The Drosophila wing is an apt venue in which to study genetic
and cellular interactions that establish coordinate axes and
allocate the predominant cell fates, vein and intervein, within
the adult wing blade. Wing vein morphogenesis in Drosophila
melanogaster was originally described by Waddington (1940).
Wing veins develop during an intricate metamorphic process
that converts a folded, monolayer wing disc into a flat, bilayered
adult wing. These processes have been described at the cellular
level by Fristrom et al. (1993). The results of numerous mutant
and genetic mosaic analyses imply that patterns and size in wing
development are controlled by cell interactions and not by cell
lineage (Diaz-Benjumea and García-Bellido, 1990; García-
Bellido and de Celis, 1992; de Celis and García-Bellido,
1994b). Sturtevant and Bier (1995) have proposed a sequential
model for the action of genes central to wing vein establishment
and differentiation. Subsequent analyses of wing vein develop-
ment have revealed that function and cross-regulation of at least
three signal transduction pathways – Delta/Notch, dpp/TKV
and DER – are required for differentiation of the correct number
of vein cells (de Celis, 1997).
The Delta (Dl) locus was first identified on the basis of the
thickenings of the distal termini of longitudinal wing veins
(‘deltas’) in flies heterozygous for Dl loss-of-function and wild-
type alleles (Dexter, 1914). Dl or Notch (N) loss-of-function
genotypes result in vein thickening of varying severity (Lindsley
and Zimm, 1992) and N gain-of-function genotypes can lead to
vein shortening (Foster, 1975; Portin, 1975). These contrasting
phenotypes imply that neurogenic signalling may function during
two stages of wing vein development. During the first stage, neu-
rogenic signalling would be required to define the provein anlage,
i.e., a broad stripe of ‘vein-competent’ cells more numerous than
those in the adult vein, that extends bilaterally beyond the adult
vein boundaries. Excessive neurogenic signalling during this
stage, associated with N gain-of-function Abruptex (NAx) alleles
(Palka et al., 1990; Heitzler and Simpson, 1993; de Celis and
García-Bellido, 1994a), would reduce the extent of the anlage and
thereby lead to development of shortened wing veins. During the
second stage, Delta and Notch would delineate vein and intervein
cell fates within the provein anlage. Reduced neurogenic sig-
nalling during this stage, associated with reductions in Dl or N
function, would lead to overcommitment to the vein cell fate
within the anlage and the development of thickened veins.
Analysis of vein mutants performed by Diaz-Benjumea and
García-Bellido (1990) supports the premise that veins are deter-
mined during the proliferative phase of larval wing disc devel-
opment and differentiate during pupal development. The fact that
clones of Dl or N mutant tissue develop as vein only when
adjacent to ‘normal’ vein tissue implies that there exists a ‘vein-
competent’ domain within which veins normally form (García-
Bellido and de Celis, 1992; de Celis and García-Bellido, 1994b).
Neurogenic signalling is central to the partitioning of cell
fates within equivalence groups in many contexts during
embryonic and postembryonic Drosophila
(Muskavitch, 1994; Artavanis-Tsakonas et al., 1995). Delta
functions as a signal in this process, as reflected in the cell non-
autonomous behavior of Delta in somatic mosaic analyses in
Development 124, 3283-3291 (1997)
Printed in Great Britain © The Company of Biologists Limited 1997
Delta and Notch are required for partitioning of vein and
intervein cell fates within the provein during Drosophila
metamorphosis. We find that partitioning of these fates is
dependent on Delta-mediated signalling from 22 to 30
hours after puparium formation at 25°C. Within the
provein, Delta is expressed more highly in central provein
cells (presumptive vein cells) and Notch is expressed more
highly in lateral provein cells (presumptive intervein cells).
Accumulation of Notch in presumptive intervein cells is
dependent on Delta signalling activity in presumptive vein
cells and constitutive Notch receptor activity represses
Delta accumulation in presumptive vein cells. When Delta
protein expression is elevated ectopically in presumptive
intervein cells, complementary Delta and Notch expression
patterns in provein cells are reversed, and vein loss occurs
because central provein cells are unable to stably adopt the
vein cell fate. Our findings imply that Delta-Notch sig-
nalling exerts feedback regulation on Delta and Notch
expression during metamorphic wing vein development,
and that the resultant asymmetries in Delta and Notch
expression underlie the proper specification of vein and
intervein cell fates within the provein.
Key words: Delta, Notch, Drosophila, neurogenic gene, wing vein
Feedback regulation is central to Delta-Notch signalling required for
Drosophila wing vein morphogenesis
Stacey S. Huppert, Thomas L. Jacobsen and Marc A. T. Muskavitch
Program in Genetics, Cell and Developmental Biology, Department of Biology, Indiana University, Bloomington, Indiana, 47405,
the notum (Heitzler and Simpson, 1991) and wing blade
(García-Bellido and de Celis, 1992). Notch functions as a
receptor, as shown by its cell-autonomous function in embryos
(Hoppe and Greenspan, 1986, 1990), nota (Heitzler and
Simpson, 1991) and wing blades (García-Bellido and de Celis,
1992). The fact that the extracellular domain of the Delta
protein is capable of interacting with the Notch protein on
opposing cell surfaces implies that Delta is a ligand for the
Notch receptor (Fehon et al., 1990). The similar phenotypes
observed in Dl and N loss-of-function mutants imply that Delta
activates the Notch receptor (Shellenbarger and Mohler, 1978;
Lehmann et al., 1983; Parody and Muskavitch, 1993).
In this paper, we investigate the functions of neurogenic sig-
nalling during pupal wing vein formation. We find that neuro-
genic signalling is required between 22 and 30 hours after
puparium formation (APF) for specification of the proper
number of vein cells, i.e., for the partitioning of vein and
intervein fates within the provein. Delta and Notch protein
expression patterns are generally complementary during wing
vein development. This observation, and the impacts of ectopic
Delta expression on vein development that we observe, imply
that asymmetries in Delta and Notch protein expression among
adjacent cells are essential for the correct partitioning of cell
fates within the provein. We show that reduction in Delta
expression or ubiquitous expression of an activated Notch
receptor in the pupal wing blade alters Delta and Notch protein
expression, reflecting the existence of a feedback mechanism
by which neurogenic signalling regulates Delta and Notch
expression in developing wing veins.
