a targeted gain-of-function screen identifies genes affecting … · 2009-03-30 · sion screen in...
TRANSCRIPT
Copyright � 2009 by the Genetics Society of AmericaDOI: 10.1534/genetics.108.094052
A Targeted Gain-of-Function Screen Identifies Genes AffectingSalivary Gland Morphogenesis/Tubulogenesis in Drosophila
Vanessa Maybeck1 and Katja Roper2
Department of Physiology, Development and Neuroscience, University of Cambridge, Cambridge CB2 3DY, United Kingdom
Manuscript received July 30, 2008Accepted for publication December 3, 2008
ABSTRACT
During development individual cells in tissues undergo complex cell-shape changes to drive themorphogenetic movements required to form tissues. Cell shape is determined by the cytoskeleton andcell-shape changes critically depend on a tight spatial and temporal control of cytoskeletal behavior. Wehave used the formation of the salivary glands in the Drosophila embryo, a process of tubulogenesis, as anassay for identifying factors that impinge on cell shape and the cytoskeleton. To this end we have performeda gain-of-function screen in the salivary glands, using a collection of fly lines carrying EP-element insertionsthat allow the overexpression of downstream-located genes using the UAS-Gal4 system. We used a salivary-gland-specific fork head-Gal4 line to restrict expression to the salivary glands, in combination with reportersof cell shape and the cytoskeleton. We identified a number of genes known to affect salivary glandformation, confirming the effectiveness of the screen. In addition, we found many genes not implicatedpreviously in this process, some having known functions in other tissues. We report the initial charac-terization of a subset of genes, including chickadee, rhomboid1, egalitarian, bitesize, and capricious, throughcomparison of gain- and loss-of-function phenotypes.
DURING development and organogenesis mosttissues arise from layers of epithelial cells that re-
organize through complex morphogenetic movements.Many adult organs consist of tubular arrangements ofepithelial sheets, and these tubules form during devel-opment through a process called tubulogenesis. Thereare a number of ways to generate tubules (Lubarsky
and Krasnow 2003). One important process is the directconversion of epithelial sheets into tubules throughwrapping (Colas and Schoenwolf 2001) or budding(Hogan and Kolodziej 2002). Cells undergoing tubulo-genesis change their shapes drastically from a cuboidalor columnar epithelial shape to a wedge shape or conicalshape and then back to a more columnar epithelialshape once positioned inside the tube. Cell shape isdetermined by the intracellular cytoskeleton, primarilyactin and microtubules. The cytoskeleton is closelycoupled to cell–cell adhesion as well as to adhesion tothe extracellular matrix. We are interested in under-standing how the cytoskeleton and thus cell shape isregulated and coordinated during tubulogenesis.
We chose to perform a gain-of-function screen ratherthan a mutagenesis-based loss-of-function screen asphenotypes observed in the latter might be subtle andthus missed or phenotypes in a given tissue might be
obscured by disruption of other tissues and many genesmight also have redundant functions. In contrast, thegain-of-function/overexpression approach allows a par-ticular tissue and gene to be targeted, and many suchscreens have been successfully conducted in the past(for examples, see Rørth et al. 1998; Molnar et al. 2006;Bejarano et al. 2008). The screen presented here usesthe formation of the salivary glands in the Drosophilaembryo as an assay system. The screen is based on acollection of transposable elements (EP elements) gen-erated by Rørth et al. (1998) that contain UAS sites thatrespond to the yeast transcription factor Gal4 that isfollowed by a promoter directing expression, whenactivated, of genes located downstream 39 of the EPinsertion site. If combined through crosses with a tissue-specific source of Gal4 (Henderson and Andrew 2000;Zhou et al. 2001), overexpression (and in some casesantisense expression) of a downstream gene will beactivated only in the target tissue, which in our case arethe embryonic salivary glands in the Drosophila embryo.
Salivary gland formation in Drosophila is probablythe simplest form of tubulogenesis (Lubarsky andKrasnow 2003). A patch of �200 cells in the ventralepidermis of the embryo within parasegment 2 becomesspecialized to form a salivary gland primordium, theplacode, with 100 cells on either side of the embryo.This fate determination occurs through a combinationof the activities of the homeotic genes sex combs reduced(scr), extradenticle (exd), and homothorax (hth) and dorsalsignaling by decapentaplegic (dpp) (Panzer et al. 1992;
1Present address: Institute of Neuroscience and Biophysics, MolecularBiophysics (INB-2), Research Center Julich, D-52425 Julich, Germany.
2Corresponding author: Department of Physiology, Development andNeuroscience, Downing St., University of Cambridge, Cambridge CB23DY, United Kingdom. E-mail: [email protected]
Genetics 181: 543–565 (February 2009)
Henderson et al. 1999; Henderson and Andrew 2000).Without scr, exd, and hth function, no salivary glandsform. Different subpopulations of cells are found in theinvaginated gland, such as the secretory cells and thecommon and individual duct cells. Their distinctiondepends on EGF signaling from the ventral midline(Kuo et al. 1996; Haberman et al. 2003). Once the cellshave become specialized at stage 10 of embryogenesis,no further cell division occurs within the primordium,and no cells are lost through apoptosis (Campos-Ortega
and Hartenstein 1985; Bate and Martinez Arias
1993; Myat and Andrew 2000a). Invagination initiatesin the dorsal posterior corner of the primordium, withall future secretory cells invaginating in a precise order,followed by invagination of the duct cells and formationof the ducts (Myat and Andrew 2000b). A key geneessential for the invagination is fork head ( fkh). Fkh isa winged-helix transcription factor, and in its absenceall of the cells fated to form the glands remain on thesurface of the embryo as they fail to undergo apicalconstriction (Myat and Andrew 2000a). Once in-side the embryo, the glands have to navigate their waythrough the surrounding tissues, including the visceralmesoderm and central nervous system, to reach theirextended final position parallel to the midline andanterior–posterior axis. They are guided by cues fromthe surrounding tissues (Kolesnikov and Beckendorf
2005; Harris and Beckendorf 2007; Harris et al.2007). Also, after initially invaginating in a posterior–dorsal direction, the glands turn and further extendinto the embryo in a direction parallel to the anterior–posterior embryonic axis in a process dependent onintegrins and downstream signals (Bradley et al. 2003;Vining et al. 2005).
A few factors that impinge on the cytoskeleton andcell shape during salivary gland morphogenesis havepreviously been identified. The actin cytoskeleton ismodified through proteins such as Btk29/Tec29 inconjunction with Chickadee (Chandrasekaran andBeckendorf 2005). Small GTPases such as Rac and Rhoaffect the invagination of the glands (Pirraglia et al.2006; Xu et al. 2008). Crumbs and Klarsicht affect thedelivery of apical membrane and thus cell shape at latestages of morphogenesis (Myat and Andrew 2002).Nonetheless, how these factors work together through-out the whole process of invagination is still not clear,and it is likely that many others remain to be identified.
