Download asense is a Drosophila neural precursor gene and is

1
Development 119, 1-17 (1993)
Printed in Great Britain © The Company of Biologists Limited 1993
asense is a Drosophila neural precursor gene and is capable of initiating
sense organ formation
Michael Brand*, Andrew P. Jarman, Lily Y. Jan and Yuh Nung Jan
Howard Hughes Medical Institute, University of California, San Francisco, CA 94143-0724, USA
*Present address: Max Planck Institut für Entwicklungsbiologie, Spemannstrasse 35/III, D-7400 Tübingen 1, FR of Germany
SUMMARY
Neural precursor cells in Drosophila arise from the
ectoderm in the embryo and from imaginal disc epithelia
in the larva. In both cases, this process requires daugh terless and the proneural genes achaete, scute and lethalof-scute of the achaete-scute complex. These genes
encode basic helix-loop-helix proteins, which are nuclear
transcription factors, as does the asense gene of the
achaete-scute complex. Our studies suggest that asense is
a neural precursor gene, rather than a proneural gene.
Unlike the proneural achaete-scute gene products, the
asense RNA and protein are found in the neural
precursor during its formation, but not in the proneural
cluster of cells that gives rise to the neural precursor cell.
Also, asense expression persists longer during neural
precursor development than the proneural gene
products; it is still expressed after the first division of the
neural precursor. Moreover, asense is likely to be downstream of the proneural genes, because (1) asense
expression is affected in proneural and neurogenic
mutant backgrounds, (2) ectopic expression of asense
protein with an intact DNA-binding domain bypasses the
requirement for achaete and scute in the formation of
imaginal sense organs. We further note that asense
ectopic expression is capable of initiating the sense organ
fate in cells that do not normally require the action of
asense. Our studies therefore serve as a cautionary note
for the inference of normal gene function based on the
gain-of-function phenotype after ectopic expression.
INTRODUCTION
Another group of genes, the neural precursor genes, are
activated in most or all newly born neural precursors (neuroblasts in the central nervous system (CNS), and sense
organ precursors, SOPs, in the peripheral nervous system
(PNS)), and may control different aspects of neuronal differentiation (Bier et al., 1989, 1992; Vaessin et al., 1991). We
wanted to determine whether this group of genes is a part of
the downstream genetic program activated by AS-C/da heterodimers. As a first step, we chose to study the expression
and activity of one putative member of this group, the asense
(ase) gene. ase is located in the AS-C and itself encodes a
bHLH protein (Alonso and Cabrera, 1988; González et al.,
1989). Previous studies (Alonso and Cabrera, 1988;
González et al., 1989) have raised the possibility that ase is
functionally distinct from the proneural members of the ASC, achaete (ac), scute (sc) and lethal-of-scute (l’sc), because
ase expression appears to be activated later than that of the
proneural genes and it appears to persist for a longer period
of time. However, the identity of the ase-expressing cells
was not established in these studies.
Using antibodies against the ase protein, we have carried
out a detailed examination of ase expression during
embryonic and larval development. The ase expression
closely follows proneural gene expression in space and time
and is found in neural precursors shortly after their
Neural precursor cells in Drosophila arise from undifferentiated ectodermal cells in the embryo or imaginal discs
(reviewed by Campos-Ortega, 1988; Ghysen and DamblyChaudière, 1989; Campos-Ortega and Jan, 1991; ArtavanisTsakonis, 1988; Simpson, 1990; Cabrera, 1992; Campuzano
and Modolell, 1992). This process requires the action of the
proneural genes such as those of the achaete-scute complex
(AS-C), which encode members of the basic helix-loophelix (bHLH) group of transcriptional factors. The proneural
genes of the AS-C are expressed in small groups of cells in
the epithelium, called proneural clusters, prior to the generation of a neural precursor; this expression is thought to
make the cells competent to follow a neural fate. Through
cell-cell interactions mediated by the neurogenic genes,
expression of the proneural genes is then restricted to a
single cell (the neural precursor) that will delaminate from
the epithelium and contribute to the formation of the nervous
system. Previous studies of genetic interactions, as well as
interactions of the protein products in vitro or in yeast,
suggest that AS-C products form heterodimers with the
product of daughterless (da) to activate genes that lead these
precursor cells to execute a neural fate (Dambly-Chaudière
et al., 1988; Murre et al., 1989; Cabrera and Alonso, 1991).
Key words: achaete-scute complex, helix-loop-helix, proneural
genes, neural precursor genes, neurogenesis, Drosophila
2
M. Brand and others
formation, suggesting that ase is a member of the neural
precursor group. We also show that normal ase expression
requires the function of the proneural genes, whereas ectopic
expression of ase induces sense organ precursor formation
even in the absence of the proneural genes normally required
for these precursors.
MATERIALS AND METHODS
Flies and culture conditions
Flies were crossed under standard conditions at 25°C, unless
otherwise noted. Different aspects of the mutant stocks employed
in this study are described by Lindsley and Grell, 1968; Lindsley
and Zimm, 1992; García Bellido, 1979; García-Alonso and García
Bellido, 1986; Caudy et al., 1988; Jiménez and Campos-Ortega,
1990. ase1 was renamed from sc2; it is a small deletion removing
only the asense gene and a putative regulatory element of the sc
gene (González et al., 1989; Jarman et al., 1993). As neurogenic
mutations, the amorphic alleles N55e11 and neuKX9 were used
(Lehmann et al., 1983).
sc10-1 was marked with white1 and maintained with In (1) dl49,
y Hw. Larvae for sc10-1 w/Y (y+, w−) were identified in the third
instar based on mouth hook and Malpighian tubule coloration.
In(1)ac3 larvae were obtained from a homozygous stock. Larvae
carrying emcpel/Df(3)emcE12, a strong viable combination of emc
alleles, were obtained as Tubby+ larvae from a cross of
emcpel/emcpel with Df(3)emcE12/TM6B, Tb.
Heat-shock protocol
Staging of embryos was done on the basis of timed collections and
morphological criteria described in Campos-Ortega and Hartenstein (1985a). Larvae were staged for the time course of heat-shock
induction of ase relative to the third instar molt, as described by
Huang et al. (1991). This method of staging was chosen because
it introduces the smallest scatter of age during third instar,
puparium formation occuring at 48±3 hours (Huang et al., 1991).
Time was converted to hours bpf (before puparium formation) by
subtraction from 48. Pupae were collected at the white prepupae
stage (= PF). Larvae and pupae were left to develop for the
indicated times at 25°C before they were subjected to heat-shock
treatment. For other staining, climbing third instar larvae were
taken from uncrowded vials maintained at 18°C.
Heat-shock treatment of larvae and pupae was done for hs-ase
with two 25 minute pulses at 39°C, separated by 30 minutes at
room temperature. Larvae were collected into a basket inside an
Eppendorff tube half-filled with water and incubated in a water
bath. Pupae were treated on a floating agar dish. For hs-Gal4-ase
and hs-Gal4-l’sc, the incubation time was reduced to 15 minutes.
Histochemical techniques
Whole-mount hybridization with digoxigenin-labelled probes was
carried out essentially as described (Tautz and Pfeiffle, 1989). A
fragment containing only the coding region of ase (from pAse-Kpn,
see below) was used as a probe; probe preparation by random
primer labelling was according to the instruction of the manufacturer (Boehringer), except that a higher concentration of random
primer (1 mg/ml) was used to increase labelling efficiency.
Polyclonal antibodies were raised by injecting two rabbits with
the peptide CLSDESMIDAIDWWEAHAPKSNGACTNLSV, corresponding to a fragment from the C terminus of the putative ase
protein. The terminal Cys residue was used to couple the peptide
under oxidizing conditions to keyhole limpet haemocyanin carrier
protein. Positive sera from one rabbit were further purified by
affinity chromatography using bacterially produced ase fusion
protein (see Jarman et al., 1993) on an Affigel column (Biorad).
Before use, the antibodies were preabsorbed with homozygous ase1
embryonic or larval tissue; final dilution of the antibody fractions
was approximately 1:100. Western blot analysis of embryonic
extracts was done as described in Blochlinger et al. (1991). 22C10
(1:50) and anti-ac (1:20) are mouse monoclonal antibodies kindly
provided by Drs Seymour Benzer and Sean Carrol, respectively.
Anti-β-galactosidase is a polyclonal rabbit antiserum (Polysciences) used at 1:10 000 to 1:50 000. Secondary antibodies
coupled to HRP (1:500) or conjugated to texas red or biotin (Multilabel grade, 1:200) were obtained from Jackson Immunoresearch,
and Avidin-Bodipy (1:50) was purchased from Molecular Probes.
Standard antibody staining protocols were used throughout as
described in Hartenstein and Campos-Ortega (1986). Imaginal
discs were dissected in 100 mM phoshate buffer pH 7.0, fixed for
30 minutes in 4% paraformaldehyde in PBS and processed for
staining. ‘Blue balancers’ with lacZ insertions were used to allow
identification of homozygous Df(2L)daKX136 and Df(1)sc19
embryos. Homozygous embryos could be recognized because of
the absence of anti-β gal staining. For double immunofluorescence
labelling, imaginal discs were simultaneously incubated with antiase and a mouse monoclonal anti-β galactosidase (Promega, 1:50)
or anti-ac antibodies, followed by donkey anti-rabbit-biotin plus
goat anti-mouse-texas red, and a third incubation with avidinbodipy. Double-labelled discs were examined on a Biorad MRC
600 confocal microscope equipped with a Krypton/Argon laser.
