Copyright ? 2007 by the Genetics Society of America
The Balance Between the Novel Protein Target of Wingless and the
Drosophila Rho-Associated Kinase Pathway Regulates Planar Cell
Polarity in the Drosophila Wing
SeYeon Chung,* Sangjoon Kim,*,1Jeongsook Yoon,* Paul N. Adler†,2,3and Jeongbin Yim*,3
*School of Biological Sciences, Seoul National University, Seoul 151-742, Korea and†Biology Department,
University of Virginia, Charlottesville, Virginia 22903
Manuscript received November 29, 2006
Accepted for publication March 14, 2007
Planar cell polarity (PCP) signaling is mediated by the serpentine receptor Frizzled (Fz) and transduced
by Dishevelled (Dsh). Wingless (Wg) signaling utilizes Drosophila Frizzled 2 (DFz2) as a receptor and also
requires Dsh for transducing signals to regulate cell proliferation and differentiation in many devel-
opmental contexts. Distinct pathways are activated downstream of Dsh in Wg- and Fz-signaling pathways.
Recently, a number of genes, which have essential roles as downstream components of PCP signaling, have
been identified in Drosophila. They include the small GTPase RhoA/Rho1, its downstream effector Drosophila
rho-associated kinase (Drok), and a number of genes such as inturned (in) and fuzzy (fy), whose biochemical
functions are unclear. RhoA and Drok provide a link from Fz/Dsh signaling to the modulation of actin
cytoskeleton. Here we report the identification of the novel gene target of wingless (tow) by enhancer trap
screening. tow expression is negatively regulated by Wg signaling in wing imaginal discs, and the balance
between tow and the Drok pathway regulates wing-hair morphogenesis. A loss-of-function mutation in tow
does not result in a distinct phenotype. Genetic interaction and gain-of-function studies provide evidence
that Tow acts downstream of Fz/Dsh and plays a role in restricting the number of hairs that wing cells form.
their apical/basal axis. Epithelia are also often polar-
ized within the plane of the tissue, which is called planar
cell polarity (PCP) or tissue polarity. In recent years,
substantial progress has been made in discerning the
mechanisms that lead to PCP in Drosophila in several
body regions, especially in the wings and the eyes (re-
viewed in Shulman et al. 1998; Adler 2002). In the
Drosophila wing, each cell produces a single distally
oriented hair. This arises from the polarized assembly
of an actin-containing prehair at the distal vertex of
each hexagonal wing cell (Wong and Adler 1993).
Mutants in PCP signaling have characteristic defects in
both the orientation and number of hairs per wing cell
(Gubb and Garcia-Bellido 1982; Wong and Adler
1993). In the eye, polarity is reflected in the mirror
image arrangement of ommatidia of opposite chiral
forms across the dorsal/ventral boundary, the equator.
The photoreceptors within each ommatidium are ar-
ranged in an asymmetric trapezoidal shape, with the
R7/R8 photoreceptor pointing toward the equator
N most multicellular organisms, epithelial cells form
highly organized tissues with cells polarized along
and R3 toward the polar side. Mutations in genes that
regulate PCP result in failure of ommatidia to acquire
the correct chirality and/or failure to rotate properly
(Zheng et al. 1995; Tomlinson and Struhl 1999;
Weber et al. 2000).
Using a number of approaches, investigators have
identified genes required for the specification of planar
cell polarity in Drosophila. This has led to the identifi-
cation of the Frizzled pathway as a key for the devel-
opment of tissue polarity (Gubb and Garcia-Bellido
1982; Vinson and Adler 1987; Wong and Adler 1993;
the key members of this pathway are frizzled (fz), which
encodes a serpentine receptor (Vinson et al. 1989);
dishevelled (dsh), which encodes a cytoplasmic protein
et al. 1995); and starry night/flamingo (stan/fmi), which
is a cadherin protein with a seven-pass transmembrane
domain (Chae et al. 1999; Usui et al. 1999).
Dsh and Fz family proteins also participate in the
ulate a wide variety of developmental events, including
cell proliferation and cell fate specification (reviewed
in Logan and Nusse 2004). Recent works also have
uncovered some common aspects of the canonical Wg-
signaling pathway and PCP signaling: The trimeric
G-protein Ga0-subunit was reported to be required for
direct transduction of Fz signals from the membrane to
of Medicine, Johns Hopkins University, Baltimore, MD 21205-2185.
of Virginia, Charlottesville, VA 22903.E-mail: firstname.lastname@example.org
3These authors contributed equally to this work.
Genetics 176: 891–903 ( June 2007)
downstream components in both pathways (Katanaev
et al. 2005), and a genetic approach based on fz alleles
revealed that even though the PCP signaling and the
canonical Wg signaling are genetically distinct, fz con-
trols these two different signaling activities by a com-
mon mechanism (Povelones et al. 2005). However, Wg
and PCP signaling differ in their requirements for more
downstream factors such as armadillo/b-catenin (arm/
b-cat) (Noordermeer et al. 1994) and the transcription
factor pangolin/Drosophila T cell Factor (pan/dTCF)
(Brunner et al. 1997; van de Wetering et al. 1997) for
the former, and small GTPase RhoA/Rho1 (Strutt et al.
1997), MAP kinase cascade components (Boutros et al.
1998; Paricio et al. 1999), and the inturned/fuzzy (in/fy)
for the latter. The pathways appear to diverge down-
stream of Dsh (Axelrod et al. 1998; Boutros et al. 1998).
The PCP signal from Fz/Dsh directs asymmetric cyto-
skeletal reorganization and polarized cell morphology
in part by activating RhoA/Rho1 (Strutt et al. 1997)
and its downstream effector, Drosophila Rho-associated
kinase, Drok (Winter et al. 2001). RhoA acts as a mo-
lecular switch that gates signaling to downstream targets,
both nuclear and cytoskeletal. In eye development,
RhoA signals via Jun N-teminal kinase and Jun to regu-
late PCP-dependent transcription (Strutt et al. 1997;
Fanto et al. 2000; Weber et al. 2000). In the wing, RhoA
activity via phosphorylating Spaghetti squash (Sqh), a
light chain (MRLC) (Karess et al. 1991; Winter et al.
2001). Mutations in Drok result in wing cells forming
multiple hairs of normal orientation. In the eye, muta-
tions in Drok lead to abnormalities in the rotation of
ommatidia (Winter et al. 2001).
gene, target of wingless (tow), is negatively regulated by a
canonical Wg-signaling pathway and that tow interacts
with Drok in PCP signaling. Loss-of-function mutations
in tow show neither a PCP phenotype nor wg-signaling-
related defects, but itsoverexpression leads towing cells
forming multiple hairs of normal polarity much like
loss-of-function mutations in Drok. We found that Tow is
a nuclear protein in pupal wing cells and that spaghetti
squash (sqh) mRNA was downregulated when Tow was
overexpressed. This downregulation is presumably at
least part of the reason for the Tow overexpression
MATERIALS AND METHODS
Drosophila stocks: Alleles used in this study were UAS-wg
and UAS-dTCFDN(gifts from R. Nusse); sev-Gal4 UAS-fz (a re-
combinant of sev-Gal4 and UAS-fz, gift from D. Strutt); UAS-
RhoA, UAS-RhoV14, UAS-Drok-CAT, and UAS-Drok-CAT-KG, sqhAX3
(gifts from L. Luo); dsh1, a-tub-Drok, Drok2, RhoA72O, zip1, zipIIX62,
ck13, ck07130, and UAS-RedStinger (Bloomington Stock Center).
stanVC31, a dominant-negative allele of stan, was induced by
EMS, and it is a recessive lethal. shaVB13is a temperature-
sensitive hypomorphic allele of shavenoid (sha; also called
kojak) (He and Adler 2002).
Enhancer trap screening: An enhancer trap screen was
performed as described by Kim et al. (2006).