We have performed a gain-of-function/overexpres-sion screen in the salivary glands with the aim of, first,identifying more genes that are required for salivarygland tubulogenesis (and thus potentially also fortubulogenesis in general) and, second, using this systemas an assay for factors affecting the cytoskeleton andthus cell shape in general. The first aim assumes thatgenes that have a function in the morphogenesis of theglands and are endogenously expressed in the glandsmight perturb their invagination if overexpressed and if
levels of expression are important. The second aimhypothesizes that overexpression of genes not endoge-nously expressed in the glands but important for cell-shape coordination in other tissues will lead to identifiablephenotypes in this screen, as defects in cell-shape changesresulting from the overexpression will affect the properinvagination of the glands. The orientation of some ofthe EP elements is also likely to lead to (over)expressionof an antisense RNA, thus potentially inducing a tissue-specific loss-of-function effect. We identified seven genesthat have previously been implicated in salivary glandmorphogenesis or function, confirming the effective-ness of the screen, and also 44 insertions that uncovergenes with potentially novel roles in the salivary glandsor functions in the regulation of cell shape and thecytoskeleton in other tissues. Of these genes, 14 arepreviously uncharacterized genes. A selection of thegenes that fall into the three categories discussed above(i.e., overexpression of a gene with a function in theglands; overexpression of a gene not expressed in theglands, revealing a function in cell-shape coordinationin other tissues; and loss of function of a gene throughtissue-specific antisense RNA expression) and recoveredin the screen is examined in more detail below, in-cluding bitesize, egalitarian, chickadee, capricious, andrhomboid1.
MATERIALS AND METHODS
Screen design and fly husbandry: fkhGal4 (Henderson andAndrew 2000; Zhou et al. 2001) was recombined on the thirdchromosome with a UAS construct containing GFP fused tothe N terminus of the EF-Gas2 region of Shot (Subramanian
et al. 2003) or the membrane-targeting domain of src fused toGFP (Kaltschmidt et al. 2000). One or the other of thesemarker lines were crossed to 1001 EP lines from the Rørthcollection obtained from the stock centers in Szeged (secondand third chromosomes; http://expbio.bio.u-szeged.hu/fly/index.php) and Bloomington (third chromosome; http://flystocks.bio.indiana.edu/). The fkhGal4 insertion was a giftfrom Deborah Andrew, the SrcGFP from Nick Brown. UAS-chickadee was from Lynn Cooley. rho[PD5], argos[lD7], flb[ik35],UAS-argos, UAS-CA-EGFR, UAS-sspi, UAS-caps, caps[PB1], trn[28.4],caps[Del1] trn[28.4] alleles were gifts from Matthew Freeman;egl[3e], egl[PR29], BicD[HA40], b BicD[18a], dp b Df(2L)TW119and UAS-egl were gifts from Simon Bullock (the UAS-egl trans-gene leads to an approximately threefold increase in levels;S. Bullock, personal communication). To analyze egl mutantembryos, egl[3e]/egl[WU50] females were mated to egl[PR29]/1males, and to analyze BicD mutant embryos, BicD[HA40]/1; bBicD[18a]/dp b Df(2L)TW119 mothers were mated to BicD[18a]/CyO males. In the detailed analyses (apart from the cases of egland BicD mutant embryos), mutant embryos were identified bythe absence of green balancer (balancer chromosomes usedwere CyO KrTGFP, TM3 Sb Ser twi-gal4 UAS-2x eGFP and TM6bTb Sb df-Gal4 UAS-YFP). All other stocks used were from theBloomington Stock Center. Crosses were maintained at 25� oncornmeal food, and embryos were collected on apple or grape-juice–agar plates.
EP lines were determined to be heterozygous or homozy-gous for the EP insertion. In the absence of visible balancerchromosomes, lines were assumed to be homozygous. Homo-
544 V. Maybeck and K. Roper
zygous lines were crossed to a homozygous driver line, andembryos were collected overnight on apple or grape-juiceplates with yeast paste. Twenty embryos between stages 10 and13 and 20 embryos between stages 13 and 15 (an ‘‘early’’ and a‘‘late’’ sample) were scored in live mounts in halocarbon oil(Halocarbon Oil 27, Sigma) after dechorionation in 50%bleach. In heterozygous balanced lines, 40 embryos in eachof the early and late groups were scored, assuming equalfertilization and survival from both genotypes through the endof embryogenesis. Lines with $20% salivary gland defects, orwith potential defects that would require quantitative analysis(i.e., changes in length), were subjected to second-pass screen-ing. In the second pass, embryos were collected as above, fixedin 2:1 heptane:4% formaldehyde in PBS and stained withrhodamine phalloidin. A larger number of embryos werescored from these collections (average .80). First-pass hitswere also crossed to w f flies, and embryos were collected andstained with rhodamine phalloidin as above to check fordominant positional effects of the EP insertion.
Immunohistochemistry, wide-field fluorescence, and con-focal analysis: Embryos were collected on grape-juice platesand processed for immunofluorescence using standard pro-cedures. Briefly, embryos were dechorionated in 50% bleach,fixed in 4% formaldehyde, and stained with phalloidin orprimary and secondary antibodies in PBT (PBS plus 0.5%bovine serum albumin and 0.3% Triton X-100). Crumbs andDE-Cadherin antibodies were obtained from the Develop-mental Studies Hybridoma Bank at the University of Iowa; theShot antibody was raised in our lab and is identical in design tothe one described in Strumpf and Volk (1998); the anti-dCrebA antibody was from Deborah Andrew (Andrew et al.1997); the antiphosphohistone H3 and anti-GFP antibodieswere from Abcam. Secondary antibodies used were AlexaFluor 488 coupled (Molecular Probes) and Cy3 and Cy5coupled (Jackson ImmunoResearch Laboratories), and rho-damine–phalloidin was from Molecular Probes. Samples wereembedded in Vectashield (Vector Laboratories). Wide-fieldfluorescence images documenting the screen results wereobtained on a Leica DMR (equipped with a MicroFire camera,Optronics), a Zeiss Axioplan 2 (equipped with a PrincetonInstruments camera), and a Zeiss Axioskop Mot 2 (equippedwith a Jenoptik C14 camera), using PictureFrame, Meta-morph, and Openlab software, respectively. Confocal imageswere obtained using an Olympus Fluoview 1000. Confocallaser, iris, and amplification settings in experiments compar-ing intensities of labeling were set to identical values. Wide-field fluorescence and confocal images were assembled inAdobe Photoshop, and confocal z-stacks and z-stack projec-tions were assembled in Image J.
In situ hybridization: In situ hybridization of whole-mountembryos was performed essentially as described by Tautz andPfeifle 1989). To combine the in situ protocol with immuno-histochemistry for GFP, the anti-GFP antibody was incubatedtogether with the anti-DIG antibody, followed by fluorescentsecondary antibody incubation to reveal the GFP after theBCIP/NBT color reaction. Images were obtained on a LeicaDMR (equipped with a MicroFire camera, Optronics) andcomposites were assembled using Adobe Photoshop.