Data were recorded and processed using commercial software
(CM, Biorad) on a Compaq PC and an IBM worm drive. Several
optical sections from a z-series were combined to form the images
of the entire discs, which were subsequently artificially colored
using Lumena software. Standard light microscopic photography
was done on a Nikon photomicroscope equipped with Nomarski
optics.
Construction of hs-ase
A genomic BamHI fragment containing the ase gene was
subcloned from the λsc53 phage (Gonzalez et al., 1989); subsequent numbering refers to the coordinates used by these authors.
The coding region was isolated from plasmid DNA using the
primer pair A 1941-T1960 and G 3985-C3404 under standard PCR con ditions, ligated to KpnI linkers (NEB) and cloned into bluescript
SK+ to give pAse-Kpn, the sequence of which was determined by
standard sequencing protocols (USB). Two conservative
nucleotide exchanges were found: A591 to C, A745 to T. The
fragment was then cloned into the KpnI site of pWH1 (Blochlinger
et al., 1991) in the sense and antisense orientation and transformed
into flies by standard P-element-mediated transformation. Go
larvae were raised at 18°C and eclosing flies were individually
tested for the presence of w+ transformants by mating them to y w
flies. 12 of 13 transformants carrying hs-ase in the sense orientation showed upon heat shock the phenotypes that we describe here;
19 transformants carrying the same KpnI fragment in the antisense
orientation did not show a phenotype with or without heat shock.
Stocks were established from these lines; in the experiments
described, we used two homozygous viable transformant lines, hsase4a and hs-ase10a alone (2 copies) or in combination (4 copies).
To obtain hs-Gal4-ase, an Asp718 I fragment containing the coding
region from pAse-Kpn was inserted into the Asp718 I site of
pUAST (A. Brand and N. Perrimon) and injected into w embryos.
Of more than 30 independent transformants, 15 were established
as stocks and tested by crossing them to a stock producing Gal4
under control of a heat-shock promotor (hs-Gal4/CyO; obtained
from Ed Giniger). All crosses showed extra sense organs after heat
shock in CyO+, but not in CyO control flies. Lines hs-Gal4-ase 3
and hs-Gal4-ase 5 were examined in more detail by heat shocking
larvae and pupae; they showed the phenotypes described in Fig. 8.
Stocks carrying hs-Gal4-l’sc were obtained in an analogous
fashion by PCR isolation of the coding region of l’sc, sequencing,
asense, a neural precursor gene
cloning into pUAST (H. Vaessin) and transforming it into flies.
Lines hs-Gal4-l’sc 2-4a and hs-Gal4-l’sc 13-3 were used.
In vitro mutagenesis
The point mutant was obtained by PCR mutagenesis. A fragment
of 500 bp was isolated using the primer pair A1941-T1960; A2474TT
GTT AAC CTG CTT CAG GCC ATT TCC TTC CCT AGC 2439.
Highlighted bases lead to exchange of Arg171 and Arg 173 to Gly,
and Val 174 to Leu; the latter exchange is assumed to be insignificant since Leu occurs in this position in many bHLH proteins. The
PCR product was digested with HindIII and HpaI and directly
inserted in place of the corresponding wild-type fragment in pAseKpn, from which the HindIII site of the polylinker had previously
been removed, to yield pAse-PM. A deletion of the basic domain
only or both the basic and HLH domain was obtained by ‘divergent
PCR’ (Hemsley et al., 1989) with primers exactly flanking the
region to be deleted and pAse-Kpn as a template. Primers ase10
(C2428-G2408) and ase11 (C2455-G2476) were used to delete the basic
domain (pAse-bd−), and ase10 and ase12 (G2621-C2641) to delete
both basic and bHLH domain (pAse-bHLH−); these three primers
also contained a unique BglII site that introduces the amino acids
Leu and Asp in place of the deletions. PCR products were cut with
BglII, ligated and transformed. The HindIII-HpaI fragment was
then inserted into the pAse-Kpn backbone, as described above for
the point mutant, to yield pAse-bd − andpAse-bHLH−. All mutated
plasmids were then sequenced to ensure the presence of the
mutations. Mutated ase coding regions were then inserted into the
Asp718 I site of pWH1 (pAse-PM, pAse-bd−) and pUAST (all three
mutants) and P-element transformants were derived as above.
Stocks of homozygous transformants were tested as described
above by crossing to hs-Gal4/CyO females and comparing the
effects of heat-shock treatment for CyO+ and CyO control flies.
The following lines were used after their protein production and
the size of the protein were confirmed by western blot analysis of
embryonic extracts (Fig. 7): ‘Point mutant’: hs-Gal4-ase PM 11.1
and hs- ase PM 1; deletion of the basic domain: hs-Gal4-ase bd−
7.3 and 7.5, and hs- ase bd− 7.1, 7.2 and 7.4; deletion of the bHLH
domain: hs-Gal4-ase bHLH− 13.1a.
RESULTS
The embryonic expression pattern of asense RNA
and protein
The distribution of ase RNA has been studied previously by
in-situ hybridization to sections (Alonso and Cabrera, 1988;
González et al., 1989). It was noted that ase is expressed
later than other genes of the AS-C, though the identity of
the ase-expressing cells was not established. We have raised
antibodies against a C-terminal peptide of the ase protein for
immunocytochemical localization of ase, and compared the
protein distribution to the RNA pattern as determined by the
more sensitive whole-mount in situ hybridization technique
(Figs 1, 2). Staining for the antigen is eliminated in embryos
deficient for the ase gene (ase1) (not shown), indicating that
ase is the only member of the bHLH family recognized by
this antibody. RNA was detected a few minutes before the
protein; otherwise, we do not observe differences between
the RNA and the protein pattern except in the eye disc. The
following description is based on analyzing both.
Onset of asense expression in precursors of CNS and
PNS
ase expression is first detected shortly after the onset of gas-
3
trulation in the nuclei of single cells that are in the process
of segregating from the neuroectodermal epithelium, but
not in the surrounding cells of the proneural cluster (Fig.
1B,C). By contrast, proneural genes such as sc, l’sc
(Cabrera et al., 1987; Romani et al., 1987) and ac (Fig. 1A)
already show strong expression in proneural clusters. While
ac is expressed in a subset of neuroblasts (Fig. 1D), the aseexpressing cells give rise to the entire first wave of segregating neuroblasts (SI) and most likely all of the second and
third wave (SII and SIII) of neuroblasts as well (Fig. 1E,F).
ase is also found in at least some of the ganglion mother
cells, progeny of the neuroblasts (Fig. 1G,H). During
germband retraction (stage 16), ase is seen in a subset of
ventroperipherally located cells of the ventral nerve chord
and in conspicuously large and superficial cells of the brain
hemispheres (Fig. 1I). These cells correspond in their
position and appearance to larval neuroblasts and the optic
lobe anlage (Truman and Bate, 1988). The restriction of
ase expression to neural precursors and the prolonged
presence after cell division set ase apart from the proneural
genes.
In the PNS, expression of ase closely resembles the early
expression of the PNS marker line A37, which marks all
sense organ precursors (SOPs) and their progeny (Ghysen
and O’Kane, 1989). A subset of these cells variably depends
on ase (Dambly-Chaudière and Ghysen, 1987; Jarman et al.,
1993). Of the first two SOPs to arise in the PNS anlage, the
anterior one (the A cell; Ghysen and O’Kane, 1989) gives
rise to a sense organ, lk or lh3, that partially depends on ase,
while the posterior one (the P cell) produces the ASC-independent chordotonal organs (Dambly-Chaudière and
Ghysen, 1987). We find ase to be expressed from early stage
10 onwards in both the A and P cells (Fig. 2A,B,E–G), as
well as in most or all precursors that form subsequently (Fig.
2C,D,H,I). At early stages, a third cell of unknown fate may
be seen in addition to the A and P cell. (Fig. 2F). We do not
detect ase RNA or protein expression in proneural clusters
in the ectoderm at any stage. Thus, even the proneural
cluster for a partially ase-dependent SOP does not show
detectable expression of ase.
Unlike ac or sc (Cubas et al., 1991; Skeath and Carroll,
1991), ase remains detectable during and after division of
the SOP (Fig. 2G–I). Some time after the first division, the
progeny of the A and P cells cease to be labelled; instead,
additional labelled cells appear in other subepidermal
positions of the PNS anlage (Fig. 2H). Due to the expression
in daughter cells (secondary SOPs), newly segregating SOPs
are not easily distinguishable after this stage. ase also
remains detectable in a subset of PNS cells during and after
germband retraction stages, when products of the other ASC
genes are no longer detectable (Fig. 2C,D,I).
Expression in the midgut anlage
Expression of ase is not confined to neural cells: from stage
10 onwards, expression is also observed in the midgut
anlage (see below and Fig. 5), as are ac and l’sc (Romani et
al., 1987; Cabrera et al., 1987). At late embryonic stages,
these cells display a distinctive non-neuronal morphology
(A. Jarman, unpublished), and at least the majority of them
correspond to precursor cells for the imaginal midgut
(Hartenstein et al., 1992).