X-gal staining of third instar larvae: Imaginal discs were
(PBS), fixed in 1% glutaraldehyde in PBS for 20 min at room
temperature, and washed twice for 10 min with PBTr (0.5%
Triton X-100 in PBS). The disc complexes were immersed in
200 ml of X-gal staining solution (10.0 mm NaH2PO4?H2O/
Na2HPO4?2H2O [pH 7.2], 150 mm NaCl, 1.0 mm MgCl2,
3.1 mm K4[FeII(CN)6], 3.1 mm K3[FeIII(CN)6], 0.3% Triton X-
100) adding 8 ml of 50 mg/ml X-gal to dimethylformamide
for several hours at 37?. Disc complexes were washed twice for
10 min each with PBTr and mounted in 70% glycerol. LacZ
expression patterns were viewed on Nikon microscopes and
photographed with an AxioCam camera. Larvae derived from
?7500 new insertions were stained for lacZ expression, and
line 4938 was selected since its expression appeared opposite
to that of wg in the wing discs.
Plasmid rescue and flanking sequence analysis: A total of
1–5 mg of the tow-lacZ genomic DNA was fully digested with
EcoRI or SacII, and the DNA fragments were self-ligated. After
precipitation, the ligated DNAs were transformed into Escher-
ichia coli strain XL-1 Blue by electroporation. Of the ligated
DNAs, the genomic DNA flanking of the P-element insertion
was selected by an ampicillin-resistant marker.
The flanking DNA was sequenced using either the IR
quence was identified using the BLAST program in the data-
base of genomic clones sequenced by the Berkeley Drosophila
Genome Project (http:/ /www.fruitfly.org/).
Clonal analysis: Flip-out clones were generated by the Flp-
mediated recombination technique (Golic 1991; Xu and
Rubin1993). Cloneswere inducedat48–60 hrafter egglaying
dissected and fixed. The genotype of the larvae used to make
wg flip-out clones was act.y1.Gal4, UAS-CD2/UAS-wg; hs-FLP,
Immunohistochemistry: Imaginal discs were dissected from
third instar larvae, fixed with fixation buffer (0.1 m PIPES,
pH 6.9, 1 mm EDTA, 1.0% Triton X-100, 2 mm MgSO4, 1%
formaldehyde), blocked in a solution [50 mm Tris–Cl (pH
6.8), 150 mm NaCl, 0.5% NP-40, 5 mg/ml bovine serum
albumin (BSA)] and stained overnight at 4? with appropriate
antibodies in a washing/incubation solution (50 mm Tris–
HCl, pH 6.8, 150 mm NaCl, 0.5% NP-40, 1 mg/ml BSA). After
washing several times, secondary antibodies were applied for
2–4 hr at room temperature or on ice. The cuticles were
mounted in antifade mounting solution (BioMeda). The
fluorescence images were obtained with a confocal micro-
scope system (Zeiss LSM510).
Primary antibodies used in these experiments were rabbit
Amersham Pharmacia Biotech and used in 1:1000.
RNA in situ hybridization: In situ hybridization of third
instar larvaewas doneas describedbefore (TautzandPfeifle
1989). Digoxigenin-labeled antisense RNA probes were syn-
thesized from full-length cDNA of tow.
Generation of tow mutants: To obtain a deletion mutant of
tow, we induced imprecise excision of the P element in the tow-
lacZ line. The flanking sequence is occasionally deleted
during P-element jumping. The source of the transposase in
this dysgenic cross was [D2-3]Ki, and it mobilized P[lacW]
892S. Chung et al.
from the insert site. First, tow-lacZ homozygous females were
crossed to male [D2-3]Ki, and the male progeny were crossed
to balancer virgins (w; TM3/TM6B). Among the progeny,
those whose P element had jumped out had white eyes; hence
we selected male progeny with white eyes, which do not have
or TM6B by crossing individually to TM3/TM6B virgins. Of
the ?500 lines, one null mutant, tow754, whose deletion range
covers the whole open reading frame (ORF) of the tow tran-
script, was confirmed by genomic PCR.
towfzinversion mutant was created by hs-FLP-mediated in-
version. The P/Piggy Bac insertion alleles towd02991and fzf07870
The relevant transposons contain FRTsequences (Parks et al.
2004). A recombinant chromosome carrying both insertions
flp; towd02991fzf07870/TM6 were generated and heat-shocked
several times during larval and pupal life. These were crossed
to w; fzR53females and the progeny were screened for phe-
notypically fz flies. Stocks were established from several such
flies and salivary gland squashes were performed to confirm
the presence of the predicted inversion. The inversion breaks
both the tow and the fz transcription units. It results in a
phenotypically null fz allele and a tow allele that appears to be
a biochemical null.
Generation of transgenic flies: Sequencing confirmed that
the fly EST GH12583 that encodes Tow carries an intact tow
gene. We cloned the full-length EST into the pUAST vector
(Brand and Perrimon 1993) to generate flies carrying the
UAS-tow construct. Various Gal4 drivers were used to analyze
the consequences of overexpressing Tow.
arm-tow-GFP is a fusion at the C-terminal end of the full-
length Tow open reading frame to the N-terminal end of GFP.
It was cloned into the pCaSpeR vector with an armadillo (arm)
promoter, and germline transformation was performed by
standard methods (Spradling and Rubin 1982).
Actin staining: Pupal wings were stained with rhodamine-
labeled phalloidin, as previously described (Wongand Adler
Mosaic analysis of Drok2clones: FLP/FRT-mediated recom-
bination was used to generate clones homozygous for Drok2
in the wing via heat-shock-induced recombination 48–72 hr
AEL. Homozygous wing clones in newly eclosed adults were
visualized using the MARCM system (Lee and Luo 1999) by
the presence of the membrane marker mCD8GFP. The geno-
Genetic interaction and sectioning of adult eyes: For ge-
netic interaction studies with sev-Gal4 UAS-fz (Strutt et al.
1997), flies heterozygous for both sev-Gal4 UAS-fz and the tow
mutation were analyzed. Adult heads were cut and fixed
overnight in 4% paraformaldehyde solution at 4?, washed
dark brown. After washing three times for 10 min in distilled
water, the samples were dehydrated with ethanol and pro-
pylene oxide. Next, the propylene oxide was exchanged with
Spurr and embedded in a mold and polymerized for 24 hr at
65?. The block was sectioned into 200- to 300-nm thick slices
with an ultramicrotome through the equatorial region of the
eye. Thesliceswerestainedwith 1%ToluidineBlueOonahot
plate for several minutes, and ommatidia were examined for
incorrect rotation and chirality. At least five sections from
independent eyes were analyzed for each genotype.
Genetic interaction assays using multiple wing hairs: Wings
were mounted in Euparal (Asco Labs) or Gary’s Magic Mount
and examined under bright-field microscopy. The multiple-
wing-hair phenotype in Figure 6 and Table 2 was induced by
heat shock at 24 hr after puparium formation (APF) at 37? in
animals carrying one copy of hs-Gal4, UAS-tow and the mu-
tation or transgene of several components in Drok pathway.
surface of an area enclosed by the L3 and L4 vein and the first
the tip of second intervein to the L3. Since overexpression of
Tow usually causes the wing blade surface to be uneven and
wavy, counting the number of multiple hairs was easier in the
central region of the wing than in other regions. The number
of multiple hair cells was counted and averaged for at least 10
Subcellular localization of Tow: ap-Gal4/1; UAS-RedStinger/
arm-tow-GFP pupae were fixed in 4% paraformaldehyde in PBS
at 4? overnight. UAS-RedStinger line expresses a variant
DsRed (red fluorescent protein) with a nuclear localization
signal under the control of UAS. The GFP signals were seen
without staining or after anti-GFP staining (1:4,000; Molecular
Probes, Eugene, OR). The wings were mounted in Prolong
Gold Mounting Media (Molecular Probes) and examined in a
ATTO CARV confocal unit attached to a Nikon camera.
Northern hybridization: Heat shock was given to pupae of
times with a recovery period of 40 min. Total RNA was isolated
with TrizolR(Gifco). Thirty micrograms of total RNA/sample
was separated by formaldehyde gel, blotted onto nitrocellulose
membrane (Schleicher & Schuell, Keene, NH), and hybridized
with32P-labeled DNA probe specific for sqh. The sqh fragment
was amplified by RT–PCR from wild-type flies using the sqh-
specific primers sqh-RTP1-59 (AGT-GCA-GCT-GGT-CGA-AAG-
TT) and sqh-RTP2-39 (ATT-CGA-GGT-AGT-CGA-ACA-GA).
