcription factor encoded by the Antennapedia
gene resulting from a chromosomal rearrangement that brings
this transcription unit under the control of a head specific
promoter, generates a classic homeotic transformation of
antennae to legs (Schneuwly et al., 1987). Investigators
studying various other genes have also inadvertently recovered
dominant alleles resulting from constitutive activation of
cDNAs carried on heat-shock vectors, which in some cases
have served as important genetic tools (Van Vactor et al.,
1991). Thus, it is likely that the enhancer piracy method will
be useful for obtaining dominant alleles of many other genes,
especially those genes for which known phenotypes result from
heat shock expression. In principle, enhancer piracy is applicable to a wide variety of situations in which regulated
expression of the gene of interest is important for the development or survival of a dispensable region of the organism.
It is worth noting that enhancer piracy, while similar in
certain respects to the GAL4 mediated conditional expression
of target genes (Brand and Perrimon, 1993), offers comple-
mentary advantages for obtaining useful genetic tools. First,
only a single genetic element is required for enhancer piracy,
whereas the GAL4 system requires both a GAL4 producing
element and a responding UAS-cDNA construct. Second,
GAL4 activation is indirect and may not achieve the high
levels of expression derived from direct activation of the target
gene by powerful genomic enhancers. These intrinsic advantages and the relatively high frequency with which dominant
phenotypes are recovered suggest that enhancer piracy will
prove to be an effective general method for obtaining simple
dominant alleles. The two component GAL4 system also offers
important advantages as conditional ectopic expression of
genes in known patterns permits analysis of ectopic expression
leading to lethality, which cannot be studied using enhancer
piracy.
Dominant alleles arising from ectopic expression of a
transgene are also of significant value for recovering large
numbers of mutant alleles of that transgene, since reverting a
dominant phenotype requires only an F1 screen. For genes such
as rho, which do not share obvious primary sequence similarity with other genes of known function, the ability to generate
large numbers of mutant alleles of the transgene provides an
important tool for identifying critical functional domains of the
encoded protein product. Preliminary data, using EMS as a
mutagen, indicate that null revertants of a HS-rho dominant
extra vein phenotype can be recovered at a frequency of once
per 2,000-3,000 chromosomes (K. Howard and E. Bier, unpublished results). Since the mutagenized rho transgene can be distinguished from the endogenous rho gene by virtue of flanking
P-element vector sequences, it is possible to characterize only
the mutated copy of rho by using the polymerase chain reaction
(PCR). Primers synthesized to unique flanking vector
sequences can be used for PCR to amplify the entire rho coding
region (only 1 kb long), which can then be directly sequenced
to identify the mutant amino acid residue. This use of enhancer
piracy should be applicable to many other genes as protein
coding portions of cDNAs are frequently short enough to be
amplified in their entire length (< 3 kb) or in two overlapping
segments (< 6 kb).
Since transposable elements have been isolated from a broad
spectrum of organisms including mariner elements in diverse
insects (Robertson, 1993), Tc elements in C. elegans and other
nematodes (Abad et al., 1991), Ac/Ds elements in maize,
tobacco, tomatoes, and Arabidopsis (Walbot, 1992; Grevelding et al., 1992; Bancroft and Dean, 1993), and the endogenous Tag1 element in Arabidopsis (Tsay et al., 1993), it is
likely that the enhancer piracy approach could be generalized
for isolating dominant phenotypes for genes of interest in these
organisms. Given the increasing efficiency of generating other
transgenic organisms such as mice (Korn et al., 1992), fish
(Bayer and Campos-Ortega, 1992), and various plants (Walbot,
1992; Topping et al., 1991), similar strategies based on random
genomic insertion of cDNA constructs driven by basal
promoters may soon be feasible in these organisms as well.
We thank Jason O’Neill for help with whole-mount in situ hybridization; Dan Lindsley, James W. Posakony, Bob Schmidt, Martin
Yanofsky, Nigel Crawford, Hugo Bellen and Kathryn S. Burton for
helpful discussions and critical comments on the manuscript; James
W. Posakony for communicating unpublished results; and Kathryn S.
Burton for assembling the figures. This work was supported by NIH
2338 R. Noll and others
Grant no. RO1-NS29870-01, NSF Grant no. IBN-9318242 and
Research Grant no. 5-FY92-1175 from the March of Dimes Birth
Defects Foundation. E. B. was supported by funds from the McKnight
Neuroscience Foundation, Sloan Foundation, and an ACS Junior
Faculty Award.