4
M. Brand and others
asense, a neural precursor gene
5
Fig. 2. asense is expressed in sense organ precursors in the embryo. Shown are whole mounts of wild-type embryos hybridized with (A)
digoxigenin-labelled probes or (B–I) anti-ase antibodies to detect ase mRNA or protein, respectively. (A,B) Onset of RNA (and protein)
expression in the PNS of an early stage 10 embryo occurs, initially with slight asynchrony, in three separate cells per hemisegment (one is
indicated by an arrow). (C,D) Additional ase-expressing cells appear in the PNS anlage (bracket) during stages 11 (C) and 12 (D).
(E–I) Higher magnification views of the process of SOP formation in the first abdominal segment. Notice that only single cells are stained
with the antibodies against ase. Staining of neuroblasts and some ganglion mother cells can also be observed ventrally. (E,F) The A and P
cells (Ghysen and O’Kane, 1989), and the third unidentified cell (*) are indicated. (G) Later, A and P give rise to pairs of cells apparently
by division; both daughter cells express ase, while expression in the third cell weakens; its fate is unknown. Dorsally, the next arising cell
is beginning to stain faintly at a different focal plane. (H) At stage 11, several additional isolated cells have arisen, whereas expression in
the pair of cells shown in (G) is diminished or gone. The remaining expression in the presumed A cell progeny is indicated. (I) At stage
13, ase-positive cells are still seen, whereas proneural gene expression is already mostly extinguished (not shown). The future
arrangement of PNS organs in four clusters (dorsal, d; lateral, l; ventral′, v′; and ventral, v) is already foreshadowed.
Postembryonic expression of asense
SOPs in the imaginal discs arise in a well-ordered spatial
and temporal manner and are readily detected by lacZ
expression due to an enhancer trap insertion, A101, in the
neuralized gene (Huang et al., 1991; Boulianne et al., 1991).
We compared ase expression with A101 expression in order
to determine whether ase is expressed in SOPs. We also
compared the temporal pattern of ase expression with that
of the proneural ac gene product (Fig. 3).
Previous studies failed to detect ase expression in the
wing disc (González et al., 1989). We now find, however,
that both ase RNA and protein are expressed in a pattern
reminiscent of SOPs (Fig. 3A). Confocal double immunofluorescence visualization of ase and A101 confirms that the
ase-expressing cells are SOPs (Fig. 3B–E). Up to about 3
hours after puparium formation (apf), the oldest stage that
we have examined, all SOPs that arise appear to express ase.
Onset of ase expression at each position is detected shortly
after A101 expression becomes detectable. Unlike A101,
which occasionally labels 2–3 cells transiently before
expression is restricted to the SOP (Huang et al., 1991), ase
is never seen in more than one cell within each proneural
cluster. In double-label studies with antibodies against ac
and ase (Fig. 3F,G), ac expression is detected in proneural
6
M. Brand and others
Fig. 3. asense is expressed in neural precursors and their progeny in the wing imaginal disc. (A) A wing disc from a third instar larva
approximately 12 hours bpf, hybridized with a probe detecting ase mRNA. Some prominent precursors are labelled. Notice that all
staining occurs in isolated cells, although in the dorsal radius, the density of cells is high since many sense organs arise in this region.
(B,C) Confocal microscope pictures of a disc at puparium formation that has been labelled in a double immunofluourescence experiment
to detect expression of A101 (red) and ase (green). ase is detected in the neural precursor cells stained by A101 (Huang et al., 1991).
(D,E) A dividing posterior scutellar (pSC) precursor is seen in the higher magnification view of the SC area showing that both daughter
cells stain with anti-ase. (F–I) Double immunofluorescent detection of ac (red) and ase (green) in a late third instar wing disc about 6
hours bpf. Expression of ac is in proneural clusters (H) in the SC and DC areas, whereas in other areas, ac expression is already confined
to the precursor. (I) ase colocalizes with ac only in the precursors. (F,G) An optical section from the same Z series through a more basal
part of the epithelium. Here, precursor aPA will soon divide; it expresses ase (G), but ac is no longer detectable (F). Abbreviations for
neural precursors: a, anterior; p, posterior; SC, scutellar; DC, dorsocentral; NP, notopleural; SA, supra-alar; tr1, 1st sensillum trichodeum,
WM, wing margin; Teg, tegula; vR, ventral radius; dR, dorsal radius; L3–2, second sensillum campaniformia of the third vein (Huang et
al., 1991).
asense, a neural precursor gene
7
Fig. 4. ase is expressed in neural precursors in imaginal discs and in the larval CNS. (A–C) Whole mounts of leg disc (A) and eyeantennal disc (B) or the larval CNS (C) from wild-type late third instar larvae stained with anti-ase antibodies. (A) A prominent group of
cells at the base of the leg disc corresponds to precursors of the chordotonal organ (arrow). Also stained are the precursors of leg
macrochaetae (arrowhead), others are not in focus. (B) Eye-antennal disc. Expression is detected in groups of cells that are probably
precursors of the Johnston’s organ located in the anlage for the second antennal segment (arrowheads). (C) Larval CNS. Expression is
seen in the neuroblasts and their progeny. Out of focus is the expression in brain lobe in a half circle of neurons.
clusters, whereas ase is expressed only in single cells that
correspond to the SOPs. Close examination of the two acdependent precursors for the dorsocentral bristles (DC
SOPs) revealed that onset of ase expression occurs in a late
stage of cluster development, when ac expression increases
in the single cell that will become the SOP. After ac
expression disappears in the segregated cell prior to divison,
ase expression persists (Fig. 3H,I). Other SOPs of the notum
do not require ac, though they appear to show the same
relative onset of ac and ase (Fig. 3F–I). As in the embryo,
ase remains detectable in both daughter cells during and
after the first SOP division (Fig. 3D,E).
Expression of ase in the leg and antennal discs is also
restricted to precursors that are labelled in A101 (Fig.
4A,B). As described previously (González et al., 1989), ase
is detected in the prominent group of chordotonal organ
precursors at the base of the leg discs (Fig. 4A), but it is
also detected in the SOPs of the bristles. We also detect ase
in large groups of cells in the antennal disc (Fig. 4B) that
probably correspond to the precursors of the Johnston’s
organs (Bryant, 1978), which are thought to be modified
chordotonal organs (McIver, 1985). The expression pattern
in the eye disc is unusual; ase RNA is readily detectable in
the mature part of the eye disc, whereas no protein
expression is detected. In the larval brain, ase protein and
RNA are found in the CNS neuroblasts and in some of their
progeny, and in a prominent horseshoe-shaped group of
cells in the optic lobe anlage (Fig. 4C). These cells most
likely correspond to the proliferation centers that give
rise to the optic lobe (reviewed in Campos-Ortega and
Hartenstein, 1985b). In ase-deficient flies, defects have
been described in this part of the brain (González et al.,
1989).
In summary, the pattern of asense RNA and protein
distribution resembles that of other genes that are activated
in all neural precursors (Vaessin et al., 1991; Bier et al.,
1992).
Proneural mutants affect the expression of
asense
The ase expression pattern shows a close temporal and
spatial relation with the expression of the proneural genes
of the AS-C. We examined ase expression in embryos and
imaginal discs of proneural mutants.
Effect on embryonic expression of ase
daughterless is required for the differentiation of all PNS
sense organs and of most CNS cells. Although the neural
precursors of these neurons form initially, they disappear
shortly thereafter (Brand and Campos-Ortega, unpublished
data; Vaessin, Brand, Jan and Jan, unpublished data). In da
mutants (Df(2L)daKX136), expression of ase is strongly
decreased even though the neuroblasts appear morphologically normal (Fig. 5B,D). This is also true in embryos
lacking the maternal contribution of da (Brand and CamposOrtega, unpublished data). Thus the expression of ase in
neural precursors depends strongly on da.
The proneural genes of the AS-C are required for the
formation of subsets of the PNS and CNS precursors
(Jiménez and Campos-Ortega, 1990). Absence of ac, sc and
l’sc (in Df(1)sc19 embryos) leads to elimination of some, but
not all ase-expressing cells in the PNS (Fig. 5F). The subset
of PNS cells that expresses ase in Df(1)sc19 appears to cor respond to the cells that do not require AS-C for their development, such as the precursors for chordotonal organs
(CHOs), located in the posterior part of the segment (Fig.
5F). These observations indicate that ase is not expressed if
the neural precursors are not formed; that is, expression
requires the prior action of the proneural genes. The ase
expression in the midgut anlage is also eliminated (Fig. 5H).
In mutants of neurogenic genes, neural precursors are
formed at the expense of epidermal precursors (reviewed by
Artavanis-Tsakonas, 1988; Campos-Ortega, 1988; CamposOrtega and Jan, 1991; Simpson, 1990). Consistent with this,
ase is expressed in most cells of the ectoderm in mutant
8
M. Brand and others
Fig. 5. Expression of ase is affected in neurogenic and proneural
mutants. Shown are whole mounts of wild-type (A,C,E,G),
daughterless deficient (B,D), ac, sc, and l’sc deficient (F,H) and
strong neuralized (I) mutant embryos stained with anti-ase
antibodies. (A,C) Expression of ase in a wild-type stage 9 embryo
at low and higher magnification. Epidermal, neural and
mesodermal primordia are indicated by brackets; arrows indicate
neuroblasts, arrowheads indicate ganglion mother cells. (B,D) A
stage 9 da embryo. Although neuroblasts are present, expression
of ase is either severely reduced (arrow) or nearly undetectable
(open arrows). (E) Stage 12 wild-type embryo, ase is expressed in
the PNS primordium (bracket). Many neural precursors and their
progeny stain with anti-ase antibodies. (F) A Df (1) sc19/Y embryo, which is deleted for ac, sc and l’sc, but not for ase. The number of
cells in the PNS primordium is severely reduced; arrowheads point to the cluster of chordotonal organ precursors that continue to stain
with anti-ase. (G,H) CNS and midgut staining in stage 12 wild-type (G) and Df (1) sc19 (H) embryos. Arrows point to midgut cells
stained with anti-ase that are absent (asterisk) in the AS-C mutant embryo. CNS staining in this mutant embryo appears normal.