The following primers were used to generate in situ probes:chic 59-TTTCCATCTACGAGGATCCC, chic 39-ATTTCGTTCAAAGCTGAGGAC; caps 59-CGGGCAATTACCATGTCGTTG,caps 39-GATGTGGCTGATGCGATTCTG; trn 59-GTGGGCATCTGGTGCATTTTG, and trn 39-GATAAAGGATGCGCAACTGGG.For the btsz probe, the cDNA clone AY229970 was used totranscribe antisense and sense probes.
Statistics: We determined the base rate of salivary glanddefects observable in our experimental stocks by countingdefects in the genotype 1/1; 1/CyO; fkhGal4TUAS-GFPmarker/1.
The base rate in this genetic background was determined as4.3% (n ¼ 748). A similar base rate was obtained in a geneticbackground where the CyO balancer chromosome was re-placed by the GFP-marked chromosome 1/1;1/btlGal4TUAS-GFP; fkhGal4TUAS-GFPmarker/1; the base rate was 4.4% (n ¼878). Thus, we exclude any effect at least of a CyO balancechromosome present on salivary gland morphogenesis.
To set a cutoff level for the rate of affected salivary glandscounted in each experiment, above which we determinedthat EP elements driven by fkhGal4 affect the salivary glandmorphogenesis, we chose an arbitrary 20% defects as a cutofffor the first-pass analysis. This yielded 187 EP lines (18.6% ofthe total lines screened) to be rescreened in the second-passanalysis, yielding 51 confirmed insertions overall. This equals�5% of the total number of EP lines screened, which alsoequals 2 SD from the mean of a normal distributed sample,indicating that we set our cutoff at a sensible level.
RESULTS
Experimental design of the gain-of-function screen:To address how the cytoskeleton and cell shape isregulated during such a process of tubulogenesis, weperformed a gain-of-function screen in the salivaryglands of the Drosophila embryo. We used a gland-specific Gal4-driver, fkhGal4 (Henderson and Andrew
2000; Zhou et al. 2001), to drive expression of either amarker of the microtubule cytoskeleton, GFP-EFGas2(Subramanian et al. 2003), or a marker of cell shape,SrcGFP (Kaltschmidt et al. 2000), in the glands only.Flies carrying these marker chromosomes (GFP-EFGas2or SrcGFP marker plus fkhGal4: marker line) werecrossed to a collection of EP-element lines containingUAS elements, leading to the expression of gene Xlocated 39 downstream from the EP-element insertionsite. We drove expression from 1001 EP elements spe-cifically in the salivary glands and screened for anyapparent problems in their morphogenesis (see Figure1 for wild-type morphogenesis and marker expressionand Figure 2A for a scheme explaining the screensetup). It has previously been shown that the properinvagination and positioning of the salivary glands de-pends on the surrounding tissues such as the visceralmesoderm (Vining et al. 2005). The tissue-specific ex-pression of genes in the screen allowed us to identifyfactors that acted within the glands themselves and didnot affect functioning of the surrounding tissues, thusgiving a phenotype due to a secondary defect.
We crossed flies of marker to flies carrying an EPinsertion on the second or third chromosome (seescheme in Figure 2A). The resulting offspring overex-pressed a gene X specifically in the salivary glands.These embryos were collected early (stages 10–13) andlate (stages 13–15) during embryogenesis and analyzedlive for any apparent defect in salivary gland invagina-tion, positioning, and shape of salivary gland cells orgland lumen. When phenotypes were observed in .20%of embryos, embryos were collected again, fixed, andcounterstained for actin using phalloidin to analyze
Salivary Gland Morphogenesis in Drosophila 545
general morphology. Positive insertions were definedas having 20–90% of embryos showing a salivary glandphenotype after the second examination. The baselinerate of obtaining salivary glands with a phenotype inembryos expressing GFP-EFGas2 or SrcGFP in the glandsunder the control of fkhGal4 was �4% (see materials
and methods). In some cases the position and orienta-tion of EP elements would be predicted to lead to theoverexpression of an antisense RNA rather than a coding-sense mRNA. Positive insertions resulting from pre-sumed antisense RNA expression are indicated as suchin Table 1.
Phenotypes identified in the screen: We screened1001 EP-element insertions on the second and thirdchromosome (a list of all lines screened can be found insupplemental Table 1). Overexpression in the salivaryglands of genes located downstream from EP elementsunder the control of fkhGal4 led to a variety of phe-
notypes that could be classified in four major classes(Figure 2): invagination defects (failure to invaginate,wide invagination; see Figure 2, B and C); gland shapeand lumen defects (shepherd’s crook, C-shape,S-shape,wrong length, enlarged lumen, aberrant lumen; seeFigure 2, D–I); positioning defects (wrong position,turning, budding, forking, hook, butterfly; see Figure 2,K–P); and gland subfate defects (no proper duct; seeFigure 2Q). These overall phenotypes suggest that thescreen detected interference at all stages of salivarygland formation.
Although the phenotypes listed above were recur-rently found in the screen, half of the positive insertionsshowed a variable phenotype, combining several ofthese individual phenotypes. The other half showed aconsistent phenotype restricted to one class or even toone specific phenotype (see Table 1, Phenotype insalivary glands). This suggests that, in the cases of genes
Figure 1.—Salivary gland devel-opment visualized using fkhGal4-driven GFP-marker expression.Salivary gland morphogenesisfrom embryonic stages 10–15. (A)Schematic of salivary gland invagi-nation, ventral view. (B–F) Low-magnification confocal sectionsof embryos stained with phalloidintorevealactin(red)andexpressingGFP-EFGas2 under the control offkhGal4 in the salivary glands(green). (C–E) Lateral views. (B)A ventral view. (F) A dorsal view.(G–K9) Close-up confocal sectionsof salivary glands labeled with phal-loidin to reveal actin (red) and ex-pressing SrcGFP under the controlof fkhGal4 (green, and as a singlechannel in G9–K9). G–K show lat-eral views. Note that GFP-EFGas2 la-bels microtubules, whereas SrcGFPis targeted to the membrane andthus reveals cell shape.
546 V. Maybeck and K. Roper
showing variable phenotypes upon overexpression inthe glands, a specific phenotype was not necessarilyreflective of only a certain process failing during theinvagination, but rather indicated that many pertur-bances at the molecular level might lead to similarphenotypes.