REFERENCES
Abad, P., Quiles, C., Tares, S., Piotte, C., Castagnoe-Sereno, P., Abadon,
M. and Dalmasso, A. (1991). Sequences homologous to Tc(s) transposable
elements of Caenorhabitidis elegans are widely distributed in the phylum
Nematoda. J. Mol. Evol. 33, 251-258.
Alberga, A., Boulay, J.-L., Kempe, E., Dennefeld, C. and Haenlin, M.
(1991). The snail gene required for mesoderm formation in Drosophila is
expressed dynamically in derivatives of all three germ layers. Development
111, 983-992.
Ashburner, M., Thompson, P., Roote, J., Lasko, P. F., Grau, Y., El Messal,
M., Roth, S. and Simpson, P. (1990). The genetics of a small autosomal
region of Drosophila melanogaster containing the structural gene for alcohol
dehydrogenase. VII. Characterization of the region around the snail and
cactus loci. Genetics 126, 679-694.
Bancroft, I. and Dean, C. (1993). Transposition patter of the maize element Ds
in Arabidopsis thaliana. Genetics 134, 1221-1229.
Bang, A. G. and Posakony, J. W. (1992). The Drosophila gene Hairless
encodes a novel basic protein that controls alternative cell fates in adult
sensory organ development. Genes Dev. 6, 1752-1769.
Bayer, T. A. and Campos-Ortega, J. (1992). A transgene containing lacZ is
expressed in primary sensory neurons in zebrafish. Development 115, 421426.
Bellen, H. J, O’Kane, C. J., Wilson, C., Grossniklaus, U., Pearson, R. K.
and Gehring, W. J. (1989). P-element-mediated enhancer detection: a
versatile method to study development in Drosophila. Genes Dev. 3, 12881300.
Bier, E., Vaessin, H., Shepherd, S., Lee, K., McCall, K., Barbel, S.,
Ackerman, L., Carretto, R., Uemura, T., Grell, E., Jan, L. Y. and Jan, Y.
N. (1989). Searching for pattern and mutation in the Drosophila genome with
a P-lacZ vector. Genes Dev. 3, 1273-1287.
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.
Brand, A. H. and Perrimon, N. (1993). Targeted gene expression as a means
of altering cell fates and generating dominant phenotypes. Development 118,
401-415.
Clifford, R. J. and Schüpbach, T. (1992). The torpedo (DER) receptor
tyrosine kinase is required at multiple times during Drosophila
embryogenesis. Development 115, 853-872.
Freeman, M., Kimmel, B. E. and Rubin, G. M. (1992). Identifying targets of
the rough homeobox gene of Drosophila: evidence that rhomboid functions
in eye development. Development 116, 335-346.
Gotwals, P. and Fristrom, J. W. (1991). Three neighboring genes interact
with the Broad-complex and the Stubble-stubbleoid locus to affect imaginal
disc morphogenesis in Drosophila. Genetics 127, 747-759.
Grevelding, C., Becker, D., Kunze, R., von Menges, A., Fantes, V., Schell, J.
and Masterson, R. (1992). High rates of Ac/Ds germinal transposition in
Arabidopsis suitable for gene isolation by insertional mutatgenesis. Proc.
Nat. Acad. Sci. 89, 6085-6089.
Klämbt, C. (1993). The Drosophila gene pointed encodes two ETS-like
proteins which are involved in the development of the midline glial cells.
Development 117, 163-176.
Korn, R., Schoor, M., Neuhaus, H., Henseling, U., Soininen, R., Zachgo, J.
and Gossler, A. (1992). Enhancer trap integrations in mouse embryonic
stem cells give rise to staining patterns in chimeric embryos with a high
frequency and detect endogenous genes. Mech Dev. 39, 95-109.
Lindsley, D. L. and Grell, E. H. (1968). Genetic variations in Drosophila
melanogaster. Carnegie Institute of Washington, Washington, D. C.
Lindsley, D. L. and Zimm, G. G. (1992). The Genome of Drosophila
melanogaster. San Diego, CA: Academic Press Inc.