(I) Embryo with the neurogenic mutation, neuKX9, ase is expressed in most cells of the neurogenic and PNS primordium.
asense, a neural precursor gene
9
Fig. 6. ase expression in late third instar imaginal discs of AS-C mutants. Shown are whole-mount imaginal discs mutant for ac and sc
(sc10-1/Y) (A–D); ac (ac3/ac3) (E); or a combination of strong viable emc alleles (emcpel/Df (3)emcE12) (F). Discs are stained with anti-ase
antibodies (A,B,D–F) or stained for ase RNA (C). (A) Eye-antennal disc. Expression remains in the putative Johnston’s organ precursors
in the antennal portion. (B) Wing disc. Most ase expression is eliminated compared to the wild-type discs shown in Fig. 3, with the
exception of 1–2 cells (arrow) located in the dorsal radius that may correspond to precursors of a chordotonal organ. (C) Everted wing
pouch of a sc10-1 mutant wing disc 10–12 hours apf, hybridized with a probe that detects ase RNA. Expression is weakly detected in a
row of large single subepidermal cells (arrow), precursors for the row of stout bristles — the only bristles formed in the wing margin of
sc10-1 flies. (D) Leg disc: expression of ase is normal in the cluster of chordotonal organ precursors, and in the few other remaining
precursors of macrochaetae. (E) In an ac mutant wing disc, expression of ase is eliminated in the area of the DC precursors (bracket).
Expression is still detected in other precursors, such as those of the aPA, SC, WM and dorsal radius dR, that mostly give rise to scdependent sense organs. Other abbreviations as in Fig. 3 legend. (F) Expression in a late third instar larval wing disc mutant for emc,
which causes ectopic expression of ac and sc. In the notum area, ase antibody staining is detected in several additional precursors,
indicated by the bracket and asterisk.
embryos that lack neuralized function (Fig. 5I) or Notch
function (not shown).
Effect on ase expression in the imaginal discs
We have examined ase expression in discs of the sc10-1
mutant, which do not express ac or sc and lack most of the
external sense organs, and those of the mutant In(1)ac3,
which express sc but not ac and have only a few of the sense
organs missing. In both mutants, ase is expressed in the precursors of most or all remaining sense organs (Fig. 6), as
was found in the embryo.
Wing discs derived from sc10-1 mutant larvae are devoid
of ase-staining cells, with the exception of 1-2 cells located
in the dorsal radius (Fig. 6B). No sense organ was detected
externally in sc10-1 mutant wings in this region. It is possible
that these ase-expressing cells give rise to an internal chordotonal organ that is known to reside in the radius (Miller,
1950). In the leg discs, the group of chordotonal organ precursors clearly expresses ase, as does a small subset of the
precursors located on the more distal segments of the leg
(Fig. 6D), consistent with the presence of a few bristles on
sc10-1 adult legs (Lindsley and Grell, 1968; Held, 1991). In
the antennal disc, the large group of putative Johnston’s
organ precursors still express ase (Fig. 6A). The majority of
cells that still express ase in sc10-1 discs may therefore be of
the chordotonal type. For technical reasons, we have not
10
M. Brand and others
been able to examine ase expression in the precursors for
the stout row of mechanosensory bristles at the wing margin,
which is not affected in sc10-1. These precursors arise around
10 hours after puparium formation. In one case, we could
detect expression of ase RNA in a row of large subepidermal cells, which we presume to be the stout row precursors
(Fig. 6C). This indicates that ase expression in these cells is
independent of ac and sc, and is consistent with the dependence of these sense organs on ase itself (Jarman et al.,
1993).
Flies lacking a functional ac gene (In(1)ac3) retain all but
the DC macrochaetae and express sc in their SOPs (Cubas
et al., 1991; Skeath and Carrol, 1991). We find this to be
reflected in the expression of ase in In(1)ac3 wing discs:
expression is lacking in the DC area, but is unaffected in the
other SOPs (Fig. 5E). In contrast, flies carrying a combination of strong viable emc alleles, emcpel/Df(3)emcE12 show
a large number of additional bristles on the notum as a consequence of ectopic expression of AS-C genes (Cubas et al.,
1991; Skeath and Carrol, 1991). Additional ase-expressing
SOPs are observed in the notum portion of wing discs of this
genotype (Fig. 5F).
In summary, ase activation only occurs upon SOP
formation and both events may therefore be part of the same
process. Also, ase-expressing imaginal SOPs can apparently
be subdivided into four groups: those that require ase (stout
row bristles), those that require ac (DCs), those that require
sc (most other macrochaetae) and those that are independent
of AS-C (such as the chordotonal organs). It appears that in
most cases ase is temporally downstream both of the AS-C
and of other proneural genes required for sense organ
formation.
Ectopic asense expression leads to appearance of
duplicated and ectopic sense organs
Ectopic expression of the proneural gene sc can result in
placement of duplicated and ectopic bristles (Rodríguez et
al., 1990). Since ase expression differs from that of
proneural genes, we asked if this leads to a different
phenotype under conditions of ectopic expression, when ase
is under the control of a heat-shock promoter (hs-ase, Fig.
7).
Construction and testing of hs-ase
Stocks carrying 2 or 4 copies of hs-ase were constructed
(Materials and Methods). Extracts from heat-treated
embryos of these stocks contain an induced protein of the
expected size (Fig. 7D) and expression is strong and ubiquitous in heat treated, but not untreated, imaginal discs (Fig.
7E). Although difficult to quantitate, the levels of protein
achieved under these conditions appear to be much higher
than wild type.
hs-ase in wild-type background
We restrict our analysis to the larval stage, because of the
greater temporal and spatial resolution. Ectopic expression
of ase after heat shock resulted in placement of many additional bristles in various parts of the fly’s body (Fig. 8) in a
manner similar to that described after ectopic expression of
sc (Rodríguez et al., 1990). The additional bristles are
accompanied by corresponding internal parts of a sense
organ, as determined by 22C10 staining of sensory neurons
in adult nota (Fig. 8H). As reported for sc (Rodriguez et al.,
1990), the type of sense organ formed, e.g. macrochaetae,
microchaetae or campaniform sensilla, varies with the time
of heat shock and the location in the fly (Table 1; Figs 8, 9).
For instance, a heat treatment before puparium formation
(bpf) typically results in placement of macrochaetae on the
notum but not on many other tissues, such as the ventral
abdomen (Fig. 8B). After puparium formation (apf),
however, numerous bristles are formed on the ventral
abdomen, but no additional ones are found on the notum
(Fig. 8D).
After heat-shock induction of ase, the extra macrochaetae
on the notum appear around the normal macrochaetae. To
test whether some of these limitations of bristle induction
might be due to an insufficient level/stability of ectopically
produced protein, we produced transformants carrying ase
under the control of a promoter containing yeast Gal4binding sites, and crossed a heat shock-Gal4 construct into
these flies (Fig. 7). In flies carrying both of these constructs
(hs-Gal4-ase), penetrance and expressivity of the phenotype
appeared to be increased, i.e. more bristles appear to be
placed at a given time and position, making it in some cases
difficult to assign the ectopic bristle to a cognate position
(Fig. 8C); this correlates with a much higher level of protein
as determined in western blots (Fig. 7D). In contrast, the
spatial and temporal sensitivity of the tissue does not appear
to be altered, indicating that the amount of ase product
cannot be the only factor that is limiting the ability of cells
to assume a neural fate.
Extra sense organs are induced by ectopic
expression of asense during and shortly after the
period for normal neural precursor formation
In order to determine the sensitive period for ectopic ase
expression, we staged groups of larvae carrying two copies
of hs-ase and subjected them to heat treatment at different
times. It appears from these studies that the extra sense
organs for a given position form at specific time windows
(Table 1; Fig. 9). The clearest case is that of the DC bristles:
the maximum overproduction of these organs is reached
during and after the period of time when their precursors
normally form (Romani et al., 1989; Skeath and Carroll,
1991; Cubas et al., 1991).
An intact DNA-binding domain is necessary to
produce ectopic sense organs
Previous studies indicate that bHLH proteins dimerize via
their HLH domain and that the basic domain preceding the
HLH domain is required for sequence specific DNA recognition (Murre et al., 1989; Davis et al., 1990). To distinguish
between the possibility that the ase product acts directly by
binding to DNA to activate the genetic program for bristle
formation and the alternative that it titrates negative regulatory factors of ac and sc such as emc by dimerization
(Garrell and Modolell, 1990; Ellis et al., 1990; van Doren et
al., 1992), we tested in our heat-shock experiments
mutations of ase that abolish the DNA-binding moiety but
retain the dimerizing HLH domain. Three different
mutations were generated: a double mutant that changes two
arginine residues of the basic domain to glycine (these
asense, a neural precursor gene
11
Fig. 7. An intact basic domain is required
for ectopic sense organ formation after
heat-shock induction of ase. (A–C) A
summary of the structure of different ase
(A,B) and l’sc (C) heat-shock constructs
and their ability to induce extra bristles
after a heat shock. (A) Top: Structure of
the hs-ase construct. Expression is
controlled via a heat-shock promoter, hsp.