Genes identified in the screen: Of 1001 EP linesscreened, 51 showed a phenotype in the salivary glandswhen crossed to fkhGal4, equalling 5.09% of the totallines analyzed. The penetrance of phenotypes variedfrom weak (2.7% of EPs tested) to strong (1.7% of EPstested) to very strong (0.7% of EPs tested; see Table 1).The genes affected could be classified according to theirpredicted function (see Table 1). Several of the genesidentified by positive insertions have previously beenimplicated in salivary gland morphogenesis or func-tion within the glands: chickadee (Chandrasekaran
and Beckendorf 2005), tec29 (Chandrasekaran
and Beckendorf 2005), doughnut on 2 (Harris andBeckendorf 2007), rhomboid1 and spitz (Kuo et al. 1996),
tapd (Abrams and Andrew 2005), and slit (Kolesnikov
and Beckendorf 2005). Three of the insertions identi-fying these could potentially induce antisense RNA ex-pression and could thus mimic a loss-of-function situation(chic, tec29, slit), and three insertions could induce over-expression (dnt, rhomboid1, spitz). Overexpression of dntcould affect the positioning cues that the migratingglands receive, whereas rhomboid1 and spitz overexpressionwould lead to excess Spitz ligand being provided, poten-tially overstimulating EGFR signaling (see below). Thesegenes served as confirmation that factors impinging onsalivary gland morphogenesis were picked up in thisscreen.
The majority of positive insertions (44 of 51 EPs),however, were inserted into genes that have not pre-viously been implicated in salivary gland morphogenesisor in tubulogenesis in general. Several of the encodedproteins have known functions in other tissues in fliessuch as Egalitarian (Navarro et al. 2004), Traf-4 (Cha
et al. 2003), RanGAP (Kusano et al. 2001), Smd3
Figure 2.—Phenotypes observed upon overexpression of genes in the salivary glands using fkhGal4. (A) Schematic of the setupof the screen. The phenotypes observed in the screen could be classified according to the categories depicted in this figure. Broadcategories are invagination defects (B and C), gland shape and lumen defects (D–I), positioning defects (K–P), and gland subfatedefects (Q). (B–Q) GFP-marker expression in green and phalloidin staining to reveal actin in red. Lateral or dorsal views andembryonic stages of embryos are indicated in each panel. The line in C indicates the area of the too-wide opening of the invag-inating gland shown; the arrow in D points to where proximal and distal cells of the gland touch due to excessive bending; thedouble arrows in H indicate the too-wide width of the gland; the arrows in M point to two buds emerging from the side of thegland; the arrows in N point to the two ends of a fork; the arrows in O point to cells of the glands that appear to touch acrossthe midline. B–G and K–Q are wide-field fluorescence images; H, I, and Q are confocal sections.
Salivary Gland Morphogenesis in Drosophila 547
TA
BL
E1
Gen
esid
enti
fied
inth
ega
in-o
f-fu
nct
ion
scre
en
EP
no
.C
yto
logy
Gen
eaf
fect
edE
ffec
t,lo
cati
on
inge
ne,
dir
ecti
on
Pen
etra
nce
of
ph
eno
typ
esa
Ph
eno
typ
ein
sali
vary
glan
ds
(fkh
Gal
4)
Fu
nct
ion
or
mu
tan
tp
hen
oty
pe,
kno
wn
fun
ctio
nin
flie
so
rgl
and
s
Cyt
osk
elet
on
and
cyto
skel
eto
nas
soci
ated
EP
(2)5
70
2L(2
6B4)
CG
13
99
3(a
ctin
/tu
bu
lin
pro
tein
fold
ing)
Ove
rexp
ress
ion
,59
-en
do
fge
ne
Stro
ng
Var
iab
leN
om
uta
nt
avai
lab
le
EP
(2)7
13
2L(2
6B1)
chic
kade
e(p
rofi
lin
)O
vere
xpre
ssio
n,
in59
regi
on
of
gen
eu
pst
ream
of
mo
stC
DS
Stro
ng
Var
iab
leL
eth
al,
Tec
29
chic
do
ub
lem
uta
nts
sho
wsa
liva
rygl
and
ph
eno
typ
e(C
ha
nd
ra
sek
ar
an
and
Bec
ken
do
rf
2005
)E
P(2
)10
11
2L(2
6B1)
chic
kade
e(p
rofi
lin
)A
nti
sen
seto
591
kbo
fge
ne
(or
ove
rexp
ress
ion
of
eIF
4a,
1.8
kbd
ow
nst
ream
)
Stro
ng
Ho
oks
,sh
eph
erd
’scr
oo
k
Let
hal
,T
ec2
9ch
icd
ou
ble
mu
tan
tssh
ow
sali
vary
glan
dp
hen
oty
pe
(Ch
an
dr
asek
ar
an
and
Bec
ken
do
rf
2005
)E
P(2
)93
82R
(59F
7)eg
alit
aria
n(d
ynac
tin
asso
ciat
ed)
Ove
rexp
ress
ion
,d
irec
tly
59o
fge
ne
Stro
ng
Var
iab
leF
emal
est
eril
e,le
thal
,m
icro
tub
ule
-bas
edtr
ansp
ort
EP
(2)2
04
72R
(57E
5)sy
nde
can
Inse
rted
in59
-en
do
fsd
cw
ron
gst
ran
d,
cou
ldd
rive
anti
sen
seto
6o
fth
e90
-kb
sdc
locu
so
ro
vere
xpre
sssa
ra7
kbd
ow
nst
ream
Stro
ng
Var
iab
leL
eth
al,
wo
rks
inco
nju
nct
ion
wit
hSl
it(S
teig
em
an
n
etal
.20
04)
EP
(3)3
56
73R
(88D
5)bi
tesi
ze(s
ynap
tota
gmin
-li
kep
rote
in)
Mid
dle
of
gen
e,co
uld
dri
vean
tise
nse
tom
ost
iso
form
sW
eak
Var
iab
leL
eth
al,
acti
no
rgan
izat
ion
atad
her
ens
jun
ctio
ns
(Pilo
tet
al.