Western blotting: hs-Gal4/1; UAS-tow/1 pupae were heat-
shocked as described for the Northern blot analyses. Total
protein was isolated and analyzed using standard methods.
Thirty micrograms of total protein/sample was separated by
4–20% gradient SDS–PAGE gel, blotted onto PVDF mem-
brane, and incubated with anti-Sqh or anti-phospho-Sqh
antibody (Cell Signaling).
Identification of Wg-responsive enhancer trap, 4938,
inserted at the tow locus: To identify novel molecules
involved in wing development, we performed an en-
hancer trap screen (Kim et al. 2006). The P[lacW] trans-
poson (Bier et al. 1989) was randomly mobilized, and
larvae derived from 7500 P-element insertions were
stained for b-gal expression. Line 4938 was further
analyzed because it drove lacZ expression in the wing
pouch but not at the dorsal/ventral (D/V) compart-
ment boundary of the wing disc (Figure 1A) where wg
is expressed. We show below that this enhancer trap is
inserted into the tow gene and that its expression is
identical to that of tow.
expression led us to test whether Wg signaling could
regulate its expression. First, we induced Wg-expressing
flip-out clones using the FLP–FRTsystem (Golic 1991;
Xu and Rubin 1993) and found that the b-gal expres-
sion of tow-lacZ was repressed nonautonomously in the
vicinity of the clones (Figure 1D). The nonautonomy is
(Zecca et al. 1996; Neumann and Cohen 1997) and
Function of target of wingless in Planar Cell Polarity893
antibody staining is not sensitiveenough to detect levels
of ligand known to be biologically active in distant cells
(Neumann andCohen1997; StriginiandCohen2000).
We also utilized the UAS/Gal4 system (Brand and
Perrimon 1993) to express a dominant-negative form
of Drosophila Tcell factor (dTCF). The wild-type dTCF
while dTCFDN(van de Wetering et al. 1997) inhibits Wg
signaling. Expression of dTCFDNin a Decapentaplegic
(Dpp) domain [i.e., a stripe that runs perpendicular
to the D/V wg stripe at the anterior/posterior (A/P)
boundary] leads to derepression of b-gal activity of
tow-lacZ within the Dpp domain (Figure 1C). Thus, Wg
signaling negatively regulates the expression of tow-lacZ
enhancer trap. In other experiments, we found the ex-
pression of wg was not affected by overexpression or
absence of tow (Figure 1B). Drosophila frizzled 2 (Dfz2) is a
receptor for canonical Wg signaling pathway (Bhanot
et al. 1996) and is expressed in the wing in a pattern very
similar to that of tow, i.e., with the lowest levels found at
between Wg and DFz2 plays a crucial role in shaping
the Wg morphogen gradient and in determining the
response of cells to the Wg signal. We found that the
expression of Dfz2 was also not affected by tow (data not
Sequencing of genomic DNA fragments flanking
the insertion showed that the P element was inserted
into the first intron of a novel gene (CG14821) that we
347 amino acids. Homologs of tow are found in other
insects but not in more distantly related organisms. Se-
Figure 1.—Expression of the tow enhancer
trap is negatively regulated by Wg signaling in
the wing disc. (A) Wild-type tow-lacZ expression
(X-gal) is found in the whole-wing pouch except
the D/V boundary (indicated by white dashed
line). (B) Wg expression (red) is not affected
by ptc-Gal4-driven overexpression of Tow (green).
The genotype of the larvae used is ptc-Gal4/UAS-
GFP; UAS-tow/1. (C) dpp-Gal4-driven UAS-dTCFDN
derepresses tow-lacZ expression at the A/P bound-
of the larvae is UAS-dTCFDN/1; dpp-Gal4/tow-lacZ.
(D) Ectopic expression of Wg (asterisks in D-I
and D-iii) represses tow-lacZ. (D-i) b-Galactosidase
activity of tow-lacZ (red). (D-ii) Wg flip-out clones
marked by Wg antibody (green). (D-iii) Merged
image of i and ii. (E-i) Structure of tow gene, en-
hancer trap P-element insertion, and deletion cre-
ated byimpreciseexcision. towtranscript hasavery
long 59-UTR, and 7.4-kb genomic DNA is deleted
in the tow754mutant, which removes the whole
ORF of the tow gene and 160 amino acids of 39
of the neighboring gene, CG10077. Homozygous
tow754mutant flies are normal. (E-ii) towfzmutant
(tow and fz double mutant) created by hs-FLP-
mediated inversion. Homozygous towfzmutant
shows typical fz mutant phenotype. Trans-heterozy-
gote of tow754and towfzshows a normal phenotype.
(F) In situ hybridization to wild-type wing disc us-
ing a tow-derived RNA probe shows a pattern very
similar to that of tow-lacZ. (G) Sequence of Tow
protein. The residues in red are Gln repeats,
and in blue are Pro repeats. Bars in A–D and E,
894 S. Chung et al.
gene, GA13273, in Drosophila pseudoobscura and that it is
also very well conserved in other Drosophila species,
such as D. simulans, D. yakuba, D. erecta, D. ananassae,
available at http:/ /species.flybase.net/). In addition, the
first 60 amino acids in the N terminus show homology
to a gene in Anopheles gambiae. Tow does not contain in-
formative motifs. However, the sequence is notable for
containing eight runs of multiple Gln residues and one
Pro-rich region and for being unusually rich in Gln
(20.2%) and Pro (10.1%) (Figure 1G).
specific probe showed the same expression pattern as
4938 lacZ staining, in that tow mRNA was expressed in
tow null mutant is viable: To obtain a loss-of-function
allele of tow, we mobilized the tow-lacZ P element. We
obtained an allele, tow754, which deletes the entire tow
open reading frame and a part of a neighboring gene,
CG10077 (Figure 1E-i). By genomic DNA sequencing,
we determined that the 7.4-kb deletion removes the
entire tow gene except for the first (noncoding) exon
and part of the first intron and 160 amino acids of the
39-end of CG10077-RA, which is the longer of its two
as no tow mRNA could be detected in tow754homozygous
flies (data not shown). Homozygotes of tow754were viable
and showed no obvious phenotype. Similar results were
obtained with trans-heterozygotes of this tow mutant and
a deficiency for the region.
Since tow754deletes a part of another gene, we made
another allele using hs-FLP-mediated inversion. We ob-
tained an inversion mutant allele, towfz, in which a geno-
mic DNA fragment between the first intron of tow and
the final intron of fz was inverted (Figure 1E-ii). Homo-
zygous towfzflies were also viable and showed a typical
fz mutant phenotype in the wings and thorax. No tow
mRNA was detected in homozygous towfzflies in RT–
PCR experiments (data not shown). Trans-heterozygotes
of tow754and towfzflies were normal, showing that Tow is
Overexpression of Tow causes planar polarity
defects manifested by multiple wing hairs: Gain-of-
function studies can often give a clue as to the function
of a gene and are particularly useful in analyzing func-
tionally redundant genes. We generated transgenic flies,
which carried a UAS-tow cDNA transgene and crossed
them to various Gal4 drivers (Brand and Perrimon
1993). Activation of transcription by Gal4 in apterous
(ap)-expressing cells results in overexpression in thorax
and dorsal wing blade. Interestingly, when Tow was
overexpressed by the ap-Gal4 driver, it caused a disor-
ganization of microchaetae and stunted macrochaetae
on the thorax as well as defects in the scutellum (Figure
2B). A disorganized microchaetae phenotype was also
seen as a consequence of the misexpression of several
planar polarity genes, and gain-of-function screening
using this bristle defect has been successfully used in
finding a novel component of planar polarity (Paricio
et al. 1999; Feiguin et al. 2001). Therefore, this pheno-
type led us to hypothesize that Tow is also involved in
PCP. It is intriguing that a gene whose expression is
regulated by Wg signaling could have a function in PCP
signaling because the canonical Wg signaling and PCP
signaling have been known to diverge into distinct path-
ways downstream of Dsh (Axelrod et al. 1998; Boutros
et al. 1998).