The arrow symbolizes the transcription
start site of the ase gene under the control
of the heat-shock promoter. The basic
domain (bd) required for DNA binding and
the helix-loop-helix domain (HLH)
required for protein interaction are
indicated by boxes. Heat shock of animals
carrying hs-ase leads to abundant protein
expression (shown in D) that is
homogeneous and nuclear, as shown in the
wing disc in (E). Heat shock during larval
and pupal stages results in bristle
formation, indicated by a plus in the
column ‘extra bristles’, as shown in more
detail in Figs 8 and 9. (A) Bottom:
Structure of the hs-ase bd− construct,
where the basic domain has been deleted.
This construct is no longer able to induce
bristle formation, although a protein of the
correct size is still produced. (B) Structure
and performance of a series of hs-Gal4-ase
constructs. Expression of the yeast
regulatory gene Gal4 is controlled by the
heat-shock promoter. Gal4 protein binds,
after induction through heat shock, to
binding sites (indicated as black boxes)
that are upstream of the start site of
transcription in the pUAST vector. Various
ase wild-type and mutated coding regions were cloned into this vector. The phenotype after heat shock of the wild-type hs-Gal4-ase
construct is stronger than for the hs-ase construct shown in A, although not qualitatively different (see text and Fig. 8). Asterisks in the
second construct indicate the exchange of two Arginine residues that are crucial for DNA binding in myogenic bHLH proteins; this
abolishes the ability to induce extra sense organs, as in the other constructs shown with a deleted basic domain, or a deleted bHLH
domain. (C) Structure of the hs-Gal4-l’sc construct. Symbols are as above; the strong bristle phenotype is indistinguishable from hs-Gal4ase bearing flies after heat shock in the presence of a hs-Gal4 construct (see Fig. 8). n.d., protein level not determined. (D) Western blot
of a 12% SDS PAGE gel to detect wild-type and mutant aseproteins in total protein extracts from heat-shocked embryos at different
times. Each lane contains the protein equivalent of five embryos from a mix of embryos containing different copy numbers of hs-ase or
hs-Gal4-ase. Lane 1: one fourth of the embryos carried hs-Gal4-ase; no heat shock (control). Lane 2: hs-Gal4-ase, 1 hour after heat
shock, Lane 3: hs-Gal4-ase, 3 hours after heat shock. Lane 4: hs-ase, 4 copies per embryo, 3 hours after heat shock. In spite of the 16-fold
higher average copy number in this line, the intensity of the ase band is lower than in the other lanes. Lane 5: one fourth of the embryos
carried the construct with a deleted basic domain (hs-Gal4-ase-bd−), 3 hours after heat shock. Lane 6: One fourth of embryos carrying the
point mutant of the basic domain (hs Gal4-ase PM). Lane 7: One fourth of the embryos carrying the deletion of both basic and HLH
domain (hs Gal4-ase bHLH−). Proteins are stable and of the size predicted by the ase open reading frame (54×103 Mr for the wild-type
protein (González et al., 1989) and 47×103 Mr for the bHLH deletion; the other mutant proteins are of expected size).
residues are crucial for MyoD to bind to DNA, Davis et al.,
1990); a deletion of the six amino acids of the basic domain
and a larger deletion of both the basic and the HLH domains
(Fig. 7). Bacterially produced mutant proteins were no
longer able to bind DNA in a bandshift assay (A. Jarman,
unpublished), although normal ase does so when produced
under the same conditions (Jarman et al., 1993). All three
mutant versions were tested for their function in the fly with
the hs-Gal4 system; the double mutant and the basic domain
deletion were also tested under direct control of the heatshock promoter. Western blot analysis of embryonic extracts
indicates that the mutant ase proteins are of the expected size
and that they are stable under the conditions of the heatshock experiment (Fig. 7D). Larvae carrying the mutant
constructs were then tested under standard heat-shock conditions, but the resulting flies showed no additional bristles.
Thus, the DNA-binding domain of ectopically expressed ase
is necessary for the production of additional bristles.
Gene activation after ectopic expression of ase
Given its ability to direct ectopic sense organ formation, we
asked if hs-ase causes activation of other genes expressed
in neural precursors, such as the A101 marker (Huang et al.,
1991). Larvae carrying two copies of hs-ase and one copy
12
M. Brand and others
Fig. 8. Ectopic sense organs are formed after ectopic ase expression. Shown are dissected body parts of flies that have been heat treated
during larval or pupal stages. (A) Notum of a wild-type fly, showing the regular arrangement of macro- and microchaetae. (B) Notum of a
fly carrying two copies of hs-ase that was heat shocked around 12 hours bpf. Additional macrochaetae appear in the dorsocentral area
(DC) and the scutellar area as a consequence of ectopic ase expression. (C) Notum of a fly carrying one copy of hs-Gal4-ase. The
phenotype in the notum is more severe than in B. Most other tissues of this fly are indistinguishable from the wild type. (D) Abdominal
pleura of a fly carrying 1 copy of hs-Gal4-ase after a heat shock 12-24 hours apf. Except for the central sternites, this area is normally
devoid of bristles. After heat shock, numerous microchaetae densely cover the area; macrochaetae are not observed. The notum of this fly
showed no extra sense organs. (E–G) Induction of ase can replace proneural gene function. (E) The notum of a fly lacking functional ac
and sc protein (sc10-1) is devoid of macrochaetae and microchaetae. (F,G) In the presence of one copy of hs-ase, sense organ formation
can be restored in sc10-1 flies. Macrochaetae are formed if the heat shock occurred during late third instar (F), whereas microchaetae are
formed if the heat shock occurred during pupal life (G). (H) 22C10 staining of an adult notum of a fly that showed additional hs-aseinduced scutellar macrochaetae externally. External portions of ectopic macrochaetae are accompanied by corresponding internal sensory
neurons and their axons (arrowhead). (I) Distal portion of a wild-type wing. Roman numerals indicate the vein number. Vein II is devoid
of sense organs, whereas vein III shows a few campaniform sensilla in reproducible locations. (J) After heat-shock induction of ase at 8
hours apf, numerous microchaetae-like sense organs are formed preferentially along vein II, but not necessarily associated with the vein
(open arrow). Earlier during the competence period for this phenotype (see Table 1, Fig. 9), macrochaetae-like sense organs appear,
though they also seem to have less tendency to form along the veins. Additional sense organs of a different type, campaniform sensilla,
are formed mainly along vein III. (K) A similar result is seen with hs-ase in a sc10-1 mutant background. Ectopic sense organs form
(arrowheads) and are preferentially, though not always, located along the wing veins.
asense, a neural precursor gene
13
Table 1. Time course of formation of ectopic sense organs after heat-shock induction of asense
−36
Time (hours) relative
to puparium
formation
−30
−24
−18
−12
−6
2
4
8
12
16
20
24
Affected sense
organ or tissue
Macrochaetae
SC
DC
PA
PS
SA
ANP
11 (8)
4 (3)
9 (7) 7.7 (2)
42 (31) 38.5 (10) 70 (28)
4 (3) 15.4 (4) 15 (6)
3 (2) 15.4 (4)
7.7 (2) 15 (6) 58 (22)
76 (29)
MsCh
MicNo
EWSO
ScMic
WVD
AbPl
Number of body
halves examined
95 (36) 100 (28) 100 (22)
11 (4) 71 (20) 91 (20)
95 (36) 100 (28) 100 (22)
9 (2)
144
42
70
74
26
40
38
28
22
100 (16) 94 (34) 46 (21)
100 (16) 28 (10)
100 (16)
88 (14) 3 (1)
40 (14) 43 (20)
92 (33) 100 (46) 65 (30)
16
36
46
46
The percentage of body halves showing formation of extra sense organs of the indicated type following a heat-shock treatment is given as a function of
time relative to puparium formation. The number of body halves with extra sense organs in each case is given in parentheses. The total number of
larvae/pupae scored at each time point is indicated on the bottom. We attempted to score only well-isolated areas where macrochaetae arise. Even so,
additional chaetae tended to avoid the location of the extant bristles and were often located in between extant chaetae (see Fig. 8B), as was observed for
heat-shock induction of sc (Rodríguez et al., 1990). For this reason, closely situated chaetae of the same area were scored together, e.g. aDC and pDC,
were scored as DC. MsCh: the mesopleural chaetae are microchaeta-like. This area is normally free of chaetae. MicNo: increased microchaetae on the
notum. EWSO: extra wing sensory organs. This group includes both chaetae and campaniform sensilla (see Fig. 8I–K). Early in the period, chaetae tended
to be more macrochaetae like and were found in vein and intervein areas, and sometimes additional wing vein material was observed. Late in the period,
chaetae tended to be microchaetae like and located along the veins, mostly vein II (Fig. 8J,K). WVD: small deltas on the tip of wing veins. AbPl:
microchaetae-like sense organs on abdominal pleura (Fig. 8D), an area normally devoid of chaetae.
of A101 were subjected to a 30 minute heat shock at 39°C
and wing imaginal discs were dissected 12 hours after the
treatment. In the DC area, additional A101-positive cells
were formed in 7 of 19 discs as a consequence of the heat
shock (Fig. 10D). Thus, although A101 expression normally
precedes that of ase, its expression is activated in the ectopic
neural precursors induced by hs-ase.