Maniatis, T., Fritsch, E. F. and Sambrook, J. (1982). Molecular Cloning: A
Laboratory Manual. Cold Spring Harbor, New York: Cold Spring Harbor
Laboratory Press.
Mayer, U. and Nüsslein-Volhard, C. (1988). A group of genes required for
pattern formation in the ventral ectoderm of the Drosophila embryo. Genes
Dev. 2, 1496-1511.
O’Kane, C. and Gehring, W. (1987). Detection in situ of genomic regulatory
elements in Drosophila. Proc. Nat. Acad. Sci. USA 84, 9123-9127.
Raz, E. and Shilo. B. Z. (1992). Dissection of the faint little ball (FLB)
phenotype – determination of the development of the Drosophila central
nervous system by early interactions in the Ectoderm. Development 114,
113-123.
Raz, E. and Shilo, B. Z. (1993). Establishment of ventral cell fates in the
Drosophila ectoderm requires DER, the EGF receptor homolog. Genes Dev.
7, 1937-1948.
Roberts, D. B. (1986). Basic Drosophila care and techniques. In Drosophila: A
Practical Approach (ed. D. B. Roberts), pp. 1-38. Washington, D. C., IRL
Press.
Robertson, H. M. (1993). The mariner transposable element is widespread in
insects. Nature 362, 241-245.
Ruohola-Baker, H., Grell, E., Chou, T. B., Baker, D., Jan, L. Y. and Jan, Y.
N. (1993). Spatially localized rhomboid is required for establishment of the
dorsal-ventral axis in Drosophila oogenesis. Cell 73, 953-965.
Rutledge, B., Zhang, K., Bier, E., Jan, Y. N. and Perrimon, N. (1992). The
Drosophila spitz gene encodes a putative EGF-like growth factor involved in
dorsal-ventral axis formation and neurogenesis. Genes Dev. 6, 1503-1517.
Schneuwly, S., Klemenz, R. and Gehring, W. (1987). Molecular analysis of
the dominant homeotic Antennapedia phenotype. EMBO J. 6, 201-206.
Scholz, H., Deatrick, J., Klaes, A. and Klampt, C. (1993). Genetic dissection
of pointed, a Drosophila gene encoding two ETS-related proteins. Genetics
135, 455-468.
Schweisguth, F. and Posakony, J. W. (1994). Antagonistic activities of
Suppressor of Hairless and hairless control alternative cell fates in the
Drosophila adult epidermis. Development 120, 1433-1441.
Speicher, S. A., Thomas, U., Hinz, U. and Knust, E. (1994). The Serrate
locus of Drosophila and its role in morphogenesis of the wing imaginal discs:
control of cell proliferation. Development 120, 535-544.
Sturtevant, M. A., Roark, M. and Bier, E. (1993). The Drosophila rhomboid
gene mediates the localized formation of wing veins and interacts genetically
with components of the EGF-R signaling pathway. Genes Dev. 7, 961-973.
Tautz, D. and Pfeiffle, C. (1989). A non-radioactive in situ hybridization
method for the localization of specific RNAs in Drosophila reveals
translational control of the segmentation gene hunchback. Chromosoma 98,
81-85.
Topping, J. F., Wei, W. and Lindsey, K. (1991). Functional tagging of
regulatory elements in the plant genome. Development 112, 1009-1019.
Tsay, Y.-F., Frank, M. J., Page, T., Dean, C. and Crawford, N. M. (1993).
Identification of a mobile endogenous transposon in Arabidopsis thaliana.
Science 260, 342-344.
Van Vactor, D. L., Cagan, R. L., Kramer, H. and Zipursky, S. L. (1991).
Induction in the developing compound eye of Drosophila: multiple
mechanisms restrict R7 induction to a single retinal precursor cell. Cell 67,
1145-1155.
Walbot, V. (1992). Strategies for mutagenesis and gene cloning using
transposon tagging and T-DNA insertional mutatgenesis. Ann. Rev. Plant
Phys. Plant Mol. Biol. 43, 49-89.
Wilson, C., Pearson, R. K., Bellen, H. J., O’Kane, C. J., Grossniklaus, U.
and Gehring, W. (1989). P-element-mediated enhancer detection: an
efficient method for isolating and characterizing developmentally regulated
genes in Drosophila. Genes Dev. 3, 1301-1313.
(Accepted 17 May 1994)