MATERIALS AND METHODS
Oregon-R (Stanford isolate) was used as a wild-type control in all
analyses. P[lArB]l(3)A326.2F3, ry506/TM6C, cu Sb e Tb ca (i.e.,
DlPlacZ/TM6C) was obtained from the Bloomington Drosophila Stock
Center (Bloomington, IN). This enhancer trap was previously shown
to exhibit expression patterns in embryos and ovaries appropriate for
the Dl locus (Wilson et al., 1989). The DlPlacZinsertion was re-
mobilized to identify local transpositions that might affect Dl regula-
tory regions (S. S. Huppert and M. A. T. Muskavitch, unpublished
data). DlEWis a transposon-induced mutation isolated during the re-
mobilization of DlPlacZ. DlRFis a heat-sensitive allele (Parody and
Muskavitch, 1993). P[ry+ hsp70-Notch(intra)] (i.e., P[hs-N(intra)])
was a generous gift from Gary Struhl (Columbia University; New
York, NY) (Struhl et al., 1993). This strain was used to examine the
impact of constitutive neurogenic signalling on wing vein develop-
ment. The 1348::GAL4 driver, isolated by G. Technau (Universität
Mainz; Mainz, Germany), was used to induce expression of wild-type
Delta (using a UAS::DeltaWT responder construct; T. L. Jacobsen and
M. A. T. Muskavitch, unpublished data) and a dominant-negative form
of Delta (using a UAS::DeltaD responder construct; T. L. Jacobsen
and M. A. T. Muskavitch, unpublished data). The expression pattern
supported by 1348::GAL4 driver was assessed using a nuclear-
targeted β-galactosidase responder construct (i.e., UAS::βGalNUC; T.
L. Jacobsen and M. A. T. Muskavitch, unpublished data).
Phenocritical period analysis
DlRF/DlPlacZanimals were generated by crossing DlPlacZ/TM6C adults
to DlRF/DlRFadults. Tubby+white prepupa growing at 18°C were
collected and kept in a humid chamber at 18°C, then shifted to 32°
for 8 hours at various times after puparium formation (APF). P[hs-
N(intra)] animals were grown at 25°C. White prepupae were collected
and kept in a humid chamber at 25°C, then shifted to 37°C for 2 hours
at various times during development. Phenotypes of adult wings were
Pupae were staged from white prepupa formation at 25°C. Wing discs
younger than 6 hours APF were dissected in PBS (3 mM NaH2PO4,
7 mM Na2HPO4, 131 mM NaCl, pH 7.5) and fixed for 25 minutes in
4% paraformaldehyde in PBS. For wing discs aged more than 6 hours
APF, staged pupae were removed from pupal cases in PBS and then
incubated in 4% paraformaldehyde in PBS for no less than 12 hours
at 4°C. Wing discs were then pulled from the pupal carcass and
cuticles were removed.
Dissected discs were incubated with primary antibody at 1:300
(anti-β-galactosidase; Promega; Madison, WI), 1:10,000 (anti-Delta,
MAb202 ascites; Parks et al., 1995), or 1:15,000 (anti-Notch,
C17.9C6 purified ascites; Fehon et al., 1990) overnight at 4°C in
TPBS (0.3% Triton X-100 in PBS) plus 5% normal goat serum
(NGS). Discs were then incubated at room temperature for 2 to 6
hours with goat anti-mouse IgG conjugated to horseradish peroxidase
(Jackson Immunochemicals; West Grove, PA) diluted 1:800 in TPBS
plus 5% NGS. For increased sensitivity, silver was precipitated onto
the peroxidase reaction product (Gallyas et al., 1982; Liposits et al.,
1984). Discs were mounted in glycerol and viewed using a Zeiss
Axioskop light microscope.
Pupae were staged and wings were fixed and dissected as above.
Dissected discs were then incubated with primary antibody at 1:16,000
(anti-Notch, C17.9C6 purified ascites; Fehon et al., 1990) and 1:4,000
(anti-Delta, GP581) overnight at 4°C in TPBS plus 5% NGS. Discs
were next incubated overnight at 4°C with goat anti-mouse IgG con-
jugated to fluorescein isothiocyanate and goat anti-guinea pig IgG con-
jugated to Texas Red (Jackson Immunochemicals; West Grove, PA),
each diluted 1:100 in TPBS plus 5% NGS. Discs were then mounted
in glycerol plus 1% n-propyl gallate. The tissue was viewed with a Bio-
Rad MRC600 confocal laser microscope. The guinea pig polyclonal
anti-serum GP581 was prepared as described in Fehon et al. (1990).
The 0.54 kb ClaI DNA fragment encoding Delta EGF-like repeats 4-
9 (amino acids 350-529) was excised from the Dl1 cDNA (Kopczyn-
ski et al., 1988) and transferred into a β-galactosidase vector to create
a fusion protein that was used to immunize guinea pigs (Pocono Rabbit
Farm & Laboratory; Canadensis, PA).
Propidium iodide staining was performed as in Li and Kaufman
Wings were removed from adults and mounted under coverslips in
Gary’s Magic Mountant (Ashburner, 1989).
Phenocritical periods for neurogenic signalling
during pupal wing vein specification
The conditional genotype DlRF/DlPlacZwas employed to
analyze effects of the reduction of Delta function (i.e., reduction
of neurogenic signalling) on wing vein formation during larval
and pupal development. Animals bearing this genotype exhibit
heat-sensitive increases in the severity of Dl mutant wing
venation phenotypes (data not shown). At the permissive tem-
perature of 18°C, the DlRF/DlPlacZgenotype yields wings with
mild ‘deltas’ where longitudinal veins intersect the margin (Fig.
1A). DlRF/DlPlacZanimals pulsed for 8 hours beginning at times
ranging from 40 to 60 hours APF at 18°C, equivalent to 20 to
30 hours APF at 25°C, display thickened veins and marginal
‘deltas’ larger than those of animals raised at 18°C. The most
severe impact on wing venation of this conditional Dl mutant
S. S. Huppert and others
3285 Delta-Notch feedback regulation
genotype results from a pulse applied from 40 to 48 hours APF
(Fig. 1B). We propose that this vein thickening phenotype
occurs because reduction in Dl function leads to reduction of
neurogenic signalling within the provein. As a result, fewer cells
are inhibited from adopting the vein cell fate than in wild-type
animals and veins become thicker. An extreme example of the
failure to inhibit cells within the provein from adopting the vein
cell fate is evident in the wings of DlEW/DlEWanimals (Fig. 1G).