Within five hours of heat-shock induction of ase, we also
find very strong ectopic expression of ac in all cells of the
notum (Fig. 10A,B). Thus, hs-ase is capable of activating at
least one proneural gene. Moreover, although hs-ase induces
relatively few bristles, it can activate ac expression in many
more cells of the wing disc.
Ectopic expression of ase can substitute for missing
proneural gene function
The observation that ac is activated as a consequence of hsase raises the possibility that hs-ase induces ectopic sense
Fig. 9. Tissues respond at characteristic times to ectopic ase
expression. Summary of the data presented in Table 1.
(A) Formation of chaetae in the DC area as an example of the
temporally restricted effects of ase expression after heat-shock
induction. The percentage of heminota with additional DC chaetae
varies with the time of heat-shock induction of ase relative to
puparium formation. The window of sensitivity begins around the
time of normal onset of ase expression in the DC SOPs, shortly
after the expression of neu and of the proneural genes ac and sc;
competence ends at puparium formation, when the DC precursors
start to divide (Mitosis) (times based on Huang et al., 1991; Cubas
et al., 1991 and Skeath and Carroll, 1991). (B) Same time scale as
in A. Bars indicate the window of time during which extra sense
organs of the type indicated on the left are induced. Given that
heat-shock time points are only at the times marked by the dashed
lines, the sensitive period for each type of sense organ may be
longer than that indicated. Notice that additional macrochaetae
(top group) form mostly before puparium formation, the normal
time of their development. SC, scutellar; DC, dorsocentral; PA,
postalar; PS, presutural; SA, supraalar; ANP, anterior notopleural;
MsCh, chaetae on mesopleura; MicNo, increase of microchaetae
on the notum; EWSO, extra wing sensory organs; ScMic, scutellar
microchaetae; WVD, deltas at wing vein tips; AbPl, microchaetae
on abdominal pleura. See Table 1 for more details.
14
M. Brand and others
8F, macrochaetae; Fig. 8G, microchaetae). Sense organs
were also formed after heat-shock induction of ase in flies
carrying a deficiency for ac and sc, (Df(1)y3PLscS1R/Y; hsase/+, not shown). Thus, ectopic expression of ase can
replace the missing proneural function of ac and sc, the
genes that are normally required for the formation of sense
organs. It is still possible, however, that other proneural
genes are activated in this case.
Ectopic expression of l’sc also leads to extra
bristle formation
Having observed that removal of ac and sc did not abolish
the ability of hs-ase to induce bristles, we asked whether
ectopic expression of l’sc, the remaining proneural gene in
the AS-C, also induces bristles. l’sc is not required for the
formation of the bristles on the notum (García-Bellido and
Santamaria, 1978), nor could we detect any l’sc RNA
expression in wild-type wing discs (not shown). Nevertheless, supernumerary bristles are encountered in hs-Gal4-l’sc
flies. Although not examined in detail, the temporal and
spatial characteristics of sense organ induction due to hsGal4-l’sc closely resemble those due to hs-Gal4-ase. Thus,
overexpression of ase or l’sc results in gain-of-function phenotypes that differ from what would be expected on the basis
of their loss-of-function phenotypes.
DISCUSSION
Fig. 10. Activation of neural markers after heat-shock induction of
ase. Whole-mount wing discs are stained with anti-ac and anti-ase
(A,B) or anti-β galactosidase antibodies for detection of A101
expression pattern (C,D). (A,B) ac expression is activated
ectopically after heat-shock induction of ase. Larvae with two
copies of hs-ase were heat shocked, dissected 5 hours after the
heat shock, and stained in a double-label immunofluorescence
experiment with anti-ac and anti-ase. (A) High expression of ac is
induced in ectopic sites; compare to Fig. 3H for a wild-type
pattern of ac. (B) Ectopic expression of ase is already weakening
at this stage. (C,D) The SOP marker A101 is activated in ectopic
SOPs after a hs-ase. (C) Wing disc from a late third instar larva of
the A101 insertion line stained for β-galactosidase expression.
The DC area, indicated by a bracket, contains the precursors for
aDC and pDC. (D) β-galactosidase staining of a disc from a larva
with two copies of hs-ase and one copy of the A101 insertion, 12
hours after ase induction. A pair of additional A101 staining cells
is observed in the DC area (arrows).
organs indirectly via the activation of proneural genes. To
test for this possibility, we placed one copy of hs-ase into
the sc10-1 mutant. The nota of sc10-1 flies are completely
devoid of sensory organs (Fig. 8E), and no expression of ase
(see above; Fig. 6B) or A101 (Cubas et al., 1991) is detected.
After heat treatment, however, sc10-1; hs-ase/+ flies show
formation of sense organs. Again, heat-shock treatment at
different times results in placement of different bristles (Fig.
ase is expressed in all neural precursors
Previous analyses of in situ hybridization to sections
indicated that the pattern of expression of ase RNA differs
from those of the proneural genes, but the identity of the
cells involved remained largely uncharacterized (Alonso
and Cabrera, 1988; González et al., 1989). We have found
that ase RNA and protein are expressed in neural precursors
of the CNS and the PNS in the embryo and in imaginal discs.
In addition, ase is expressed in at least one nonneural cell
type, the imaginal midgut precursors, where it also closely
follows the expression of the proneural genes. The
expression of ase in neural precursors requires the function
of daughterless, and it is downstream of AS-C function in
at least a temporal sense. This is consistent with analysis of
the ase promoter, which indicates a potentially direct action
of AS-C/da proteins on ase transcription (Jarman et al.,
1993). Compared with the proneural genes, ase differs in its
later onset of expression during precursor development, the
exclusive expression in all neural precursors and the
continued expression after the first precursor division. These
properties of ase suggest that it belongs to the group of
neural precursor genes (Vaessin et al., 1991; Bier et al.,
1992). In the following, we discuss differences and commonalities between ase and the proneural genes of the ASC.
Proneural clusters and ase expression
We were unable to detect ase expression at the proneural
cluster stage, either in the embryonic CNS, PNS or imaginal
disc tissue. It seems unlikely that ase expression in proneural
clusters has been overlooked for the following reasons. First,
at stages and under conditions where clustered expression
asense, a neural precursor gene
of the proneural ac, sc or l’sc genes are clearly detectable,
we consistently detect ase in only single cells. Second, the
pattern of RNA and protein expression are identical, thus
making it unlikely that only a specific form of ase is recognized by the antibody, as may be the case for an antibody
directed against the l’sc protein (Cabrera, 1990). Third, the
expression of an ase-lacZ fusion gene is also only detected
in single cells (Jarman et al., 1993). However, of those SOPs
that depend on ase for their proper formation (DamblyChaudière and Ghysen, 1987; Jarman et al., 1993), we can
reliably identify only the A cell that gives rise to lk/h3, and
ac expression in the proneural cluster associated with this
sense organ is very faint and transient (Ruiz-Gómez and
Ghysen, 1993). Thus, although we do not detect ase
expression in any of the proneural clusters, we cannot rule
out that we have overlooked very transient expression in this
particular case.
If ase expression is indeed absent in proneural clusters for
ase-dependent sense organ precursors, how might one
reconcile this paradoxical observation? Formation of asedependent sense organs is somewhat suppressed if ac, sc and
l’sc function are eliminated (in Df(1)sc19 embryos, DamblyChaudière and Ghysen, 1987, or in the stout row of the adult
wing margin, M. Guo, M. B., A. P. J., L. Y. J. and Y. N. J.,
unpublished). One or more proneural genes may be
expressed in the proneural clusters of ase-dependent SOPs
and may be involved in the selection of neural precursors
from these clusters of cells; ase expression in the selected
precursors may then be essential for their further development. For different precursors, the function of ase could be
supplemented to different extents by the expression of the
proneural genes in the precursor. Loss of ase, if not compensated for completely, may then result in the formation of
aberrant bristles, as observed in the case of the malformed
bristles of the wing margin (Jarman et al., 1993), or it may
lead to not fully penetrant absence of the sense organ, as has
been found in the few embryonic sense organs.
Over-expression of asense gives rise to a gain-offunction phenotype characteristic of proneural
genes
Our results show that ectopic expression of ase can promote
formation of sense organs and that this expression is
effective only at the time when sense organs normally would
form. In addition, ectopic sense organs are induced by hsase in the same tissues that respond to a heat-shock
induction of sc (Rodríguez et al., 1990). Also, induction of
ase restores bristles at or around their normal positions on
the notum in the absence of ac and sc, and can thus functionally replace the proneural ac and sc genes. Consistent
with this notion, we find that hs-ase leads to apparently inappropriate activation of ac and A101(neuralized), two genes
that are normally active in early stages of precursor
formation, prior to the normal activation of ase. Thus, under
conditions of overexpression ase acts like a proneural gene.