2006
)
Sig
nal
ing
EP
(2)5
78
2L(2
4E1)
traf
-4(T
NF
-rec
epto
r-as
soci
ated
fact
or
4,p
revi
ou
sly
call
edT
raf-
1in
flie
s)
Ove
rexp
ress
ion
,m
idd
leo
fge
ne,
up
stre
amo
fm
ost
CD
SV
ery
stro
ng
Few
erce
lls
ingl
and
s,o
vere
xpre
ssio
nre
po
rted
lyin
du
ces
apo
pto
sis
(Ku
ra
na
ga
etal
.20
02)
Lar
val
leth
al(K
ur
an
ag
a
etal
.20
02)
EP
(2)2
16
72L
(29A
1)bt
k29
/tec
29
(Btk
fam
ily
kin
ase)
39-e
nd
of
gen
ew
ron
gst
ran
d,
cou
ldd
rive
anti
sen
seto
.30
of
40kb
btk2
9A
locu
s
Wea
kV
aria
ble
Let
hal
;T
ec29
isim
po
rtan
tfo
rsa
liva
rygl
and
inva
gin
atio
n(C
ha
nd
ra
sek
ar
an
and
Bec
ken
do
rf
2005
)E
P(2
)11
73
2L(3
7E1)
ran
GA
PO
vere
xpre
ssio
n,
mid
dle
of
gen
eb
ut
up
stre
amo
fC
DS
Stro
ng
Ho
ok,
cro
ok
Via
ble
;p
oss
ible
lin
kb
etw
een
nu
clea
rtr
ansp
ort
,ac
tin
,an
dp
rofi
lin
(Min
ak
hin
a
etal
.20
05)
EP
(2)2
15
82L
(37D
2)do
ugh
nu
ton
2(R
YKfa
mil
yre
cep
tor
tyro
sin
eki
nas
e)O
vere
xpre
ssio
nW
eak
Var
iab
leIm
po
rtan
tfo
rsa
liva
rygl
and
po
siti
on
ing
asis
drl
ano
ther
RYK
(Ha
rr
is
and
Bec
ken
do
rf
2007
) (con
tin
ued
)
548 V. Maybeck and K. Roper
TA
BL
E1
(Co
nti
nu
ed)
EP
no
.C
yto
logy
Gen
eaf
fect
edE
ffec
t,lo
cati
on
inge
ne,
dir
ecti
on
Pen
etra
nce
of
ph
eno
typ
esa
Ph
eno
typ
ein
sali
vary
glan
ds
(fkh
Gal
4)
Fu
nct
ion
or
mu
tan
tp
hen
oty
pe,
kno
wn
fun
ctio
nin
flie
so
rgl
and
s
EP
(3)3
54
23L
(61C
1)pt
pmeg
(tyr
osi
ne
ph
osp
hat
ase)
orm
thl9
(G-p
rote
in-c
ou
ple
dre
cep
tor)
Ove
rexp
ress
ion
of
ptpm
ego
ran
tise
nse
of
mth
l9(i
ntr
on
icto
ptpm
eg)
Stro
ng
Bu
dd
ing
Via
ble
;P
tpm
egis
FE
RM
do
mai
np
rote
in
EP
(3)3
70
43L
(62A
2)rh
ombo
id1
(EG
Fsi
gnal
ing,
intr
amem
bra
ne
pro
teas
e)O
vere
xpre
ssio
nV
ery
stro
ng
Ab
erra
nt
du
ctm
orp
ho
gen
esis
,p
ote
nti
ally
du
eto
ove
rpro
life
rati
on
(see
Fig
ure
s4
and
5)
Mu
tati
on
sin
rho
or
spit
zle
adto
tran
sfo
rmat
ion
of
du
ctce
lls
into
secr
eto
ryce
lls
(Ku
oet
al.
1996
)E
P(2
)22
01
2L(3
7F2)
spit
z(s
ecre
ted
EG
Fli
gan
d)
Inse
rted
into
mid
dle
of
spi;
cou
ldd
rive
anti
sen
seto
4kb
of
som
esp
im
RN
As
or
ove
rexp
ress
msb
1l
5kb
do
wn
stre
am
Wea
kV
aria
ble
Mu
tati
on
sin
rho
or
spit
zle
adto
tran
sfo
rmat
ion
of
du
ctce
lls
into
secr
eto
ryce
lls
(Ku
oet
al.
1996
)E
P(3
)54
93L
(62E
7)m
issh
apen
(Ste
20ki
nas
e)O
vere
xpre
ssio
n,
dir
ectl
y59
of
gen
eW
eak
To
o-w
ide
and
lum
py
lum
enL
eth
al,
lin
ked
ton
ucl
ear
mo
vem
ent
via
Bic
D(H
ou
alla
etal
.20
05)
and
cell
-sh
ape
chan
ges
du
rin
gm
orp
ho
gen
esis
(Ko
ppen
etal
.20
06)
Nu
cleu
s,tr
ansc
rip
tio
n(f
acto
rs)
EP
(2)2
17
62R
(48E
4)Sm
d3(s
nR
NP,
spli
cin
g)O
vere
xpre
ssio
n,
59o
fge
ne
Stro
ng
Ho
oks
Let
hal
(Sc
hen
kel
etal
.20
02)
EP
(2)4
74
2L(2
1B5)
kism
et(h
elic
ase)
Ove
rexp
ress
ion
,in
59re
gio
no
fge
ne
up
stre
amo
fal
lC
DS
Wea
kT
oo
-wid
elu
men
Let
hal
,se
gmen
tsp
ecifi
cati
on
(Da
ub
resse
etal
.19
99)
EP
(2)9
93
2R(5
0E1)
com
bgap
(zin
c-fi
nge
rp
rote
in)
Inse
rted
into
59-e
nd
of
CG
30
09
6w
ron
gst
ran
d,
anti
sen
seto
cg2
kbaw
ay
Ver
yst
ron
gV
aria
ble
Let
hal
;h
edge
ho
gsi
gnal
ing
inle
gp
atte
rnin
g(S
ven
dsen
etal
.20
00)
EP
(3)4
86
3L(7
5E1)
ftz-
f1(f
tz-t
ran
scri
pti
on
fact
or1
)In
sert
edin
tolo
cus,
sho
uld
ove
rexp
ress
lon
ger
iso
form
Wea
kV
aria
ble
Let
hal
(Flo
ren
ce
etal
.19
97)
EP
(3)7
11
3L(6
4E8)
bre-
1(n
ucl
ear
fact
or
do
wn
stre
amo
fN
otc
h)
Ove
rexp
ress
ion
,d
irec
tly
59o
fge
ne
Wea
kV
aria
ble
Let
hal
(Br
ay
etal
.20
05)
Pro
tein
syn
thes
isan
dd
egra
dat
ion
EP
(2)4
63
2R(4
7F7)
Tap
d(t
ran
slo
con
-as
soci
ated
pro
tein
d)
Ove
rexp
ress
ion
,59
-en
do
fge
ne
Wea
kV
aria
ble
Let
hal
,d
ow
nst
ream
of
dC
reb
-Ain
the
glan
ds
(Ab
ra
ms
and
An
dr
ew
2005
)E
P(2
)20
63
2L(3
7B7)
ned
d8(r
egu
lati
on
of
pro
teo
lysi
s)O
vere
xpre
ssio
n,
59-e
nd
of
gen
eW
eak
Bu
tter
fly
Let
hal
,u
biq
uit
inli
ke,
coo
per
ates
wit
hcu
llin
3(Z
hu
etal
.20
05)
EP
(2)1
18
72L
(33C
1)C
G5
31
7(r
ibo
som
alsu
bu
nit
)O
vere
xpre
ssio
n,
inse
rted
in59
-en
do
fJh
I-21
,w
ron
gst
ran
d,
sho
uld
dri
veC
G53
1740
0b
pd
ow
nst
ream
Stro
ng
Shep
her
d’s
cro
ok
ND
(con
tin
ued
)
Salivary Gland Morphogenesis in Drosophila 549
TA
BL
E1
(Co
nti
nu
ed)
EP
no
.C
yto
logy
Gen
eaf
fect
edE
ffec
t,lo
cati
on
inge
ne,
dir
ecti
on
Pen
etra
nce
of
ph
eno
typ
esa
Ph
eno
typ
ein
sali
vary
glan
ds
(fkh
Gal
4)
Fu
nct
ion
or
mu
tan
tp
hen
oty
pe,
kno
wn
fun
ctio
nin
flie
so
rgl
and
s
Mem
bra
ne
traf
fic
EP
(2)2
02
82R
(48F
8)ga
rz(a
rf-G
EF,
GB
F1)
Ove
rexp
ress
ion
,59
-en
do
fge
ne
Wea
kSe
vere
ho
oks
ER
-to
-Go
lgi
traf
fick
ing
inm
amm
als
(Szu
let
al.