In order to analyze the impacts of constitutive neurogenic
signalling during wing vein formation, we induced ubiquitous
expression of a Notch variant that yields N gain-of-function
phenotypes (Lieber et al., 1993; Rebay et al., 1993; Struhl et
al., 1993). The P[hs-N(intra)] transgene encodes a heat-shock-
regulated Notch variant, Notch(intra), constituted solely of the
Notch receptor intracellular domain (Struhl et al., 1993). When
this variant is expressed during embryonic neuroblast specifi-
cation, adoption of neuroblast fate is inhibited excessively and
almost all ventral ectodermal cells give rise to epidermis
(Struhl et al., 1993). At 25°C, Notch(intra) has no effect on
vein formation and the adult wing is wild type in character
(Fig. 1C). However, when Notch(intra) expression is induced
by 2 hour heat pulses at 37°C initiated at times
from 22 to 32 hours APF, vein loss is observed
in the adult wing. The most extensive vein loss
that we observe results from a heat pulse from
26 to 28 hours APF (Fig. 1D). Animals pulsed
in this manner exhibit shortening of L2, L4 and
L5, and some shortening of L3. We attribute
phenotype to excessive inhibition of adoption
of the vein cell fate within the provein.
The interval during which neurogenic sig-
nalling appears to be required for partitioning
of fates within the provein is correlated with
the second phase of cell layer apposition
observed for intervein cells (Fristrom et al.,
1993). During this time, the wing veins
become apparent progressively. Cells on either
side of the ‘intervein bands’ begin to extend
basal processes and the vein cells are the only
regions remaining unapposed (Fristrom et al.,
1993, 1994). Neurogenic signalling apparently
blocks the responses of lateral provein cells to
inductive cues that specify the vein cell fate
and these lateral cells then extend basal
processes during the later stages of intervein
Delta and Notch protein expression in
larval and pupal wing discs
Delta and Notch exhibit complex and dynamic
expression patterns that are correlated with the
numerous embryonic and imaginal defects
observed in Dl and N mutants (Fehon et al.,
1991; Kooh et al., 1993). We have examined
Delta and Notch protein expression patterns
throughout wing vein development in order to
expression patterns and the timing of neuro-
In third larval instar wing discs, Delta is
expressed in most cells within the wing pouch, although higher
expression is associated with two stripes of cells along the
prospective wing margin, and with regions wherein longitudinal
wing veins L3, L4 and L5 will develop (Kooh et al., 1993) (Fig.
2A). Notch protein accumulates on cell surfaces and is expressed
throughout the wing pouch (Fig. 2B), although expression is
lower in cells in which Delta is expressed (Fehon et al., 1991;
Kooh et al., 1993). At 6 hours APF at 25°C, when the wing disc
has everted and the dorsal and ventral surfaces of the wing pouch
have become apposed, the Delta expression pattern is similar to
that observed in third instar larvae (Fig. 2C). The Notch
expression pattern becomes refined such that Notch protein levels
are higher in wing blade cells adjacent to Delta-expressing cells
(Fig. 2D). Delta protein expression does not appear in the region
within which L2 will develop until approximately 22 hours APF
at 25°C (data not shown). By 30 hours APF at 25°C, the devel-
oping wing blade has the general shape of the adult wing and the
vein pattern is pre-figured by Delta expression. Delta is highly
expressed in all vein cells and is concentrated within subcellular
vesicles at this time (Fig. 2E). This protein expression pattern is
consistent with the report of de Celis (1997), indicating that Dl
Fig. 1. Phenotypic effects associated with perturbations in neurogenic signalling during
pupal wing vein development. (A,B) Adult wing phenotypes of DlPlacZ/DlRFanimals
reared at permissive temperature, 18°C (A), or pulsed to 32°C from 40 to 48 hours APF
(B). (C,D) Phenotypes of P[hs-N(intra)] animals grown at 25°C (C), or grown at 25°C
and heat-pulsed to 37°C from 26 to 28 hours APF (D). (E) Vein loss phenotype in a
UAS::DeltaWT/+; 1348::GAL4/+ wing from an animal grown at 25°C. (F) Vein
thickening phenotype in a 1348::GAL4/+; UAS::DeltaD/+ wing from an animal grown
at 25°C. (G) Extreme vein thickening phenotype of DlEW/DlEWanimals grown at 25°C.
transcription is limited to vein territories in the pupal wing. Notch
is more highly expressed in the cells (i.e., lateral provein cells;
Fig. 2F) that flank Delta-expressing cells (i.e., vein cells), and
remains on the surfaces of lateral provein cells.
Simultaneous staining for Delta and Notch proteins reveals
that Delta and Notch protein expression patterns are comple-
mentary, on a cell-by-cell basis, in pupal wing discs (Fig. 3).
Delta is visible on the surfaces of vein cells at 24 hours APF
(Fig. 3A) and accumulates in subcellular vesicles within these
cells at 30 hours APF (Fig. 3B). Notch expression is higher in
the intervein cells that immediately flank the Delta-expressing
vein cells than in other intervein cells. These patterns become
apparent by 24 hours APF (Fig. 3A) and are even more evident
by 30 hours APF (Fig. 3B). This implies that Delta is expressed
maximally in wing vein cells during the partitioning of fates
within the provein (22 to 30 hours APF) and that Notch
expression during the same period is reduced in central provein
cells and elevated in lateral provein cells that will adopt the
intervein fate. We have assessed the number of provein cells
that express Delta and Notch by using propidium iodide to
mark cell nuclei (data not shown). We find that Delta is highly
expressed within proveins in stripes three cells in width in wing
discs from wandering third instar larvae, prepupae at 6 hours
APF, and pupae at 24 hours and 30 hours APF. Notch protein
levels within proveins are lowest in regions that are three cells
in width throughout wing vein development. However, the
initially broad regions of cells that express high levels of Notch
in the lateral proveins narrow to stripes two to three cells in
width by 30 hours APF. Thus, the distributions of cells with
high levels of Delta and Notch proteins appear to be comple-
mentary throughout metamorphic vein development.
Dynamic changes in the subcellular localization of Delta are
also apparent during larval and pupal wing vein development.
Delta expression in the wing pouch first becomes apparent
S. S. Huppert and others
Fig. 2. Delta and Notch protein expression in
wild-type wing discs. Whole-mount wing discs
staged at 25°C and stained with an antibody
against either Delta or Notch, as described in
Materials and Methods. Discs from third instar
larvae stained for Delta (A) or Notch (B). L,
wing margin; ∆, longitudinal wing veins L3, L4
and L5. (Inset A) The intersection of the dorsal
wing margin stripe (L) and the L5 provein (∆).