We overexpressed Tow in the wing using several wing-
specific Gal4 drivers, all of which resulted in multiple
hair cells (Figure 2D). In wild-type wings, each cell pro-
duces a single, distally oriented wing hair, and multiple-
hair-cell phenotypes are associated with PCP defects.
Most PCP genes, including fz and dsh, exhibit both
stereotypical orientation and multiple-hair-cell defects
(Gubb and Garcia-Bellido 1982; Wong and Adler
Figure 2.—Overexpression of Tow causes disorganization
of microchaetae on the thorax and multiple hairs in the wing.
(A and B) Microchaetae phenotype in adult thorax. Anterior
is left. (A) Thorax of ap-Gal4/CyO. Microchaetae are regularly
oriented and point posteriorly. (B) Thorax of ap-Gal4/1; UAS-
tow/1. Misexpression of Tow results in disorganization of mi-
crochaetae and defects in the scutellum (arrow). (C–F) Adult
(C–E) and pupal (F) wing phenotypes. In all of the wings,
proximal is to the left. (C) Wild-type adult wing. (D) Multiple-
hair-cell phenotype of MS1096-Gal4-driven UAS-tow adult
wing. A few multiple hairs are indicated by red arrows. (E)
Multiple hairs on the adult wing in the Drok2mutant clone.
The clonal border is marked by a white dashed line. (F) Phal-
loidin staining of the MS1096-Gal4-driven UAS-tow pupal wing
reveals multiple F-actin prehairs. White arrows represent sev-
eral multiple F-actin bundles.
Function of target of wingless in Planar Cell Polarity895
1993), but others such as Drok show only the multiple-
wing-hair phenotype (Winter et al. 2001; Figure 2E).
To investigate the cellular basis of the Tow overex-
pression phenotype, we used phalloidin staining to
examine the distribution of F-actin during hair mor-
phogenesis. When we used MS1096-Gal4 to drive Tow
expression mainly in the dorsal region of the wing
(Capdevila and Guerrero 1994), a majority of wing
cells formed more than one F-actin bundle (Figure 2F).
Notably, most prehairs maintained a roughly distal ori-
entation. The similarities between the pupal and adult
phenotypes indicate that the multiple-hair-cell pheno-
type of adult Tow-overexpressing wings is likely the
result of a failure to restrict F-actin bundle assembly to a
single site during prehair formation. Due to the simi-
larity of their multiple-hair-cell phenotypes, we thought
it possible that Tow might be involved in the Drok
pathway. In addition, MS1096-Gal4-driven Tow overex-
and in the wing blade near the most distal region, which
suggests that ectopic expression of Tow in the wing mar-
gin where it is not normally expressed results in forma-
tion of extra bristle sense organs in those regions (data
Tow acts downstream of Frizzled/Dishevelled in planar
cell polarity: The proper level of Fz/Dsh signaling is
reported that both overexpression and loss of function
of many PCP genes result in polarity defects in the eye
and the wing (Krasnow and Adler 1994; Axelrod et al.
1998; Gubb et al. 1999; Usui et al. 1999; Treeet al. 2002).
The multiple-hair-cell phenotype of Tow gain of function
led us to hypothesize that Tow might also act downstream
of Fz/Dsh. To assess the genetic interaction of the Fz
pathway and tow, we employed two kinds of assays in the
eye and in the wing.
In the Drosophila eye, PCP is reflected in the mirror-
symmetric arrangement of ommatidial units relative to
the dorso-ventral midline, the equator. This pattern is
generated posterior to the morphogenetic furrow when
ommatidial preclusters rotate 90? toward the equator,
adopting opposite chirality, depending on their dorsal
or ventral positions (Gubb 1993; Figure 3A). Polarity
defects are manifested in the loss of mirror-image sym-
metry, with the ommatidia misrotating and adopting
random chirality or remaining symmetrical (Gubb 1993;
Theisenet al. 1994; Zheng et al. 1995; Strutt et al. 1997;
Boutros et al. 1998; Wolff and Rubin 1998).
The sevenless (sev)-driven gain-of-function Fz or Dsh
phenotype (sev-Gal4 UAS-fz or sev-Gal4 UAS-dsh) has
previously been used successfully to identify and study
new components of the Fz/Dsh planar polarity pathway
(Strutt et al. 1997; Boutros et al. 1998). We employed
the same assay to determine if Tow could be a new com-
ponent of the Fz pathway. sev-driven overexpression of
Fz in the eye results in planar polarity defects charac-
terized by abnormal ommatidial chirality and misrota-
tion due to overactivation of planar polarity signaling
(Strutt et al. 1997; Figure 3B). Null tow754flies showed
no eye polarity defects and the overexpression of Tow
Figure 3.—tow null mutation
dominantly suppresses a polarity-
specific Fz gain-of-functionpheno-
type. (A–D) Tangential sections of
the equatorial regionofadult eyes.
Anterior is left, and dorsal is up.
Each panel contains eye sections
(top) and a schematic of the same
area with arrows reflecting omma-
tidial polarity (bottom). (A) sev-
Gal4/UAS-tow. It has completely
normal eyes. Equator is indicated
by gray lines, and it shows omma-
tidia of normal dorsal and ventral
respectively. (B) sev-Gal4 UAS-fz/1.
Transient sevenless-driven overex-
pression of Fz in the eye results in
planar polarity defects. Red arrows
represent misrotated ommatidia;
green arrows without flags repre-
sentsymmetrical nonchiral omma-
tidia. Circlesmark unscorable
ommatidia, usually due to missing
photoreceptors or abnormal mor-
tow754. The mutation in tow is able to dominantly suppress the gain-of-function sev-Gal4 UAS-fz eye phenotype. It is almost completely
tidia look different in the different genotypes due to fixation or staining variation.
896 S. Chung et al.
(Figure 3A). However, we found that a null mutation of
tow acted as a strong dominant suppressor of the gain of
function of the sev-Gal4 UAS-fz eye phenotype (Figure
3C; Table 1), as is the case for PCP genes such as RhoA
(Strutt et al. 1997).
We confirmed these results using another tow allele,
towfz. Introduction of a towfzchromosome could also
suppress the sev-Gal4 UAS-fz eye phenotype (Table 1).
Since towfzis a double mutant for tow and fz, and a fz
mutant itself can dominantly suppress the sev-Gal4-
driven Fz overexpression phenotype (Strutt et al.
1997; Table 1), we tried to rescue fz activity in the towfz
mutant by introduction of an arm-fz-GFP chromosome
(Strutt 2001). The suppression of the sev-Gal4 UAS-fz
eye phenotype by towfzwas still seen when the arm-fz-GFP
chromosome was present (Table 1). In addition, co-
overexpression of Tow and Fz resulted in enhanced po-
larity defects; notably, the frequency of misrotation of
ommatidiawasincreased(Figure 3D; Table 1).Thus the
sev-Gal4 UAS-fz phenotype is sensitive to both an in-
crease and a decrease in tow dose. These data suggest
or it could function in parallel in eye planar polarity.
In addition, two kinds of tow alleles showed the same
in the tow754mutant does not seem to affect the genetic
interaction with fz.