Similarly, l’sc is not normally required for development of
adult sense organs on the notum (García-Bellido and Santamaria, 1978), and it is not normally expressed there (M.
Brand, unpublished). Nevertheless, hs-l’sc behaves like hsase or hs-sc with respect to temporal and spatial sensitivity
of the responding tissue. This demonstrates the ability of
15
different AS-C functions to replace one another under the
gain-of-function condition. Some functional redundancy has
been observed in the analysis of loss-of-function phenotypes
of the AS-C (Jiménez and Campos-Ortega, 1987; DamblyChaudière and Ghysen, 1987), but our gain-of-function
results suggest complete interchangeability of the AS-C
members in initiating sense organ development.
Although ase, l’sc or sc overexpression can place sense
organs on some parts of the body normally devoid of sense
organs, such as the thoracic pleura, parts of the wing blade
or the ventral abdomen, the placement of sense organs is far
from random, pointing to constraints on neural competence.
Some of these constraints are known. It is likely that the
extramacrochaetae (emc) product needs to be reduced in
order to allow ase protein to become effective. emc is known
to be a negative regulator of AS-C gene function (Moscoso
del Prado and García-Bellido, 1984; García-Alonso and
García-Bellido, 1986; Marí-Beffa et al. 1991). This is consistent with the finding that the emc RNA (Cubas and
Modolell, 1992; Van Doren et al., 1992) and protein (Brand
et al, unpublished) appear to be low in areas of active neurogenesis. Evidence from in vitro experiments suggests that
emc can interfere with binding of ac or sc to da protein (Van
Doren et al., 1992). However, simply raising the level of ase
protein through the use of the more efficient hs-Gal4-ase
constructs does not overcome such temporal or spatial constraints; it only appears to increase penetrance and expressivity of sense organ formation. We are currently testing the
possibility that da becomes limiting under the conditions of
AS-C over expression.
Comparison of the neurogenic and the myogenic
networks of bHLH proteins
Our studies of members of bHLH proteins in fly neurogenesis indicate that a detailed knowledge of the pattern of
expression and of the loss-of-function phenotype are
important for interpreting the gain-of-function phenotypes.
Clearly, under conditions of ectopic expression, ase and l’sc
can activate the neural fate for cells that do not normally
require their function. A similar situation was noticed in the
case of the bHLH gene hairy, where misexpression leads to
artifactual interference with the sex determination function
of sc (Parkhurst and Ish-Horowicz, 1991). This behaviour
could perhaps be a general property of bHLH proteins. In
vertebrate myogenesis, all of the known myogenic factors
with bHLH domains are able to switch cultured cells to a
myoblast fate, and thus appear to be interchangeable like the
genes of the AS-C. These genes have thus been called
‘master regulators’ of myogenesis (reviewed by Olson,
1990; Weintraub et al., 1991). Nevertheless, like the AS-C
in Drosophila neurogenesis, the analysis of the distribution
of the myogenic factors in vivo indicates temporally and
spatially distinct patterns of expression in skeletal muscle
primordia (Sassoon et al., 1989; Bober et al., 1991; Ott et
al., 1991). Moreover, targeted gene disruption of some
myogenic genes gives surprisingly mild effects: indeed, the
results vary from abnormal rib development in the case of
myf-5 disruption to normal muscle development in the case
of MyoD disruption (Braun et al., 1992; Rudnicki et al.,
1992). Similarly, body wall muscles that normally express
the MyoD like gene hlh-1 in C. elegans can differentiate in
16
M. Brand and others
the absence of that gene (Chen et al., 1992). Thus, as a
group, the myogenic bHLH proteins are important for
muscle development. However, assigning the role of a
‘master regulatory gene’ to an individual member solely
based on gain-of-function phenotype can be misleading.
In fly neurogenesis, the loss-of-function phenotypes and
the distribution of most of the gene products are known. As
is the case for the myogenic regulators, the expression
patterns and gain-of-function phenotypes of ac, sc, l’sc and
ase are more widespread than the requirement for these
genes, as evidenced by the loss-of-function phenotypes.
Likewise, differences in temporal onset of normal
expression such as observed between the proneural genes
and ase seem not to affect the similar performance of these
genes under the gain-of-function condition. Members of the
AS-C have the same or very similar protein properties, but
differ in their mode of regulation. This suggests that the task
of the ‘master regulator’ of neurogenesis is shared by several
genes rather democratically.
We would like to thank our former and present colleagues in the
Jan lab, as well as Christine Dambly Chaudière and Alain Ghysen,
for stimulating discussions and advice. Special thanks goes to
Harald Vaessin, who also made the pUAST-l’sc construct. Juan
Modolell kindly sent us the genomic ase clone, A. Brand and N.
Perrimon made the pUAST vector available to us and Chris Turk
prepared the ase peptide. Thanks also to Larry Ackermann and
Sandy Barbel for technical assistance and for preparing the figures,
to Tim Mitchison for allowing us to use the confocal microscope.
Thanks to Nadean Brown, Sean Carrol for flies and ac antibodies,
Y. Hiromi and P. Gergen for blue balancers, and Ed Giniger for
hs-Gal4 flies. This work was supported by a postdoctoral fellowship from the Deutsche Forschungsgemeinschaft to M. B., a NATO
postdoctoral fellowship to A. P. J, and the Howard Hughes Medical
Institute. L. Y. J. and Y. N. J. are Howard Hughes Medical Institute
investigators.
REFERENCES
Alonso, M. C. and Cabrera,C. V. (1988). The achaete-scute complex of
Drosophila melanogaster comprises four homologous genes. EMBO J. 7,
2585-2591.
Artavanis-Tsakonas, S. (1988). The molecular biology of the Notch locus
and the fine tuning of differentiation in Drosophila. Trends in Genetics 4,
95-100.
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., Vaessin, H., Younger-Shepherd, S., Jan, L. Y. and Jan, Y. N.
(1988). deadpan, an essential pan-neural gene in Drosophila, encodes a
helix-loop-helix protein with a structure similar to the hairy product.
Genes Dev. 6, 2137-2151.
Blochlinger, K., Jan, L. Y. and Jan, Y. N. (1991). Transformation of
sensory organ identity by ectopic expression of Cut in Drosophila. Genes
Dev. 5, 1124-1135.
Bober, E., Lyons, G. E., Braun, T., Cossu, G., Buckingham, M., and
Arnold, H. H. (1991). The muscle regulatory gene, Myf-5, has a biphasic
pattern of expression during early mouse development. J. Cell Biol. 113,
1255-1265.
Boulianne, G. L., de la Concha, A., Campos-Ortega, J. A., Jan, L. Y.,
and Jan, Y. N. (1991). The Drosophila neurogenic gene neuralized
encodes a novel protein and is expressed in precursors of larval and adult
neurons. EMBO J. 10, 2975-2983.
Braun, T., Rudnicki, M. A., Arnold, H. H. and Jaenisch, R. (1992).
Targeted inactivation of the muscle regulatory gene Myf-5 results in
abnormal rib development and perinatal death. Cell 71, 369-382.
Bryant, P. (1978). Pattern formation in imaginal discs. In The Genetics and
Biology of Drosophila (ed. M. Ashburner and T. R. F. Wright). pp. 229335. New York: Academic Press.
Cabrera, C. V., Martínez-Arias, A. and Bate,M. (1987). The expression
of three members of the achaete-scute gene complex correlates with
neuroblast segregation in Drosophila. Cell 50, 425-433.
Cabrera, C. V. (1990). Lateral inhibition and cell fate during neurogenesis
in Drosophila: the interactions between scute, Notch and Delta.
Development 109, 733-742.
Cabrera, C. V. and Alonso, M. C. (1991). Transcriptional activation by
heterodimers of the achaete-scute and daughterless gene products of
Drosophila. EMBO J. 10, 2965-2973.
Cabrera, C. V. (1992). The generation of cell diversity during early
neurogenesis in Drosophila. Development 115, 893-901.
Campos-Ortega, J. A. (1988). Cellular interactions during early
neurogenesis in Drosophila melanogaster. Trends in Neurosci. 11, 400405.
Campos-Ortega, J. and Hartenstein, V. (1985a). The Embryonic
Development of Drosophila melanogaster. New York: Springer-Verlag.
Campos-Ortega, J. A., and Hartenstein, V. (1985b). Development of the
Nervous System. In Comprehensive Insect Physiology, Biochemistry and
Pharmacology, Vol 5 (ed. L. I. Gilbert and D. A. Kerkut), pp. 49-84. New
York/London: Pergamon Press.
Campos-Ortega, J. A. and Jan,Y. N. (1991). Genetic and molecular basis
of neurogenesis in Drosophila melanogaster. Annu. Rev. Neurosci. 14,
399-420.
Campuzano, S. and Modolell, J. (1992). Patterning of the Drosophila
nervous system: the achaete-scute gene complex. Trends Genet. 8, 202208
Caudy, M., Vässin, H., Brand, M., Tuma, R., Jan, L. Y. and Jan, Y. N.
(1988). daughterless, a gene essential for both neurogenesis and sex
determination in Drosophila, has sequence similarities to myc and the
achaete-scute complex. Cell 55, 1061-1067.
Chen, L., Krause, M., Draper, B., Weintraub, H. and Fire, A. (1992).
Body wall muscle formation in Caenorhabditis elegansembryos that lack
the MyoD homolog hlh-1. Science 256, 240-243.