Discs from 6 hour APF pupae stained for Delta
(C) or Notch (D). L3, L4 and L5 marked as in
panel A. (Inset C) From a region in the L5
provein. Discs from 30 hour APF pupae stained
for Delta (E) or Notch (F). L, L2, L3, L4 and
L5. Inset E, is from the region of distal L5.
Fig. 3. Delta and Notch protein expression in pupal wing discs. Discs
stained for Delta and Notch, as described in Materials and Methods.
Delta signal, red; Notch signal, green. (A) 24 hour APF disc with
presumptive L3 and L4 vein cells marked (∆). (B) 30 hour APF disc
with presumptive L4 and L5 vein cells marked (∆).
3287 Delta-Notch feedback regulation
during the mid-second larval instar. Doherty et al. (1996) report
that Delta accumulates in the membranes of cells at the
dorsal/ventral boundary during the second larval instar,
primarily in cells within the ventral compartment. Subcellular
localization of Delta in the developing wing
blade varies dynamically during the third
larval instar and early pupal life, presumably
as the result of successive cycles of cell surface
targeting and endocytic down-regulation
(Kooh et al., 1993; Parks et al., 1995). During
the late third larval instar, we find that Delta is
localized in subcellular vesicles and is also
accumulating on cell surfaces (Fig. 2A inset).
Next, Delta is predominantly vesicular
between 2 and 3 hours APF (data not shown).
Around 6 hours APF, the majority of Delta
protein is again found on cell surfaces (Fig. 2C
inset). Still later, by 30 hours APF, the majority
of Delta protein in vein cells has been down-
regulated into subcellular vesicles (Fig. 2E
inset). This latter phase of Delta protein down-
regulation appears to be correlated with the
end of Delta-dependent signalling within the
provein, and completion of the partitioning of
provein cells between vein and intervein fates.
Neurogenic signalling affects levels
and patterns of Delta and Notch
In order to investigate regulatory mechanisms
that bear on the Delta-Notch signalling
pathway during wing vein morphogenesis, we
have analyzed the effects on Delta and Notch
protein expression of constitutive Notch
receptor activity and reduced Delta signal
activity. This analysis focused on the pheno-
critical period during which we find that neu-
rogenic signalling is required for final parti-
tioning of vein and intervein cell fates within
We employed the P[hs-N(intra)] genotype to investigate the
impacts of constitutive neurogenic signalling on Delta protein
expression. Constitutive Notch receptor activity was induced
by the pupal heat pulse regime that we had found resulted in
Fig. 4. Delta and Notch protein
expression associated with altered
neurogenic signalling in pupal wing
discs. Discs were staged at 25°C and
dissected at 30 hours APF. (A,C,E)
Delta protein; (B,D,F) Notch protein.
(A-D) Images of the region in which
longitudinal veins L4 and L5 (∆ or
L) and the posterior crossvein
intersect. (E,F) Images of the distal
portions of L3 (L) and L4. (A,B)
Wild-type discs; (C,D) P[hs-
N(intra)]/P[hs-N(intra)] discs from
animals grown at 25°C and heat-
pulsed to 37°C from 26 to 28 hours
APF, and allowed to recover 2 hours
at 25°C before dissecting and
staining. (E,F) DlEW/DlEWdiscs from
animals grown at 25°C.
Fig. 5. Effects of ectopic Delta expression on Delta and Notch expression in 30 hours
APF wing discs. Animals grown at 25°C. (A,C,E) Delta protein; (B,D,F) Notch protein.
(A,B) Discs from wild-type animals with longitudinal veins L4 and L5 marked (∆ or
L). (C,D) Discs from UAS::DeltaWT/+; 1348::GAL4/+ animals. (C) Region of L4
marked (L) where Delta is highest in the lateral provein cells. (D) Longitudinal veins
L4 and L5 marked (∆) where Notch accumulates throughout the provein. (E,F) Discs
from 1348::GAL4/+; UAS::DeltaD/+ animals with longitudinal veins L4 and L5
marked (∆ or L).
the most extensive vein loss in P[hs-N(intra)] animals: a heat
pulse at 37°C from 26 to 28 hours APF (Fig. 1D). Notch
receptor activation during this interval leads to drastic reduc-
tions in Delta expression (Fig. 4C) in comparison to wild-type
Delta expression levels (Fig. 4A). We also observe expression
and nuclear localization of the Notch intracellular domain
fragment, encoded by the P[hs-N(intra)] transposon, through-
out the wing blade (Fig. 4D), although it is unclear whether
this nuclear localization is relevant to the N gain-of-function
phenotype that we observe in the wing. These observations
imply that Delta expression is negatively regulated by Notch
receptor activity within the pupal wing pouch.
The effects of reductions in Delta signalling activity were
examined in animals homozygous for the DlEWallele. This
allele appears to be a regulatory mutation that disrupts Dl
function during wing and eye development, but supports normal
Dl function in other imaginal contexts (data not shown). Adults
homozygous for the DlEWallele exhibit severe thickening of all
veins (Fig. 1G), which appears to arise because all cells within
the provein adopt the vein cell fate. Delta protein expression is
severely reduced in DlEWmutant wing blades at 30 hours APF
(Fig. 4E) compared to wild-type wing blades (Fig. 4A). Notch
expression in lateral proveins fails to increase above levels
observed in intervein regions in DlEWwing blades (Fig. 4F
compared to Fig. 4B), implying that wild-type levels of Delta
signalling are required for elevated levels of Notch expression
in lateral provein cells in the pupal wing blade.
Asymmetries in Delta-Notch signalling are required
for vein cell specification
To evaluate the functional significance of complementary
patterns of Delta and Notch protein expression during wing
vein development, we employed the GAL4-UAS driver-
responder system (Brand and Perrimon, 1993) to generate
ectopic expression of wild-type and mutated Delta proteins in
the wing blade, using the 1348::GAL4 driver. When
UAS::DeltaWT/+; 1348::GAL4/+ animals are raised at 25°C,
vein loss is observed in adult wings (Fig. 1E). We observe
variable reductions of veins L2, L4 and L5. These data support
the hypothesis that increased levels of Delta-mediated sig-
nalling preclude adoption of the vein cell fate. Expression of a
dominant-negative form of Delta (DeltaD; T. R. Parody and M.
A. T. Muskavitch, unpublished data) under control of the
1348::GAL4 driver causes vein thickening (1348::GAL4/+;
UAS::DeltaD/+; Fig. 1F). All veins seem to be affected by the
ectopic expression of this dominant-negative Delta variant. We
infer that reduced levels of Delta signalling activity lead to an
increase in the number of provein cells that adopt the vein cell
fate, at the expense of intervein cells.