2007
)E
P(2
)23
13
2L(3
5F1)
syn
taxi
n5
(SN
AR
Ep
rote
in)
Ove
rexp
ress
ion
,d
irec
tly
59o
fge
ne
Wea
kD
egen
erat
ing
glan
ds
Let
hal
,m
emb
ran
efu
sio
n,
cyto
kin
esis
(Xu
etal
.20
02)
Cel
lsu
rfac
ean
dex
trac
ellu
lar
EP
(2)8
27
2R(5
8D4)
CG
36
24
(Ig
do
mai
np
rote
in)
Ove
rexp
ress
ion
,d
irec
tly
59o
fge
ne
Stro
ng
Var
iab
leN
DE
P(2
)93
72R
(52D
1)sl
it(a
xon
guid
ance
rece
pto
r)M
idd
leo
fge
ne,
wro
ng
stra
nd
,an
tise
nse
to20
of
50kb
gen
e?St
ron
gH
oo
ksL
eth
al;
slit
has
bee
nsh
ow
nto
be
invo
lved
insa
liva
rygl
and
po
siti
on
ing
(Ko
lesn
ik
ov
and
Bec
ken
do
rf
2005
)E
P(2
)21
20
2L(2
2A3)
CG
14
35
1(L
RR
and
Igd
om
ain
tran
smem
bra
ne
pro
tein
)O
vere
xpre
ssio
n,
dir
ectl
y59
of
gen
eW
eak
Var
iab
leN
D;
BL
AST
sho
ws
sim
ilar
ity
toSl
itE
P(2
)24
63
2L(3
5D4)
glio
tact
in(t
ran
smem
bra
ne
pro
tein
of
sep
tate
jun
ctio
ns)
Ove
rexp
ress
ion
,d
irec
tly
59o
fge
ne
Wea
kV
aria
ble
Let
hal
;im
po
rtan
tfo
rtu
be
size
con
tro
lin
trac
hae
(Pa
ul
etal
.20
03)
EP
(3)5
52
3L(7
0A3)
capr
icio
us
(tra
nsm
emb
ran
eL
RR
pro
tein
)O
vere
xpre
ssio
n,
dir
ectl
y59
of
gen
eV
ery
stro
ng
Biz
arre
lyb
ran
chin
gan
db
ud
din
glu
men
Let
hal
(Sh
ish
id
oet
al.
1998
)
En
zym
esE
P(2
)21
99
2R(5
1B1)
tou
tve
lu(g
luco
sam
inyl
-tr
ansf
eras
e)In
sert
edin
intr
on
of
bo
thtt
van
dla
mC
(wh
ich
isin
tro
nic
tott
v),
cou
ldo
vere
xpre
ss�
10o
f60
kbtt
v(�
50%
of
CD
S)o
r1.
1kb
anti
sen
seto
Lam
C
Wea
kV
aria
ble
Let
hal
;m
uta
nts
dis
rup
th
h,
wn
t,an
dd
pp
sign
alin
g(B
or
nem
an
net
al.
2004
)
EP
(2)1
15
72R
(59B
6)C
G9
84
9(p
ote
nti
alp
rote
ase
of
the
sub
tila
sefa
mil
y)In
sert
edin
to59
-en
do
fC
G38
00,
wro
ng
stra
nd
,o
vere
xpre
ssio
no
fC
G98
4960
0b
paw
ay
Wea
kV
aria
ble
ND
EP
(3)3
63
93L
(65A
10)
CG
10
16
3(p
ho
sph
oli
pas
eA
1)In
sert
ed59
of
Bes
t2,
wro
ng
stra
nd
,co
uld
dri
vean
tise
nse
toC
G1
01
63
800
bp
away
Wea
kT
oo
larg
ean
dir
regu
lar
lum
en
ND
Mit
osi
s,m
eio
sis,
and
germ
lin
eE
P(2
)81
22L
(35C
1)va
sao
rvi
g(v
asa
intr
on
icge
ne)
In39
regi
on
of
vig,
wh
ich
isin
tro
nic
tova
sa,
wo
uld
ove
rexp
ress
391
kbo
fvi
go
ran
tise
nse
tova
sa
Wea
kL
um
py
lum
enva
sa:
leth
al,
germ
cell
det
erm
inat
ion
;vi
g:N
D
EP
(3)3
41
3R(8
2D2)
tacc
(cen
tro
som
alp
rote
in)
Mid
dle
of
gen
e,co
uld
dri
vean
tise
nse
toal
lla
rge
iso
form
so
fta
ccW
eak
Var
iab
leL
eth
al(B
ar
ro
set
al.
2005
)
(con
tin
ued
)
550 V. Maybeck and K. Roper
TA
BL
E1
(Co
nti
nu
ed)
EP
no
.C
yto
logy
Gen
eaf
fect
edE
ffec
t,lo
cati
on
inge
ne,
dir
ecti
on
Pen
etra
nce
of
ph
eno
typ
esa
Ph
eno
typ
ein
sali
vary
glan
ds
(fkh
Gal
4)
Fu
nct
ion
or
mu
tan
tp
hen
oty
pe,
kno
wn
fun
ctio
nin
flie
so
rgl
and
s
Oth
erE
P(2
)23
56
2R(5
7A6)
mir
-31
0/-
31
3cl
ust
erO
vere
xpre
ssio
n,
inse
rted
200
bp
59o
fcl
ust
erV
ery
stro
ng
To
o-w
ide
irre
gula
rlu
men
Dis
rup
tio
no
fm
ir-3
10
clu
ster
affe
cts
do
rsal
clo
sure
(Lea
ma
net
al.
2005
)E
P(2
)25
86
2R(5
7A6)
mir
-31
0/-
31
3cl
ust
erO
vere
xpre
ssio
n,
inse
rted
100
bp
59o
fcl
ust
erSt
ron
gIr
regu
lar
lum
enD
isru
pti
on
of
mir
-31
0cl
ust
eraf
fect
sd
ors
alcl
osu
re(L
ea
ma
net
al.
2005
)E
P(2
)25
87
2R(5
7A6)
mir
-31
0/-
31
3cl
ust
erO
vere
xpre
ssio
n,
inse
rted
100
bp
59o
fcl
ust
erW
eak
Irre
gula
rlu
men
Dis
rup
tio
no
fm
ir-3
10
clu
ster
affe
cts
do
rsal
clo
sure
(Lea
ma
net
al.