We also examined genetic interactions between tow
dsh1allele, which is defective for PCP function without
affecting canonical Wg signaling (Axelrod et al. 1998;
Boutros et al. 1998). In a dsh1mutant, typical polarity
defects such as swirling hair patterns and multiple wing
hairs are seen (Wong and Adler 1993; Figure 4A). To
assess the genetic interactions between dsh and tow, we
region, the dorsal surface of the central region of the
In dsh1hemizygous males, an average of 4.8 cells with
multiple hairs were present in this region (Figure 4, A
and E). This multiple-hair-cell phenotype was sup-
pressed by the reduction of tow dosage, leading to an
average of only 0.8 multiple hair cells in the defined
region, when one copy of the tow754chromosome was
introduced (Figure 4, B and E). The polarity of wing
hairs in dsh1hemizygotes was not affected by tow dosage.
ap-Gal4-driven UAS-tow resulted in multiple hairs over
the entire dorsal surface of the wing, and in our target
region an average of 14.9 cells with multiple hairs were
by ap-Gal4 in a dsh1hemizygous background, the aver-
age number of cells exhibiting multiple hairs was sub-
stantially increased to 44.3/wing region (Figure 4, D
and E). ap-Gal4-driven overexpression of Tow caused the
wing blade to be very uneven and wavy, so the polarity
Quantification of genetic interactions with the sev-Gal4
(% 6 SD)
arm-fz-GFP/1; sev-Gal4, UAS-fz/1
arm-fz-GFP/1; sev-Gal4, UAS-fz/fzK2165.3 (64.4)
arm-fz-GFP/1; sev-Gal4, UAS-fz/towfz73.8 (63.8)
Quantification of genetic interaction of sev-Gal4 UAS-fz with
tow. The quantification of allelic combinations is based on the
scoring of five independent eyes per genotype. The percent-
age figure shown is the average number of correctly oriented
ommatidia, with the standard deviation calculated across all
eyes of a given genotype scored.
Figure 4.—tow mutant suppresses dsh1multiple-hair-cell
phenotype, and its gain-of-function phenotype is enhanced
in a dsh1background. (A–D) Representative multiple-hair-cell
phenotypes (circled in red) on a dorsal surface of the central
region of the adult wing. The genotypes are (A) dsh1/Y, (B)
dsh1/Y; tow754/1, (C) ap-Gal4/1; UAS-tow/1, and (D) dsh1/Y;
ap-Gal4/1; UAS-tow/1. The dsh1polarity phenotype looked
like it was enhanced because ap-driven overexpression of
Tow caused the wing blade to be very uneven. (E) The num-
ber of multiple hair cells. Error bars represent standard er-
rors. In all of the adult wings, proximal is to the left.
Function of target of wingless in Planar Cell Polarity 897
was overexpressed by a weaker driver such as hs-Gal4,
was increased and no apparent enhancement of polarity
wings of dsh1/Y; ap-Gal4/1 or dsh1/Y; UAS-tow/1 flies, no
dsh1alleles when fz activity was rescued by the arm-fz-GFP
chromosome (data not shown). In a previous study, the
overexpression of Drok reduced the average number of
multiple hair cells in dsh1hemizygotes, and a reduction
cells, which led to the suggestion that Drok was activated
by Fz signaling (Winter et al. 2001). We found an oppo-
site dose response between tow and dsh1. Taken together,
these experiments suggest that Tow functions down-
stream of Fz/Dsh perhaps to inhibit Drok activity.
Genetic interaction of tow with other PCP genes: To
see if tow genetically interacts with other known PCP
genes, we crossed the tow754null mutant to several PCP
mutants and examined the phenotypes of the trans-
heterozygotes or double homozygotes. The dominant
allele of stan, stanVC31, causes a swirling polarity pattern
and a small number of multiple hair cells on the wing.
The number of multiple hair cells was increased by a
reduction in tow dosage (Figure 5, A and B). The intro-
duction of one copy of the tow754null allele also can
enhance the phenotype of the sha mutant. Mutations in
sha result in a delay in hair morphogenesis and, in some
cells, no hair or only several small hairs form (He and
Adler 2002). The morphogenesis of the hair involves
temporal control by sha and spatial control by the genes
of the Fz pathway, and there is a strong genetic in-
teraction between mutations in these genes (Ren et al.
2006). tow mutation caused an increased number of
small hairs and multiple wing hairs in the hypomorphic
shaVB13background (Figure 5, C and D). We also looked
for interactions with other PCP genes such as Van Gogh
[Vang; also known as strabismus (stbm)], which encodes a
putative membrane protein (Tayloret al. 1998; Wolff
and Rubin 1998); fy, which encodes a novel four-pass
prickle (pk), which encodes a protein with LIM domains
(Gubb et al. 1999). However, there was no distinct inter-
action between the tow mutant and VangTBS42, a domi-
alleles of fy (Collier and Gubb 1997); and pkD, a domi-
nant allele of pk (Adler et al. 2000).
tow genetically interacts with the Drok pathway:
Transient overexpression of Tow driven by hs-Gal4 with
heat shock at 24 hr APF caused the formation of mul-
tiple hair cells throughout the wing blade. Once again,
the multiple hairs had largely normal polarity. In the
flies that carried one copy of both hs-Gal4 and UAS-tow,
an average of 10.9 cells with multiple hairs was found
in the defined region (Figure 6A; Table 2). When Drok
was simultaneously overexpressed [using a a-tubP-Drok
transgene (Winter et al. 2001)] with Tow, the average
number of multiple hair cells was reduced by more than
one-half to 4.6/wing region (Figure 6B; Table 2). This
suppression was confirmed with co-overexpression of
Figure 5.—Genetic interaction of tow and other PCP com-
ponents (A–D) Representative wing-hair phenotype of the
dorsal compartment of the adult wing. (A) stanVC31/1 wing
in the anterior region. (B) Trans-heterozygotes of tow754and
stanVC31showing increased multiple wing hairs. (C) shaVB13ho-
mozygous adult wing in the center of the most posterior re-
gion. (D) Reduction of tow dosage increased number of
small hairs and multiple wing hairs in a shaVB13background.
pathway. (A–D) Representative multiple-hair-cellphenotype
(circled in red) of various genotypes in a defined region of
the adult wing. In A–D, heat shock was given at 37? 24 hr
APF. (A) hs-Gal4/1; UAS-tow/1 wing. (B and C) Overexpres-
sion of Drok via a a-tubP-Drok transgene results in a reduced
number of multiple hairs, and reduction of Drok dosage
causes an increase in the number of multiple hair cells. (B)
hs-Gal4/1; UAS-tow/a-tub-Drok. (C) Drok2/1; hs-Gal4/1; UAS-
tow/1. (D) hs-Gal4/zip1; UAS-tow/1. zipper mutation results
in a substantial increase of multiple hairs. In all of the wings,
proximal is to the left.
898 S. Chung et al.
the catalytic domain of Drok (Drok-CAT) (Winter et al.
hair cells/region (Table 2). Complementary dosage
responses were seen with loss-of-function mutations in
Drok. The presence of a Drok2null allele (Drok2/1) en-
hanced the multiple-hair-cell phenotype of hs-Gal4 UAS-
tow by2.3-fold to 24.7 multiplehair cells/region (Figure
6C; Table 2). Similarly, when a kinase-dead form of
Drok, Drok-CAT-KG (Winter et al. 2001), which is not
a dominant-negative form, was co-overexpressed with
Tow, the phenotype was slightly enhanced to 13.5 mul-
tiple hair cells/region (Table 2).
Drok is a downstream effector of RhoA/Rho1, which
functions as a molecular switch that gates signaling
from Fz/Dsh to downstream targets (Strutt et al. 1997;
Winter et al. 2001). The presence of a RhoA72Oallele
(Strutt et al. 1997) slightly increased the number of
multiple hairs caused by hs-Gal4 UAS-tow to 13.3/region
(Table 2). On the otherhand, co-overexpression of Tow
and wild-type RhoA via UAS-RhoA (Lee et al. 2000) also
resulted in a reduced number of multiple hair cells
(2.1/wing region, Table 2). A constitutively active form
of RhoA, UAS-RhoV14(Lee et al. 2000), caused lethality
when overexpressed by hs-Gal4, so we could not exam-
ine the effect.
An important substrate of Drok is sqh, which encodes
the Drosophila homolog of the nonmuscle MRLC. Sqh
is phosphorylated by Drok in vivo (Karess et al. 1991;
Jordan and Karess 1997). Reducing the wild-type sqh
gene dosage from two to one by introducing a single
copy of the sqhAX3null allele (Jordan and Karess 1997)
also slightly enhanced the multiple-wing-hair pheno-
type caused by Tow overexpression (Table 2).