Given the dramatic effects of 1348::GAL4-mediated Delta
expression on wing vein morphogenesis, we characterized the
1348::GAL4 driver-mediated expression pattern (data not
shown) in a wild-type genetic background using a responder
construct that encodes a nuclear-targeted form of β-galactosidase
(i.e., UAS::βGalNUC; see Materials and Methods). Initial
patchy 1348::GAL4-mediated expression is evident throughout
the wing blade circa 12 hours APF. Expression begins to be
limited to intervein areas circa 18 hours APF and is restricted to
interveins by 30 hours APF (see also de Celis, 1997). Initial
intervein-restricted expression evident circa 18 hours APF
appears highest in regions corresponding to ‘intervein bands’
(Fristrom et al., 1994), and expression becomes more evenly dis-
tributed among intervein cells as pupal development proceeds.
We also examined the impacts of ectopic 1348::GAL4-
mediated Delta signalling on Delta and Notch expression during
pupal vein development. Elevated intervein expression of Delta
in UAS::DeltaWT/+; 1348::GAL4/+ wing blades (Fig. 5C)
leads to higher levels of Notch expression in intervein cells 30
hours APF (Fig. 5D) in comparison to wild-type wing blades
(Fig. 5B). This suggests that Delta-mediated signalling
increases expression of the Notch receptor in pupal wing blade
cells. We also find that this ectopic Delta signalling leads to
elevated Delta expression in two stripes of cells – the lateral
provein cells – (Fig. 5C, note the arrowheads) that flank the
region within which maximal Delta expression occurs in wild-
type wing blades at 30 hours APF (Fig. 5A). This, in turn,
causes Notch expression within presumptive vein cells to
increase (Fig. 5D, note the arrowheads where Notch accumu-
lates to uniformly high levels across the provein region),
compared to that observed in wild-type wing blades (Fig. 5B).
It therefore appears that Delta expression in intervein cells leads
S. S. Huppert and others
Fig. 6. A model for Delta-mediated signalling in the provein.
(A,B) Wild-type situation; (C,D) the situation in which Delta is
driven ectopically in intervein (IV) cells by 1348::GAL4. (A,C) Delta
protein expression; (B,D) Notch protein expression. (A) Delta
protein levels are maximal in the central provein cells (CPV); very
low levels are detected in the lateral provein cells (LPV). (B) Notch
expression is complementary to that of Delta; maximal protein levels
are observed in the LPV and the lowest levels are detected in the
CPV. (C) When Delta is expressed under the control of 1348::GAL4,
Delta accumulates to higher levels in the LPV (presumptive intervein
cells) than in the CPV (presumptive vein cells, V). (D) This leads to
a higher level of Notch expression in the CPV compared to that in
wild-type animals (B). Elevated Notch expression in CPV cells leads
to reception of a neurogenic signal by those cells. This signal inhibits
them from stably adopting the vein cell fate. High levels of Notch are
also observed in the intervein, where 1348::GAL4 mediates ectopic
Delta expression. Note: shadings of circles (i.e., cells) represent
levels of protein expression relative to those in adjacent cells. More
darkly shaded cells express higher levels of a given protein than more
lightly shaded cells.
3289 Delta-Notch feedback regulation
to the reconfiguration of Delta and Notch expression patterns
within and around the provein such that neurogenic signalling
can lead to inhibition of adoption of the vein cell fate through-
out the provein. Ectopic expression of a dominant-negative
form of Delta in 1348::GAL4/+; UAS::DeltaD/+ wing blades
(Fig. 5E) leads to reduced Notch expression within the lateral
provein cells, such that Notch accumulates at uniformly low
levels across the provein (Fig. 5F). Therefore, reduced neuro-
genic signal reception in lateral provein cells results in a failure
to elevate Notch expression in lateral provein cells and a failure
to inhibit adoption of the vein cell fate within the lateral provein.
The roles of neurogenic signalling in wing vein
Our phenocritical period analysis implies that Delta-mediated
neurogenic signalling is required for the final resolution of cell
fates within the provein between 22 and 30 hours APF at 25°C.
This correlates with a temperature-sensitive period for wing vein
development defined by Shellenbarger and Mohler (1978) using
the l(1)Nts1mutation. Rebay et al. (1993) also observed wing
vein thickening during this time, after inducing expression of a
membrane-associated Notch variant that lacks the extracellular
domain. This vein-thickening phenotype is similar to the so-
called ‘Confluens’ phenotype, which results from hyperploidy
for the wild-type N gene (Welshons, 1965) and probably reflects
titration of factors limiting for Delta-Notch signalling (Muska-
vitch, 1994). In contrast, expression of the constitutively active
Notch(intra) variant produces the opposite phenotype, wing vein
loss, as would be expected if constitutive Notch signalling
inhibits adoption of the vein cell fate throughout the provein. We
also note that a comparable, but earlier, induction of wing vein
loss was observed by Lyman and Yedvobnick (1995) following
expression of a different constitutively active Notch intracellu-
lar domain variant (Lieber et al., 1993) from 16 to 24 hours APF.
Sturtevant and Bier (1995) also found that reductions in Dl or N
function during the early pupal stage led to broadening of
expression domains for rhomboid, a marker for wing veins, in
the wing blade. These results, in aggregate, indicate that Delta
signalling and Notch receptor activity function during the same
interval during pupal wing vein development and imply that
Delta acts as the signal for the Notch receptor during the neu-
rogenic signalling process required to limit the number of vein
cells that are specified among the ‘vein-competent’ cell popula-
tion within the provein. These phenocritical periods correlate
with the developmental interval during which wing veins are
progressively emerging, as cells determined to become intervein
cells become apposed by extending basal processes (Fristrom et
al., 1993). Thus, Delta-Notch-mediated neurogenic signalling
functions to partition cells within the provein between vein and
intervein cell fates, presumably by inhibiting the lateral provein
cells from adopting the vein cell fate until the proper number of
intervein cells have been specified.
A feedback mechanism regulates levels of Delta and
Notch protein expression in cells involved in
Heitzler and Simpson (1991) proposed that Delta and Notch
expression could be subject to feedback regulation by Delta-
Notch signalling, leading to complementary patterns of Delta
and Notch expression, based on somatic mosaic analysis of
Delta and Notch functions during microchaeta development.