2005
)E
P(2
)12
21
2L(2
7F4)
mir
-27
5,
mir
-30
5O
vere
xpre
ssio
n,
2kb
up
stre
amo
fge
nes
Stro
ng
Shep
her
d’s
cro
ok
ND
EP
(2)2
08
32R
(45F
1)C
G1
88
8.
6kb
away
,o
vere
xpre
ssio
nW
eak
Var
iab
leN
DE
P(2
)11
63
2L(3
3E4)
vir-
1(v
iru
s-in
du
ced
RN
A1)
Ove
rexp
ress
ion
,d
irec
tly
59o
fge
ne
Wea
kB
ud
din
gN
DE
P(2
)12
39
2L(2
5F5)
CG
14
00
5o
rC
G7
23
9In
sert
edin
59-e
nd
of
CG
9171
wro
ng
stra
nd
,an
tise
nse
toC
G14
005
300
bp
do
wn
stre
amo
ro
vere
xpre
ssio
no
fC
G72
392
kbd
ow
nst
ream
Wea
kV
aria
ble
ND
;b
oth
CG
1400
5an
dC
G72
39co
nse
rved
on
lyam
on
gD
roso
ph
ilid
ae
EP
(2)2
21
92L
(33E
4)C
G6
40
5O
vere
xpre
ssio
n,
inse
rted
1.5
kbu
pst
ream
of
CG
64
05
Wea
kV
aria
ble
ND
;tw
om
amm
alia
no
rth
olo
gs
EP
(2)2
19
02R
(55E
1)C
G3
03
32
Ove
rexp
ress
ion
,1
kb59
of
CG
3033
2St
ron
gH
oo
ksN
DE
P(2
)21
82
2R(5
4A2)
CR
30
23
4(c
yto
soli
ctR
NA
gen
e)O
vere
xpre
ssio
nV
ery
stro
ng
Var
iab
leN
DE
P(3
)31
33R
(98E
5)C
G1
52
3(r
elat
edto
WD
40re
pea
t-co
nta
inin
gp
rote
in32
)O
vere
xpre
ssio
n,
inse
rted
dir
ectl
y59
of
gen
eSt
ron
gV
aria
ble
ND
EP
(2)2
26
92R
(53D
11)
CG
34
46
0A
nti
sen
seto
CG
3446
0?,
EP
is�
2.5
kbaw
ayW
eak
Var
iab
leN
D
EP
(2)3
83
2L(2
3C4)
No
thin
gd
ow
nst
ream
for
.10
kb;
nex
tC
Gis
CG
1726
514
kbaw
ay
Inse
rted
39o
fC
G35
58W
eak
Sho
rt,
exp
and
edat
turn
ND
EP
(2)2
17
32L
(35B
2)n
oth
ing
do
wn
stre
amfo
r.
10kb
Inse
rted
into
59-e
nd
of
no
ocel
li,
wro
ng
stra
nd
Wea
kV
aria
ble
ND
EP
(2)2
14
6a
Un
clea
rU
ncl
ear
Gen
om
ep
osi
tio
no
fE
Pu
ncl
ear
Stro
ng
Sear
chin
gce
lls
ND
EP
(2)2
26
5U
ncl
ear
Un
clea
rG
eno
me
po
siti
on
of
EP
un
clea
rV
ery
stro
ng
Sho
rt,
stra
igh
t,ex
pan
ded
ND
EP
(2)9
85
Un
clea
rU
ncl
ear
Gen
om
ep
osi
tio
no
fE
Pu
ncl
ear
Stro
ng
Bu
dd
ing,
bra
nch
ing
ND
All
the
gen
esid
enti
fied
inth
esc
reen
,so
rted
acco
rdin
gto
thei
rp
rop
ose
dfu
nct
ion
,ar
eli
sted
.aP
enet
ran
ceo
fp
hen
oty
pes
:w
eak,
20–3
0%o
fem
bry
os
sho
win
gp
hen
oty
pe;
stro
ng,
30–5
0%o
fem
bry
os
sho
win
gp
hen
oty
pe;
very
stro
ng,
.50
%o
fem
bry
os
sho
win
gp
he-
no
typ
e(w
ith
thre
eca
ses
,70
%an
dfo
ur
case
s.
90%
).
Salivary Gland Morphogenesis in Drosophila 551
(Schenkel et al. 2002), Nedd 8 (Zhu et al. 2005), andTout-velou (The et al. 1999). Fourteen of the 51 hitswere EPs inserted in previously uncharacterized genes,many with close orthologs in other species, includingvertebrates.
Analysis of individual genes in detail: In the follow-ing section we will discuss a subset of the genes identifiedin the screen. Three of these were genes endogenouslyexpressed in the glands (chickadee, rhomboid1, and egl),and thus overexpression could have interfered withtheir proper function in the glands. One gene (btsz) wasexpressed in the glands and was identified through aninsertion that would have led to production of a tissue-specific antisense RNA, thus potentially mimicking aloss-of-function situation. The last gene (caps) was notendogenously expressed in the glands and thus theoverexpression identified a potential requirement else-where for proper cell and tissue shape.
Genes endogenously expressed in glands identifiedthrough overexpression: Chickadee encodes the Dro-sophila Profilin protein. Profilins are actin–monomer-sequestering proteins, which have been implicated inpromoting both actin polymerization and depolymer-ization (Yarmola and Bubb 2006). Drosophila Profilinfulfills essential functions at all stages of developmentand also in the female germline (Verheyenand Cooley
1994). Loss of profilin has been associated with theinability to constrict apical surfaces in the morphoge-netic furrow of the larval eye disc (Benlali et al. 2000).With respect to salivary gland morphogenesis, it hasbeen reported that tec29 chic double mutants showdisorganized actin in the salivary gland placode anddisplay a delay in invagination (measured by theremaining placode area at stage 14). This study alsoreports that chic mutant embryos have normal glands(Chandrasekaran and Beckendorf 2005). Two EPinsertions into chic showed phenotypes in the glandswhen driven by fkhGal4, EP(2)713, and EP(2)1011.EP(2)713 should overexpress the entire chickadee cod-ing sequence (see supplemental Figure 1), whereasEP(2)1011 is inserted in the opposite direction andcould drive expression of an antisense RNA to the 59
most 1 kb of the chic pre-mRNA (or it could driveexpression of eIF4a, situated 1.8 kb away; see scheme inFigure 3A). Expression of either EP led to aberrantlyshaped glands (Figure 3, B–E), with EP(2)1011 givingfrequent hook-shaped and shepherd’s crook-shapedglands (Figure 3, D and E). Overexpression of a UAS-chickadee construct using fkhGal4 led to embryos showinga disorganized epidermis in the regions where fkhGal4was expressed, with a loss of apical Crumbs accumula-tion in the epithelial cells of the epidermis (Figure 3,F–G9). The glands nonetheless invaginated and, withinthe invaginated portion of the glands, Crumbs waslocalized apically. This suggests that epithelial integrityand/or polarity might be impaired if levels of Profilinare imbalanced. Effects on junctional Armadillo in the
absence of Profillin have been described (Townsley
and Bienz 2000). In chic mutant embryos (either chic221
or chic01320) at a stage when the first cells had justinvaginated from the salivary gland placode, cell shapeswithin the placode often appeared irregular comparedto wild type, although Crumbs was still localized apicallyat this stage (compare H, H9, and I in Figure 3). At laterstages the salivary glands invaginated but were irregularin shape, and the placodal and surrounding epidermalcells on the surface of the embryo appeared disruptedwith absent or mislocalized Crumbs labeling (Figure 3,K and L and O and P at stage 12 and M and N at stage 14;for comparison, a matching wild-type epidermal scan atstage 14 is shown in Q–Q$). Other apical markers suchas the spectraplakin Shot and DE-Cadherin also ap-peared disrupted at these stages (Figure 3, V–V$ forDE-Cadherin and data not shown). Nonetheless, withinthe invaginated portion of the glands Crumbs waslocalized apically, probably because early apical Crumbslocalization in the placodal cells was unperturbed. Thus,in contrast to an earlier report (Chandrasekaran andBeckendorf 2005), either elevation or disruption ofChickadee/Profilin levels appeared to affect salivarygland invagination to some extent. As Profilin has beenshown to promote both actin polymerization and depoly-merization, depending on the context and tissue (Yarmola
and Bubb 2006), an imbalance of Profilin levels (eitherdecreased or increased) could affect the critical corticalfunction of actin during cell-shape changes required toallow the invagination and/or the cell rearrangementson the surface of the embryo during invagination.