Zipper (Zip) also functions in hair morphogenesis
and interacts with Fz/Dsh (Winter et al. 2001). Loss of
one copy of the zip gene enhanced the dsh1and hs-Fz
multiple-hair-cell phenotypes, suggesting that myosin
II functions positively downstream of Fz/Dsh in hair
development. We found that reducing the dose of zip
caused by overexpression of Tow (Figure 6D, Table 2).
This was confirmed for two independent zip alleles, zip1,
a null mutant, and zipIIX62, a hypomorph (Table 2).
We also found an interaction with crinkled (ck), which
encodes a myosin VIIA. In the mouse, a mutation in
myosin VIIA (shaker-1 mutant) caused stereocilia disor-
ganization and the formation of multiple stereocilia
bundles (Self et al. 1998). ck was previously found to be
involved wing-hair morphogenesis and to interact with
Fz/Dsh signaling in Drosophila. Mutations in ck cause
multiple wing hairs as well as a split-hair phenotype
(Gubb et al. 1984; Turner and Adler 1998), and a re-
phenotype (Winter et al. 2001). The genetic interaction
of ck with components of the Fz/Dsh pathway has oppo-
site effects to that of zip, and the seemingly antagonistic
relationship between myosin II and myosin VIIA sug-
or stoichiometry of these two myosins is critical for the
common process that they regulate (Winter et al. 2001).
We also found a complementary genetic interaction of
Genetic interaction with tow gain-of-function phenotype
Multiple wing hairs/sector
Average Standard error
Drok2/1; hs-Gal4/1; UAS-tow/1
sqhAX3/1; hs-Gal4/1; UAS-tow/1
induced overexpression of Tow at 24 hr APF.
aStatistical significance was determined by Student’s t-test. Note that, in most cases, the P-value is ,0.0001,
which indicates a highly significant difference.
bThe P-value between two samples is between 0.0001 , P , 0.05, which represents a likely difference.
Function of target of wingless in Planar Cell Polarity899
of the Tow overexpression phenotype by the introduc-
tion of the zip mutant chromosome, reduction in the ck
dose suppressed the multiple-wing-hair phenotype due
independent alleles, ck13and ck07130(Table 2).
Tow is localized in the nucleus in pupal wing cells
and overexpression of Tow downregulates the level of
sqh mRNA: GFP was fused to the C terminus of Tow to
generate a tool for studying the subcellular localization
of the Tow protein. The tow-lacZ signal was uniformly
distributed in the pupal wing blade, so we expressed
Tow-GFP under the control of the ubiquitous promoter
distribution in pupal wings at various times between 24
and 36 hr APF. The GFP signalin pupalwingswas always
found in what we suspected to be nuclei. To confirm
this, we colocalized arm-Tow-GFP and a nuclear fluores-
cent protein (using the UAS-RedStinger line, which ex-
presses a variant red fluorescent protein with a nuclear
localization signal under the control of UAS). Indeed,
the two signals colocalized, consistent with Tow being
primarily a nuclear protein in the pupal wing cells
Due to the lack of the mutant phenotype for tow, we
could not test the function of Tow-GFP by simply look-
ingatthe rescueofatowmutation.Hence, weexpressed
mutation can suppress the multiple-hair-cell phenotype
caused by the hs-Gal4-driven overexpression of Tow by
about two-thirds (Table 2). Introduction of the arm-tow-
GFP chromosome slightly but significantly increases the
number of multiple hair cells, so we assume that arm-
Tow-GFP mimics endogenous Tow function. However,
we note that we still cannot rule out the possibility that
arm-Tow-GFP also localized to ectopic sites where Tow is
not normally present.
The nuclear localization of Tow-GFP led us to test
whether the multiple-wing-hair phenotype caused by
Tow overexpression could be due to repression of the
mRNA level of one or more components of the Drok
pathway. Total RNA was isolated from the hs-Gal4 UAS-
tow pupae to which heat shock was given 24 hr APF, and
Northern hybridization results showed that the mRNA
level of sqh, a substrate of Drok, was repressed in the
Tow-overexpressing flies (Figure 7B). Western blotting
experiments confirmed that Sqh protein and phos-
phorylated Sqh level were also repressed in the Tow-
overexpressing flies (data not shown). On the other
hand, no difference in mRNA level was detected in
homozygous tow754or towfzmutants (data not shown), or
trans-heterozygotes ofthese(Figure 7B),consistent with
the normal phenotypes of those flies.
A novel gene whose expression is regulated by the
canonical Wg-signaling pathway has a function in PCP
signaling: tow was identified in an enhancer trap screen,
and its expression pattern in the wing imaginal disc
was the opposite to that of wg. The expression of tow
was negatively regulated by Wg signaling: When Wg was
omously, and when Wg signaling was blocked by misex-
pression of a dominant-negative form of dTCF (van de
Wetering et al. 1997), the expression of tow was dere-
pressed in the Gal4-expressing domain. The expression
of wg or its receptor Dfz2 was not affected by tow, which
suggested that tow is a downstream target of Wg sig-
naling. From its expression pattern and the regulation
by Wg signaling, we expected that tow might function in
the Wg-signaling pathway. Our data, however, showed
that Tow has little, if any, function in Wg signaling. The
only evidence supporting a role for tow in Wg signaling
was that the overexpression of Tow in the wing resulted
in several extra bristles at the anterior wing margin and
in the wing blade near the distal margin. The spec-
ification of sensory bristle cells is dependent on wg ac-
tivity (Phillips and Whittle 1993; Couso et al. 1994).
Since the expression of wg or Dfz2 was not affected
by the overexpression of Tow, this extra bristle phe-
notype might be caused by Tow in some way regulating
the expression of one or more downstream genes. On
Figure 7.—Tow protein is localized in the nu-
cleus, and overexpression of Tow represses sqh
mRNA. (A) Subcellular localization of Tow in pu-
pal wing cells of arm-tow-GFP. (A-i) GFP signal
(green) showing subcellular localization of Tow
protein. (A-ii) ap-Gal4-driven UAS-RedStinger
wing, which results in the red fluorescent protein
in the nucleus. A-i and A-ii are merged in A-iii.
GFP and nuclear signal show the same localiza-
tion patterns. (B) hs-Gal4-driven overexpression
of Tow represses the level of sqh mRNA.
900S. Chung et al.
the other hand, surprisingly, we obtained evidence
for tow acting in Fz-dependent planar cell polarity
dependent polarity pathway have indicated that diver-
gent signaling mechanisms are involved (Axelrod et al.
1998; Boutros et al. 1998). Although both the Wg- and
Fz-signaling pathways require the function of Dsh, they
demonstrate differential requirements for more down-
stream factors. Our study, however, showed that tow
expression was negatively regulated by Wg signaling
in wing discs and that it functioned in the regulation of
the actin cytoskeleton and hair morphogenesis dur-
ing PCP signaling. In both the wing and the eye, the
interactions suggested that Tow functioned as a down-
stream regulator/component of PCP signaling and the
Fz/Dsh pathway. The significance of the regulation of
tow expression by Wg in wing discs for the function of
tow during hair morphogenesis is unclear as these two
events are temporally separated. Further studies will be
required to determine this.
Interactions between tow and other PCP genes: We
tested for genetic interactions between tow and other
PCP genes and found that the reduction in tow dose
enhanced the multiple-hair-cell phenotype of a domi-
nant stan allele. This interaction is the opposite of what
we saw betweentow and dsh.The basis for thisdifference
is unclear but might be due to Tow affecting the expres-
sion of different target genes for these two phenotypes.
That tow mutations could interact with PCP genes by
altering gene expression would not be surprising. The
PCP system is sensitive to the relative level of compo-
nents, and interactions were seen previously between
PCP genes and mutations in the grainy head transcrip-
tion factor (Lee and Adler 2004). Our experiments
provided evidence for tow regulating sgh expression
and this could explain at least part of the interaction
The sha gene encodes a putative actin-binding pro-
tein whose expression increases dramatically just prior
pression of sha playing a key role in regulating hair
initiation, shamutant wing cells either fail toform ahair
or have delayed hair formation. Those cells that form
delayed hairs produce multiple short ones. Previously,
it was found that decreases in PCP gene function
enhanced the sha mutant phenotype. We found that
reduction in tow dose also enhanced a sha mutant phe-
notype. This could be due to an indirect effect of tow on
the PCP pathway and/or to tow regulating sha expres-
sion directly or by an alternative mechanism.