Seydoux and Greenwald (1989) had previously suggested that
expression of lin-12, a Notch homolog, could be subject to sig-
nalling-dependent feedback regulation, based on somatic
mosaic analysis in C. elegans. Wilkinson et al. (1994) subse-
quently found that lin-12 receptor activity represses activation
of transcriptional regulatory sequences of lag-2, which encodes
an apparent ligand for the lin-12 receptor (Tax et al., 1994;
Henderson et al., 1994), implying that signalling activity exerts
negative feedback regulation on ligand expression for the lag-
2/lin-12 ligand/receptor pair.
We present three lines of evidence supporting the hypothesis
that Delta-Notch signalling exerts feedback control on Delta and
Notch expression in wing vein morphogenesis. First, Delta and
Notch protein expression patterns are generally complementary
within the developing wing blade. In general, cells that express
higher levels of Delta and lower levels of Notch are flanked by
cells that express higher levels of Notch and lower levels of
Delta. This complementarity becomes more pronounced as
pupal development proceeds and proveins are partitioned into
vein and intervein cell types. We observe similar regionally com-
plementary patterns of Delta and Notch expression in the pupal
notum during sensory organ precursor specification (Parks et al.,
1997). Second, elevated Notch expression is dependent on Delta
activity in two cell populations in the pupal wing. The DlEW
mutation, which drastically reduces Delta expression in pre-
sumptive vein cells, leads to the absence of elevated Notch
expression in lateral provein cells. In addition, when a dominant-
negative form of Delta is expressed in lateral provein cells,
Delta-mediated signalling is hindered, leading to the absence of
elevated Notch expression in lateral provein cells. Elevated
Notch expression in lateral provein cells is therefore dependent
on Delta activity in presumptive vein cells. Elevation of Delta
expression in intervein cells leads to elevated Notch expression
in these cells, relative to the levels of Notch expression observed
in intervein cells in wild-type pupal wings. These observations
imply that Delta-mediated signalling from presumptive vein
cells positively regulates Notch expression in lateral provein
cells and that Delta-mediated signalling in intervein cells can
positively regulate Notch expression in these cells. Third, con-
stitutive activation of the Notch receptor, mediated by uniform
expression of a gain-of-function Notch(intra) variant (Struhl et
al., 1993), represses Delta expression in the pupal wing blade.
Analogous repression of Delta expression is observed after
ectopic Notch activation in the pupal notum (Parks et al., 1997)
and retina (A. L. Parks and M. A. T. Muskavitch, unpublished
data). These findings imply that Delta-Notch signalling activates
Notch expression and represses Delta expression in lateral
provein cells during pupal wing vein development, and indicate
that Delta-Notch signalling exerts feedback regulation on Delta
and Notch expression within the provein anlage. Our data
suggest that feedback regulation of ligand and receptor
expression may be a general property of Notch-mediated sig-
nalling mechanisms in Drosophila.
Our findings regarding Delta and Notch expression in pre-
sumptive vein cells raise the intriguing possibility that these
cells, which exhibit elevated Delta expression and reduced Notch
expression, are refractory to this Delta-Notch-mediated feedback
regulation. The fact that presumptive vein cells are capable of
responding to Notch receptor activity is demonstrated by the
repression of Delta expression in these cells that we observe
following ectopic induction of constitutive Notch receptor
activity in pupal wings. However, levels of Delta in a wild-type
genetic background may be sufficient to eliminate, in a cell-
autonomous manner, the ability of these cells to receive a
normally regulated Delta-Notch signal. Such cell-autonomous
inhibition of neurogenic signal reception has been described by
Micchelli et al. (1997). Cells that express high levels of Delta or
Serrate along the developing wing margin do not exhibit the
elevated expression of cut and wingless normally induced by the
receipt of neurogenic signals, even though expression of margin-
like levels of cut and wingless are induced in cells that lack Delta
and Serrate and are adjacent to highly expressing cells. Within
the provein, analogous inhibition of neurogenic signal reception
in presumptive vein cells would lead to the observed persistence
of elevated levels of Delta expression and a failure to induce
increased Notch expression in central provein cells.
Asymmetry in Delta and Notch protein expression is
required for wing vein development
Asymmetries in Delta and Notch expression – and in the gen-
eration and reception of neurogenic signals – appear to underlie
the partitioning of vein and intervein cell fates within the pupal
provein. The juxtaposition of cells that express elevated levels
of Delta or Notch, respectively, establishes a boundary
condition that arises as a result of feedback regulation by neu-
rogenic signalling of Delta and Notch expression in lateral
provein cells. Presumptive vein (i.e., central provein) cells, on
one side of this boundary, express elevated levels of Delta and
send the neurogenic signal. Presumptive intervein (i.e., lateral
provein) cells, on the other side, express elevated levels of
Notch and receive this signal, which inhibits adoption of the
vein cell fate by lateral provein cells.
Disruption of the wild-type pattern of asymmetric Delta and
Notch expression within the provein, by means of ectopic
elevation of Delta expression in lateral provein cells using the
GAL4-UAS system, can lead to the failure of presumptive vein
cells to stably adopt the vein cell fate. Elevation of Delta
expression in the lateral provein disrupts the Delta-Notch
boundary condition that exists in wild-type proveins (Fig. 6).
As a result of this disruption, lateral provein (i.e., presumptive
intervein) cells express higher levels of Delta than central
provein (i.e., presumptive vein) cells, and central provein cells
exhibit elevated levels of Notch expression compared to their
counterparts in wild-type proveins. This implies that Delta
expression in lateral provein cells can induce elevated Notch
expression in central provein cells when levels of Delta
expression in presumptive intervein cells exceed levels in pre-
sumptive vein cells. This newly induced ability of central
provein cells to receive a neurogenic signal can apparently lead
to inhibition of adoption of the vein cell fate within the
proveins for L2, L4 and L5. Adult veins are therefore shortened
or gapped because central provein, as well as lateral provein,
cells are now inhibited from adopting the vein cell fate.
Qualitative characteristics of Delta signalling during
The fact that Delta expression in central provein cells affects
specification of lateral provein cells in wild-type wing discs
provides additional evidence for the nonautonomy of Delta
function in vivo (Heitzler and Simpson, 1991; Parks et al.,
1995). We also find that expression of Delta in intervein (i.e.,
lateral provein) cells, in 1348::GAL4/+;UAS::DeltaWT/+
animals, inhibits adoption of the vein cell fate by central provein
cells. This outcome must also reflect nonautonomous Delta
function within the provein because expression of constitutively
active Notch receptor [i.e., Notch(intra)] under control of the
1348::GAL4 driver does not inhibit the development of vein
cells (S. S. Huppert, T. L. Jacobsen, and M. A. T. Muskavitch,
unpublished data). If the silencing of vein cell fate in
1348::GAL4/+;UAS::DeltaWT/+ animals were the result of
cell-autonomous neurogenic signalling, then expression of the
constitutively active Notch receptor under control of the same
driver would also inhibit adoption of the vein cell fate.