Cubas, P., and Modolell, J. (1992). The extramacrochaetae gene provides
information for sensory organ patterning. EMBO J. 11, 3385-3393.
Cubas, P., de Celis, J.-F., Campuzano, S., and Modolell, J. (1991).
Proneural clusters of achaete-scute expression and the generation of
sensory organs in the Drosophila imaginal wing disc. Genes Dev. 5, 9961008.
Dambly-Chaudière, C. and Ghysen, A. (1987). Independent subpatterns
of sense organs require independent genes of the achaete-scute complex
in Drosophila larvae. Genes Dev. 1, 297-306.
Dambly-Chaudière, C., Ghysen, A., Jan, L. Y. and Jan, Y. N. (1988).
The determination of sense organs in Drosophila: interaction of scute
with daughterless. Roux’s Arch. Dev. Biol. 197, 419-423.
Davis, R. L., Cheng, P.-F., Lassar, A. B. and Weintraub, H. (1990). The
MyoD DNA binding domain contains a recognition code for musclespecific gene activation. Cell 60, 733-746.
Ellis, H. M., Spann, D. R. and Posakony, J. W. (1990).
extramacrochaetae, a negative regulator of sensory organ development in
Drosophila, defines a new class of Helix-Loop-Helix Proteins. Cell 61,
27-38.
García-Alonso, L. A., and García-Bellido, A. (1986). Extramacrochaetae,
a trans-acting gene of the achaete scute complex involved in cell
communication. Roux’s Arch Dev. Biol. 197, 328-338.
García-Bellido, A. (1979). Genetic analysis of the achaete-scute system of
Drosophila melanogaster. Genetics 91, 491–520.
García-Bellido, A. and Santamaria, P. (1978). Developmental analysis of
the achaete-scute system of Drosophila melanogaster. Genetics 88, 469486.
Garrell, J. and Modolell, J. (1990). The Drosophila extramacrochaetae
locus, and antagonist of proneural genes that, like these genes, encodes a
Helix-Loop-Helix protein. Cell 61, 39-48.
Ghysen, A., and Dambly-Chaudière, C. (1989). Genesis of the
Drosophila peripheral nervous system. Trends Genet. 5, 251-255.
Ghysen, A. and O’Kane, C. (1989). Neural enhancer-like elements as
specific cell markers in Drosophila. Development 105, 35-52.
González, F., Romani, S., Cubas, P., Modolell, J. and Campuzano, S.
(1989). Molecular analysis of the achaete-scute complex of Drosophila
melanogaster, and its novel role in optic lobe development. EMBO J. 8,
3553-3562.
asense, a neural precursor gene
Hartenstein, V., and Campos-Ortega, J. A. (1986). The peripheral
nervous system of mutants of early neurogenesis in Drosophila
melanogaster. Roux’s Arch. Dev. Biol. 195, 210-211.
Hartenstein, A. Y., Rugendorff, A., Tepass, U., and Hartenstein, V.
(1992). The function of the neurogenic genes during epithelial
development in the Drosophila embryo. Development 116, 1203-1220.
Held, L. I. (1991). Arrangement of bristles as a function of bristle number
on a leg segment in Drosophila melanogaster. Roux’s Arch. Dev. Biol.
199, 48-62.
Hemsley, A. N., Arnheim, N., Toney, M. D., Cortopassi, G., and Galas,
D. J. (1989). A simple method for site directed mutagenesis using the
polymerase chain reaction. Nucleic Acids Res. 17, 6545-6551.
Huang, F., Dambly-Chaudière, C. and Ghysen, A. (1991). The
emergence of sense organs in the wing disc of Drosophila. Development
111, 1087-1095.
Jarman, A. P., Brand, M., Jan, L. Y. and Jan, Y. N. (1993). The
regulation and function of the helix-loop-helix gene, asense, in
Drosophila neural precursors. Development 119, 19-29.
Jiménez, F. and Campos-Ortega, J. A. (1987). Genes in subdivision 1B of
the Drosophila melanogaster X-chromosome and their influence on
neural development. J. Neurogen. 4, 179.
Jiménez, F. and Campos-Ortega, J. A. (1990). Defective neuroblast
commitment in mutants of the achaete scute complex and adjacent genes
of D. melanogaster. Neuron 5, 81-89.
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.
Lindsley, D. L. and Grell, E. H. (1968). Genetic Variations of Drosophila
melanogaster. Carnegie Inst. of Washington Publication No 627, Wash.
DC
Lindsley, D. L. and Zimm, G. G. (1992). The Genome of Drosophila
melanogaster. Academic Press, San Diego.
Marí-Beffa, M., de Celis, J. F., and García-Bellido, A. (1991). Genetic
and developmental analyses of chaetae pattern formation in Drosophila
tergites. Roux’s Arch. Dev. Biol. 200, 132-142.
McIver, S. B. (1985). Mechanoreception. In Comprehensive Insect
Physiology, Biochemisty and Pharmacology, Vol 6 (ed. L. I. Gilbert and
D. A. Kerkut). New York/London: Pergamon Press.
Miller, A. (1950). The internal anatomy and histology of the Imago of
Drosophila melanogaster. In Biology of Drosophila (ed. M. Demerec), pp.
420-534. New York: Wiley.
Moscoso del Prado, J. and García-Bellido, A. (1984). Genetic regulation
of the achaete-scute complex of Drosophila melanogaster. Roux’s Arch.
Dev. Biol. 193, 242-245.
Murre, C., Schonleber-McCaw, P., Vaessin, H., Caudy, M., Jan, L. Y.,
Jan, Y. N., Cabrera, C. V., Buskin, J. N., Hauschka, S. D., Lassar, A.
B., Weintraub, H. and Baltimore, D. (1989). Interactions between
heterologous Helix-Loop-Helix proteins generate complexes that bind
specifically to a common DNA sequence. Cell 58, 537-544.
Olsen, E. N. (1990). MyoD family: a paradigm for development? Genes
Dev. 4, 1454-1461.
17
Ott, M. O., Bober, E., Lyons, G., Arnold, H. H. and Buckingham, M.
(1991). Early expression of the myogenic regulatory gene, Myf-5, in
precursor cells of skeletal muscle in the mouse embryo. Development 11,
1097-1107.
Parkhurst, S. M. and Ish-Horowicz, D. (1991). Mis-regulating
segmentation gene expression in Drosophila. Development 111, 11211135.
Rodríguez, I., Hernandez, R., Modolell, J. and Ruiz-Gómez, M. (1990).
Competence to develop sensory organs is temporally and spatially
regulated in Drosophila epidermal primordia. EMBO J. 9, 3583-3592.
Romani, S., Campuzano, S, and Modolell, J. (1987). The achaete-scute
complex is expressed in neurogenic regions of Drosophila embryos.
EMBO J. 6, 2085-2092.
Romani, S., Campuzano, S., Macagno, E. and Modolell, J. (1989).
Expression of achaete and scute genes in Drosophila imaginal discs and
their function in sensory organ development. Genes Dev. 3, 997-1007.
Rudnicki, M. A., Braun, T., Hinuma, S., and Jaenisch, R. (1992).
Inactivation of MyoD leads to upregulation of the myogenic HLH gene
Myf-5 and results in apparently normal muscle development. Cell 71,
383-390.
Ruiz-Goméz, M. and Ghysen, A. (1993). The expression and role of a
proneural gene, achaete, in the development of the larval nervous system
of Drosophila. EMBO J. 12, 1121-1130.
Sassoon, D., Lyons, G., Wright, W. E., Lin, V., and Weintraub, H.
(1989). Expression of two myogenic regulatory factors myogenin and
MyoD1 during mouse embryogenesis. Nature 341, 303-307.
Simpson, P. (1990). Notch and the choice of cell fate in Drosophila
neuroepithelium. Trends in Genetics 6, 343-345.
Skeath, J. B., and Carroll, S. B. (1991). Regulation of achaete-scute gene
expression and sensory organ pattern formation in the Drosophila wing.
Genes Dev. 5, 984-995.
Tautz, D. and Pfeifle, C. (1989). A non-radioactive in situ hybridization
method for the localization of specific RNAs in Drosophila embryos
reveals translational control of the segmentation gene hunchback.
Chromosoma 98, 81-85.
Truman, J. W., and Bate, M. (1988). Spatial and temporal patterns of
neurogenesis in the central nervous system of Drosophila melanogaster.
Dev. Biol. 125, 145-157.
Vaessin, H., Grell, E., Wolff, E., Bier, E., Jan, L. Y. and Jan, Y. N.
(1992). prospero is expressed in neuronal precursors and encodes a
nuclear protein that is involved in the control of axonal outgrowth in
Drosophila. Cell 67, 941-953
Van Doren, M., Powell, P. A., Pasgernak, D. and Posakony, J. W.
(1992). Spatial patterning of proneural clusters in the Drosophila wing
imaginal disc: auto- and cross-regulation of achaete is antagonized by
extramacrochaetae. Genes Dev. 6, 2592–2605.
Weintraub, H., Davis, R., Tapscott, S., Thayer, M., Krause, M.,
Benezra, R., Blackwell, T., Turner, D., Rupp, R. and Hollenberg, S.
(1991). The MyoD gene family: nodal point during the specification of the
muscle cell lineage. Science 251, 761-766.
(Accepted 1 June 1993)