Proper level of Tow is important in regulation of the
actin cytoskeleton: Mutations in a number of genes
involved in PCP signaling result in multiple-wing-hair
phenotypes as well as a swirling pattern of hairs on the
Drosophila wing. Drok is thought to act as an effector of
small GTPase RhoA and to link Fz/Dsh signaling to the
actin cytoskeleton (Winter et al. 2001). Drok mutations
alter only the number of hairs produced and not the
hair orientation; thus these two processes likely diverge
upstream of Drok.
The gain-of-function phenotype of Tow (distally ori-
Drok pathway. Genetic interactions revealed that Drok
and tow have an antagonistic relationship. Increasing
Drok expression suppresses and decreasing Drok ex-
pression enhances the multiple-wing-hair phenotype
caused by Tow overexpression. RhoA is upstream of
Drok and functions to activate Drok. Thus, it is not
surprising that overexpression of RhoA also suppresses
the Tow multiple-hair-cell phenotype while mutation of
RhoA enhances it. The phosphorylation of Sqh/MRLC
by Drok leads to Sqh activation and, consistent with
this, a reduction in sqh dose enhances the Tow multiple-
hair-cell phenotype.The two cellular myosin genes zip
(myosin II) and ck (myosin VIIA) also interact with Tow
in a manner consistent with the interactions of these
genes with Drok. These data suggest that the balance
between Drok and Tow is important for the regulation
of the actin cytoskeleton.
The arm-tow-GFP transgene revealed that Tow is local-
ized in the nucleus in pupal wings. From the subcellular
localization of Tow, we assumed that it might influence
the expression of components involved in the Drok
pathway. The overexpression of Tow downregulates sqh,
a substrate of Drok. This leads to a decrease in the level
of Sqh and phosphorylated Sqh protein. This providesa
putative mechanism that can explain the multiple-hair-
cell phenotype caused by Tow overexpression, which is
similar to the Drok mutant phenotype. In a Drok mutant,
thelevel ofphosphorylated Sqhwas reduced,indicating
Sqh phosphorylation (Winter et al. 2001). Genetic in-
teractions of tow with Drok pathway components also fit
such a model. The mutation of genes in the Drok
pathway enhanced the multiple-hair-cell phenotype
caused by Tow overexpression, and the overexpression
or the introduction of transgenes of Drok pathway com-
We conclude that the proper level of Tow (and/or the
relative level of Tow and Drok) is important in the
regulation of the actin cytoskeleton. However, it re-
mains to be resolved whether the downregulation of sqh
a tow mutant phenotype suggests that there could be
other genes whose function in the regulation of sqh is
redundant with tow.
and the Harvard Exelixis Stock Collection for providing the fly stocks
and the Developmental Studies Hybridoma Bank for the antibody.
This research was supported by a grant (M103KV010002 04K2201
00230) from the Brain Research Center of the 21st Century Frontier
Research Program funded by the Ministry of Science and Technology,
General Medical Science (GM-37136) to P.N.A.
Function of target of wingless in Planar Cell Polarity 901
Adler, P. N., 2002
ila. Dev. Cell 2: 525–535.
Adler, P. N., and H. Lee, 2001
actions in planar polarity. Curr. Opin. Cell Biol. 13: 635–640.
Adler, P. N., J. Taylor and J. Charlton, 2000
non-aunotomy of frizzled and Van Gogh clones in the Drosophila
wing is a consequenceof disruption in local signaling.Mech. Dev.
Axelrod, J. D., J. R. Miller, J. M. Shulman, R. T. Moon and N.
Perrimon, 1998 Differential recruitment of Dishevelled pro-
vides signaling specificity in the planar cell polarity and Wingless
signaling pathways. Genes Dev. 12: 2610–2622.
Bhanot, P., M. Brink, C. Harryman Samos, J.-C. Hsieh, Y. Wang
et al., 1996A new member of the frizzled family from Drosophila
functions as a Wingless receptor. Nature 382: 225–230.
Bier, E., H. Vaessin, S. Shepherd, K. Lee, K. McCall et al.,
1989 Searching for pattern and mutation in the Drosophila ge-
nome with a P-lacZ vector. Genes Dev. 3: 1273–1287.
Boutros, M., N. Paricio, D. I. Strutt and M. Mlodzik, 1998
hevelled activates JNK and discriminates between JNK pathways
in planar polarity and wingless signaling. Cell 94: 109–118.
Brand, A. H., and N. Perrimon, 1993
means of altering cell fates and generating dominant pheno-
types. Development 118: 401–415.
Brunner, E., O. Peter, L. Schweizer and K. Basler, 1997
encodes a Lef-1 homologue that acts downstream of Armadillo to
transduce the Wingless signal in Drosophila. Nature 385: 829–833.
repression of Drosophila frizzled 2 expression shapes the wingless
morphogen gradient in the wing. Cell 93: 767–777.
Capdevila, J., and I. Guerrero, 1994
signaling molecule decapentaplegic induces pattern duplica-
tions and growth alterations in Drosophila wings. EMBO J. 13:
Chae, J., M. Kim, J. H. Goo, S. Collier, D Gubb et al., 1999
Drosophila tissue polarity gene starry night encodes a member
of the protocadherin family. Development 126: 5421–5429.
Collier, S., and D. Gubb, 1997Drosophila tissue polarity requires
the cell-autonomous activity of the fuzzy gene, which encodes a
novel transmembrane protein. Development 124: 4029–4037.
Couso, J. P., S. A. Bishop and A. Martinez Arias, 1994
less signaling pathway and the patterning of the wing margin in
Drosophila. Development 120: 621–636.
Fanto, M., U. Weber, D. I. Strutt and M. Mlodzik, 2000
signaling by Rac and Rho GTPases is required in the establish-
ment of epithelial planar polarity in the Drosophila eye. Curr.
Biol. 10: 979–988.
Feiguin, F., M. Hannus, M. Mlodzik and S. Eaton, 2001
kyrin repeat protein Diego mediates Frizzled-dependent planar
polarization. Dev. Cell 1: 93–101.
Golic, K. G., 1991Site-specific recombination between homolo-
gous chromosomes in Drosophila. Science 252: 958–961.
Gubb, D., 1993Genes controlling cellular polarity in Drosophila.
Dev. Suppl., 269–277.
Gubb, D., and A. Garcia-Bellido, 1982
termination of cuticular polarity during development in Drosoph-
ila melanogaster. J. Embryol. Exp. Morphol. 68: 37–57.
Gubb, D., M. Shelton, J. Roote, S. McGill and M. Ashburner,
1984The genetic analysis of a large transposing element of Dro-
sophila melanogaster. The insertion of a w1 rst1 TE into the ck lo-
cus. Chromosoma 91: 54–64.
Gubb, D., C. Green, D. Huen, D. Coulson, G. Johnson et al.,
1999 The balance between isoforms of the Prickle LIM domain
protein is critical for planar polarity in Drosophila imaginal discs.
Genes Dev. 13: 2315–2327.
He, B., and P. N. Adler, 2002 The genetic control of arista lateral
morphogenesis in Drosophila. Dev. Genes Evol. 212: 218–229.
Jordan, P., and R. Karess, 1997Myosin light chain-activating phos-
phorylation sites are required for oogenesis in Drosophila. J. Cell
Biol. 139: 1805–1819.
Karess, R. E., X. J. Chang, K. A. Edwards, S. Kulkarni, I. Aguilera
et al., 1991The regulatory light chain of nonmuscle myosin is
encoded by spaghetti-squash, a gene required for cytokinesis in
Drosophila. Cell 65: 1177–1189.
Planar signaling and morphogenesis in Drosoph-
Frizzled signaling and cell–cell inter-
Targeted gene expression as a
Targeted expression of the
A genetic analysis of the de-
Katanaev, V. L., R. Ponzielli, M. Semeriva and A. Tomlinson,
2005Trimeric G protein-dependent Frizzled signaling in Dro-
sophila. Cell 120: 111–122.