Serrate, another Notch ligand (Rebay et al., 1991), can sub-
stitute for Delta in some developmental contexts (Gu et al.,
1995; T. L. Jacobsen and M. A. T. Muskavitch, unpublished
data). However, we find that expression of Serrate under
control of the 1348::GAL4 driver exerts no discernible effects
on Delta and Notch expression patterns or on the specification
of vein cell fates (data not shown). This implies that Serrate
cannot function as an effective ligand for Notch in the sig-
nalling process that underlies partitioning of vein and intervein
cell fates within the provein during metamorphosis. In this
developmental context, then, Serrate cannot substitute for
Delta as an activator of neurogenic signalling.
Our observation that veins develop in the virtual absence of
pupal Delta expression in the DlEWgenetic background implies
that Delta function, per se, is not required for direct specifica-
tion of the vein cell fate or vein cell differentiation in meta-
morphic wing discs. Within the pupal provein, Delta function
is required solely to prevent adoption of the vein cell fate by
lateral provein cells. This is consistent with the observation that
morphologically and functionally normal anchor cells develop
even when lag-2 activity is severely reduced by mutation in C.
elegans (Lambie and Kimble, 1991). In these two instances,
neurogenic signaling activity is required to modulate the
adoption of cell fates, but is not required for the differentiation
of particular cell types after they have stably adopted their
Finally, our data reveal that there are qualitative differences
within the provein among the relationships between signalling
activity and feedback regulation of receptor expression for the
three signal transduction pathways known to influence wing vein
development during metamorphosis. Delta/Notch signalling
activity leads to up-regulation of receptor expression in lateral
provein cells. In contrast, dpp/TKV (de Celis, 1997) and DER
(Sturtevant et al., 1994)-dependent signalling are believed to lead
to the down-regulation of receptor expression in central provein
cells. This implies that ligand-dependent signalling, for TKV
and DER pathways, within the provein may be self-limiting and
required for only brief periods during wing vein development.
In contrast, Delta-dependent signalling within the provein may
be required for an extended period of time during metamorpho-
sis for the correct partitioning of vein and intervein cell fates.
We would like to thank Gary Struhl and Gerhard Technau for fly
stocks, Kris Klueg and Annette Parks for reading the manuscript, and
Susan Fugate and Kari Huppert for revisions in the Reference section.
This work was supported by grant GM33291 from the National Insti-
tutes of Health to M. A. T. M.
S. S. Huppert and others
3291Delta-Notch feedback regulation Download full-text
Artavanis-Tsakonas, S., Matsuno, K. and Fortini, M. E. (1995). Notch
signalling. Science 268, 225-232.
Ashburner, M. (1989). Appendix C: Mounting and embedding media. In
Drosophila: A Laboratory Manual, Vol. 2, pp. 337. New York: Cold Spring
Harbor Laboratory Press.
Brand, A. H. and Perrimon, N. (1993). Targeted gene expression as a means
of altering cell fates and generating dominant phenotypes. Development 118,
de Celis, J. F. (1997). Expression and function of decapentaplegic and thick
veins during the differentiation of the veins in the Drosophila wing.
Development 124, 1007-1018.
de Celis, J. F. and García-Bellido, A. (1994a). Modifications of the Notch
function by Abruptex mutations in Drosophila melanogaster. Genetics 136,
de Celis, J. F. and García-Bellido, A. (1994b). Roles of the Notch gene in
Drosophila wing morphogenesis. Mech. Dev. 46, 109-122.
Dexter, J. S. (1914). The analysis of a case of continuous variation in
Drosophila by a study of its linkage relationships. Am. Nat. 68, 712-758.
Diaz-Benjumea, F. J. and García-Bellido, A. (1990). Genetic analysis of the
wing vein pattern of Drosophila. Roux’s Arch. Dev. Biol. 198, 336-354.
Doherty, D., Feger, G., Younger-Shepherd, S., Jan, L. Y. and Jan, Y. N.
(1996). Delta is a ventral to dorsal signal complementary to Serrate, another
Notch ligand, in Drosophila wing formation. Genes Dev. 10, 421-434.
Fehon, R. G., Johanson, K., Rebay, I. and Artavanis-Tsakonas, S. (1991).
Complex cellular and subcellular regulation of Notch expression during
embryonic and imaginal development of Drosophila: Implications for Notch
function. J. Cell Biol. 113, 657-669.
Fehon, R. G., Kooh, P. J., Rebay, I., Regan, C. L., Xu, T., Muskavitch, M. A.
T. and Artavanis-Tsakonas, S. (1990). Molecular interactions between the
protein products of the neurogenic loci Notch and Delta, two EGF
homologous genes in Drosophila. Cell 61, 523-534.
Foster, G. G. (1975). Negative complementation at the Notch locus of
Drosophila melanogaster. Genetics 81, 99-120.
Fristrom, D., Gotwals, P., Eaton, S., Kornberg, T. B., Sturtevant, M., Bier,
E. and Fristrom, J. W. (1994). blistered: a gene required for vein/intervein
formation in wings of Drosophila. Development 120, 2661-2671.
Fristrom, D., Wilcox, M. and Fristrom, J. (1993). The distribution of PS
integrins, laminin A, and F-actin during key stages in Drosophila wing
development. Development 117, 509-523.
Gallyas, F., Göres, T. and Merchenthaler, I. (1982). High-grade
intensification of the end-product of the diaminobenzidine reaction for
peroxidase histochemistry. J. Histochem. Cytochem. 30, 183-184.
García-Bellido, A. and de Celis, J. F. (1992). Developmental genetics of the
venation pattern of Drosophila. Ann. Rev. Gen. 26, 277-304.
Gu, Y., Hukriede, N. A. and Fleming, R. J. (1995). Serrate expression can
functionally replace Delta activity during neuroblast segregation in the
Drosophila embryo. Development 121, 855-865.
Heitzler, P. and Simpson, P. (1991). The choice of cell fate in the epidermis of
Drosophila. Cell 64, 1083-1092.