Kim, S., S. Chung, J. Yoon, K.-W. Choi and J. Yim, 2006
pression of Tollo/Toll-8 antagonizes Dpp signaling and induces
cell sorting in Drosophila wing. Genesis 44: 541–549.
Klingensmith, J., R. Nusse and N. Perrimon, 1994
segment polarity gene dishevelled encodes a novel protein re-
quired for response to the wingless signal. Genes Dev. 8: 118–130.
Krasnow, R. E., and P. N. Adler, 1994
a dual function in tissue polarity. Development 120: 1883–1893.
Krasnow, R. E, L. L. Wong and P. N. Adler, 1995
component of the frizzled signaling pathway in Drosophila. Devel-
opment 121: 4095–4102.
Lee, H., and P. N. Adler, 2004The grainy head transcription factor is
essential for the function of the frizzled pathway in the Drosophila
wing. Mech. Dev. 121: 37–49.
Lee, T., and L. Luo, 1999Mosaic analysis with a repressible cell
marker for studies of gene function in neuronal morphogenesis.
Neuron 22: 451–461.
Lee, T., C. Winter, S. S. Marticke, A. Lee and L. Luo,
2000Essential roles of Drosophila RhoA in the regulation of
neuroblast proliferation and dendritic but not axonal morpho-
genesis. Neuron 25: 307–316.
Logan, C. Y., and R. Nusse, 2004
velopment and disease. Annu. Rev. Cell Dev. Biol. 20: 781–810.
Neumann, C. J., and S. M. Cohen, 1997
less organizes the dorsal-ventral axis of the Drosophila wing. De-
velopment 124: 871–880.
Noordermeer, J., J. Klingensmith, N. Perrimon and R. Nusse,
dishevelled and armadillo act in the Wingless signaling path-
way in Drosophila. Nature 367: 80–83.
Paricio, N., F. Feiguin, M. Boutros, S. Eaton and M. Mlodzik,
1999The Drosophila STE20-like kinase Misshapen is required
downstream of the Frizzled receptor in planar polarity signaling.
EMBO J. 18: 4669–4678.
Park, W. J., J. Liu, E. J. Sharp and P. N. Adler, 1996
tissue polarity gene inturned acts cell autonomously and encodes
a novel protein. Development 122: 961–969.
Parks, A. L., K. R. Cook, M. Belvin, N. A. Dompe, R. Fawcett, et al.
2004 Systematic generation of high-resolution deletion cover-
age of the Drosophila melanogaster genome. Nat. Genet. 36:
Phillips, R. G., and J. R. S. Whittle, 1993
diates determination of peripheral nervous system elements in
late stages of Drosophila wing disc development. Development
Povelones, M., R. Howes, M. Fish and R. Nusse, 2005
idence that Drosophila frizzled controls planar cell polarity and
armadillo signaling by a common mechanism. Genetics 171:
Ren, N., B. He, D. Stone, S. Kirakodu and P. N. Adler, 2006
shavenoid gene of Drosophila encodes a novel actin cytoskeleton
interacting protein that promotes wing hair morphogenesis. Ge-
netics 172: 1643–1653.
Self, T., M. Mahony, J. Fleming, J. Walsh, S. D. M. Brown et al.,
1998 Shaker-1 mutations reveal roles for myosin VIIA in both
development and function of cochlear hair cells. Development
Shulman, J. M., N. Perrimon and J. D. Axelrod, 1998
naling and the developmental control of cell polarity. Trends
Genet. 14: 452–458.
Spradling, A., and G. Rubin, 1982
ments into Drosophila germline chromosomes. Science 218:
Strigini, M., and M. Cohen, 2000
the Drosophila wing. Curr. Biol. 10: 293–300.
ment of cell polarity in the Drosophila wing. Dev. Cell 7: 367–375.
Strutt, D. I., U. Weberand M. Mlodzik, 1997
tissue polarity and Frizzled signaling. Nature 387: 292–295.
Strutt, H., and D. Strutt, 1999
sophila eye. Curr. Opin. Genet. Dev. 9: 442–446.
Tautz, D., and C. Pfeifle, 1989A non-radioactive in situ hybridiza-
tion method for the localization of specific RNAs in Drosophila
A single Frizzled protein has
Dishevelled is a
The Wnt signaling pathway in de-
Long-range action of Wing-
wingless expression me-
Transposition of cloned P ele-
Wingless gradient formation in
The role of RhoA in
Polarity determination in the Dro-
902S. Chung et al.
embryos reveals translational control of the segmentation gene Download full-text
hunchback. Chromosoma 98: 81–85.
Taylor, J., N. Abramova, J. Charlton and P. N. Adler, 1998
Gogh: a new Drosophila tissue polarity gene. Genetics 150:
Theisen, H., J. Purcell, M. Bennett, D. Kansagara, A. Syed et al.,
dishevelled is required during wingless signaling to establish
both cell polarity and cell identity. Development 120: 347–360.
Tomlinson, A., and G. Struhl, 1999
tion from a gradient: sequential roles of the receptors Frizzled
and Notch in establishing planar polarity in the Drosophila
eye. Development 126: 5725–5738.
Tree, D., J. M. Shulman, R. Rousset, M. P. Scott, D. Gubb et al.,
2002Prickle mediates feedback amplification to generateasym-
metric planar cell polarity signaling. Cell 109: 371–381.
Turner, C. M., and P. N. Adler, 1998
microtubule cytoskeletons in the morphogenesis of epidermal
hairs during wing development in Drosophila. Mech. Dev. 70:
Usui, T., Y. Shima, Y. Shimada, S. Hirano, R. W. Burgess et al.,
1999Flamingo, a seven-pass transmembrane cadherin, regu-
lates planar cell polarity under the control of Frizzled. Cell 98:
et al., 1997Armadillo coactivates transcription driven by the
product of the Drosophila segment polarity gene dTCF. Cell 88:
Vincent, J. P., C. H. Girdham and P. H. O’Farrell, 1994
autonomous, ubiquitous marker for the analysis of Drosophila
genetic mosaics. Dev. Biol. 164: 328–331.
Decoding vectorial informa-
Distinct roles for the actin and
Vinson, C. R., and P. N. Adler, 1987
and the transmission of polarity information by the frizzled gene
of Drosophila. Nature 329: 549–551.
Vinson, C. R., S. Conover and P. N. Adler, 1989
sue polarity locus encodes a protein containing seven potential
transmembrane domains. Nature 338: 263–264.
Weber, U., N. Paricio and M. Mlodzik, 2000
Frizzled-induced R3/R4 cell fate distinction and planar polarity
determination in the Drosophila eye. Development 127: 3619–
Winter, C. G., B. Wang, A. Ballew, A. Royou, R. Karess et al.,
2001 Drosophila Rho-associated kinase (Drok) links Frizzled-
mediated planar cell polarity signaling to the actin cytoskeleton.
Cell 105: 81–91.
Wolff, T., and G. M. Rubin, 1998
ulate tissue polarity and cell fate decisions in Drosophila. Devel-
opment 125: 1149–1159.
Wong, L. L., and P. N. Adler, 1993
sophila regulates the subcellular location for prehair initiation
in pupal wing cells. J. Cell Biol. 123: 209–221.
Xu, T., and G. M. Rubin, 1993Analysis of genetic mosaics in de-
veloping and adult Drosophila tissues. Development 117:
Zecca, M., K. Basler and G. Struhl, 1996
action of a Wingless morphogen gradient. Cell 87: 833–844.
Zheng, L., J. Zhang and R. W. Carthew, 1995
ror-symmetric pattern formation in the Drosophila eye. Develop-
ment 121: 3045–3055.
Directional non-cell autonomy
A Drosophila tis-
strabismus, a novel gene that reg-
Tissue polarity genes of Dro-
Direct and long-range
frizzled regulates mir-
Communicating editor: A. J. Lopez
Function of target of wingless in Planar Cell Polarity 903