Heitzler, P. and Simpson, P. (1993). Altered epidermal growth factor-like
sequences provide evidence for a role of Notch as a receptor in cell fate
decisions. Development 117, 1113-1123.
Henderson, S. T. Gao, D., Lambie, E. J. and Kimble, J. (1994). lag-2 may
encode a signaling ligand for the GLP-1 and LIN-12 receptors of C. elegans.
Development 120, 2913-2924.
Hoppe, P. E. and Greenspan, R. J. (1986). Local function of the Notch gene
for embryonic ectodermal pathway choice in Drosophila. Cell 46, 773-783.
Hoppe, P. E. and Greenspan, R. J. (1990). The Notch locus of Drosophila is
required in epidermal cells for epidermal development. Development 109,
Kooh, P. J., Fehon, R. G. and Muskavitch, M. A. T. (1993). Implications of
dynamic patterns of Delta and Notch expression for cellular interactions
during Drosophila development. Development 117, 493-507.
Kopczynski, C. C., Alton, A. K., Fechtel, K., Kooh, P. J. and Muskavitch,
M. A. (1988). Delta, a Drosophila neurogenic gene, is transcriptionally
complex and encodes a protein related to blood coagulation factors and
epidermal growth factors of vertebrates. Genes Dev. 2, 1723-1735.
Lambie, E. and Kimble, J. (1991). Two homologous genes, lin-12 and glp-1,
have overlapping functions. Development 112, 231-240.
Lehmann, R., Jiménez, F., Dietrich, U. and Campos-Ortega, J. A. (1983).
On the phenotype and development of mutants of early neurogenesis in
Drosophila melanogaster. Roux’s Arch. Dev. Biol. 192, 62-74.
Li, K. and Kaufman, T. C. (1996). The Homeotic Target Gene centrosomin
Encodes an Essential Centrosomal Component. Cell 85, 585-596.
Lieber, T., Kidd, S., Alcamo, E., Corbin, V. and Young, M. W. (1993).
Antineurogenic phenotypes induced by truncated Notch proteins indicate a
role in signal transduction and may point to a novel function for Notch in
nuclei. Genes Dev. 7, 1949-1965.
Lindsley, D. L. and Zimm, G. G. (1992). The Genome of Drosophila
melanogaster. New York: Academic Press.
Liposits, Z. S., Sétáló, G. Y. and Lerkó, B. F.(1984). Application of the silver-
gold intensified 3,3-diaminobenzidine chromatin to the light and electron
microscopic detection of the luteinizing hormone releasing hormone
detection system of the rat brain. Neuroscience 13, 513-525.
Lyman, D. F. and Yedvobnick, B. (1995). Drosophila Notch receptor activity
suppresses Hairless function during adult external sensory organ
development. Genetics 141, 1491-1505.
Micchelli, C. A., Rulifson, E. J. and Blair, S. S. (1997). The function and
regulation of cut expression on the wing margin of Drosophila: Notch,
Wingless, and a dominant negative role for Delta and Serrate. Development
Muskavitch, M. A. T. (1994). Delta-Notch signaling and Drosophila cell fate
choice. Dev. Biol. 166, 415-430.
Palka, J., Schubiger, M. and Schwaninger, H. (1990). Neurogenic and
antineurogenic effects from modifications at the Notch locus. Development
Parks, A. L., Huppert, S. S., Muskavitch, M. A. T. (1997). The dynamics of
neurogenic signalling underlying bristle development in Drosophila
melanogaster. Mech. Dev. 63, 61-74.
Parks, A. L., Turner, F. R. and Muskavitch, M. A. T. (1995). Relationships
between complex Delta expression and the specification of retinal cell fates
during Drosophila eye development. Mech. Dev. 50, 201-216.
Parody, T. R. and Muskavitch, M. A. T. (1993). The pleiotropic function of
Delta during postembryonic development of Drosophila melanogaster.
Genetics 135, 527-539.
Portin, P. (1975). Allelic negative complementation at the Abruptex locus of
Drosophila melanogaster. Genetics 81, 121-133.
Rebay, I., Fehon, R. G. and Artavanis-Tsakonas, S. (1993). Specific
truncations of Drosophila Notch define dominant activated and dominant
negative forms of the receptor. Cell 74, 319-329.
Rebay, I., Fleming, R. J., Fehon, R. G., Cherbas, L., Cherbas, P. and
Artavanis-Tsakonas, S. (1991). Specific EGF repeats of Notch mediate
interactions with Delta and Serrate: implications for Notch as a
multifunctional receptor. Cell 67, 687-699.
Seydoux, G. and Greenwald, I. (1989). Cell autonomy of lin-12 function in a
cell fate decision in C. elegans. Cell 57, 1237-1245.
Shellenbarger, D. L. and Mohler, J. D. (1978). Temperature-sensitive periods
and autonomy of pleiotropic effects of l(1)Nts1, a conditional Notch lethal in
Drosophila. Dev. Biol. 62, 432-446.
Struhl, G., Fitzgerald, K. and Greenwald, I. (1993). Intrinsic activity of the
Lin-12 and Notch intracellular domains in vivo. Cell 74, 331-345.
Sturtevant, M. A. and Bier, E. (1995). Analysis of the genetic hierarchy
guiding wing vein development in Drosophila. Development 121, 785-801.
Sturtevant, M. A., O’Neill, J. W. and Bier, E. (1994). Down-regulation of
Drosophila Egf-r mRNA levels following hyperactivated receptor signaling.
Development 120, 2593-2600.
Tax, F., Yeagers, J. J. and Thomas, J. H. (1994). Sequence of C. elegans lag-2
reveals a cell-signalling domain shared with Delta and Serrate of Drosophila.
Nature 368, 150-154.
Waddington, C. H. (1940). The genetic control of wing development in
Drosophila. J. Genetics 41, 75-139.
Welshons, W.J. (1965). Analysis of a gene in Drosophila. Science 150, 1122-
Wilkinson, H. A., Fitzgerald, K. and Greenwald, I. (1994). Reciprocal
changes in expression of the receptor lin-12 and its ligand lag-2 prior to cell
commitment in a C. elegans cell fate decision. Cell 79, 1187-1198.
Wilson, C., Pearson, R. K., Bellen, H. J., O’Kane, C. J., Grossniklaus, U.
and Gehring, W. J. (1989). P-element-mediated enhancer detection: an
efficient method for isolating and characterizing developmentally regulated
genes in Drosophila. Genes Dev. 3, 1301-1313.
(Accepted 25 June 1997)