Rhomboid1 and EGF-receptor signaling are known toinfluence cell-fate decisions within the salivary glandprimordium. The EGF-ligand Spitz is secreted from theventral midline cells with Rhomboid being the intra-membrane protease essential for its release (Shilo
2005). Spitz diffuses a few cell diameters laterally toinduce the most ventral cells within the salivary glandplacode to become duct cells, whereas the other placodalcells become secretory cells (see scheme in Figure 4A).This switch in fate transmitted by EGF is in part achievedthrough repression of the fkh transcription factor. Fkhin turn represses three duct-specific genes: trachealess(trh), dead ringer (dri), and serrate (ser) (Kuo et al. 1996;Haberman et al. 2003). Thus, by the end of embryogen-esis, rhomboid1 and other spitz group mutants have salivaryglands that are composed entirely of secretory cells andare completely enclosed within the embryo without anyductal connection to the outside (Figure 5, A–B9; Kuo
et al. 1996).Overexpression of rhomboid1 using EP(3)3704 in the
glands led to the formation of glands that werepositioned too anteriorly with no obvious distinctionin shape between duct and secretory cells and nocommon duct-like structure formed at stage 15 (Figure4, C–C$). An identical phenotype was observed when arhomboid1 transgene under the control of the UAS
552 V. Maybeck and K. Roper
Figure 3.—Chickadee (Profilin) is important for salivary gland invagination. chickadee encodes the Drosophila Profilin protein.(A) Scheme of the chic locus indicating the position and orientation of the two EP lines that showed phenotypes when driven inthe salivary glands. B and C and D and E show phenotypes observed in the screen for EP713 and EP1011 respectively; wide-fieldfluorescent images of live embryos are shown. (F–G9) Overexpression of chickadee using a UAS-chickadee construct led to invagi-nation problems and aberrantly shaped glands. (F) An internal confocal stack of a gland (14 mm thick), labeled with Crumbs toreveal the apical surface/lumen of the glands (red) and showing the SrcGFP marker in green. (G and G9) A surface stack of thesame embryo (with Crumbs in red and SrcGFP in green in G and Crumbs as a single channel in G9; 3 mm thick). The arrowG9points to disrupted epidermis in the region of the placode from which the glands have started to invaginate. Note the absenceof Crumbs from the apical surface of cells in this region. (H and H9) A low and high magnification view, respectively, of the slightlydisorganized epidermis of a chic mutant embryo labeled for crumbs, with H9 showing the salivary gland placode and gland in aprojection (18-mm thick stack). (I) A wild-type placode and gland (17-mm thick stack) at the same stage as in H9. Note that thehighly organized arrangement of apical constriction of the placodal cells is less apparent in the chic mutant (bracket in H9 and inI). K and L and O and P show the aberrant glands and disrupted epidermis in two different chic alleles (chic01320 and chic221) at stage12; labeling for Crumbs is in green (and also shown as a single channel in K9 and O9) and for phalloidin in red. Note the dis-ruption and absence of apical Crumbs labeling in the region of the salivary gland placode (arrows in K9, L, O9, and P point to theseareas). (K) A projection of a 35-mm thick stack. (L) A 5-mm-thick surface projection. (O) A projection of a 26-mm-thick stack. (P) A3-mm-thick surface projection. (M and N) The disrupted epidermis in the region from which salivary gland cells invaginated in astage 14 embryo. Crumbs is green in M and a single channel in M9, and phalloidin is red in M and a single channel in N. M is aprojection of a 34-mm-thick stack, and N is a 5-mm-thick surface stack. For comparison, a stage 14 wild-type embryo is shown in Q–Q$. Crumbs is green in Q and a single channel in Q9, and phalloidin is red in Q and a single channel in Q$; Q is a 5-mm-thicksurface stack. The arrows in M9 and N point to the disrupted region, the white lines in M and Q indicate the ventral midline (theview in M–N is slightly oblique). (R–V$) chic221 mutant embryos at stages 12–14. R and R9 show lateral views of a placode, whereas Sshows an internal stack of the gland. T and T9 are ventral views of the two placodes, with U showing an internal stack of the glands.(V–V$) A surface stack of a mutant embryo (5 mm thick). Crumbs is green in R, T, and V and a single channel in R9, T9, and V9;phalloidin is red. Both S and U show Crumbs labeling to outline the lumen of the gland. DE-Cadherin (DE-Cad) labeling is in redin V and as a single channel in V$.
Salivary Gland Morphogenesis in Drosophila 553
Figure 4.—rhomboid1 overex-pression disrupts salivary glandmorphogenesis, but is not suffi-cient to induce salivary duct fate.(A) Scheme of the rho locus indicat-ing the position of the EP identi-fied in the screen. (B) Schemedepicting the known involvementof EGF signaling in salivary glandmorphogenesis. EGF is releasedfrom themidline (red line) and in-duces, in the cells close to the mid-line (light green), the repressionof fkh, which in turn leads to sup-pression of secretory fate in thesecells, inducing them to adopt ductcell fate. Fkh expression remainshigh in the remaining salivarygland primordium (dark green),thus inducing these cells to formthe secretory part of the gland(Kuo et al. 1996). (C–C$) Overex-pression of rhomboid1 in the sali-vary glands using EP(3)3704 ledto glands that, at stage 15 of em-bryogenesis, were located too faranterior with secretory cells thatappeared cuboidal instead of co-lumnar, no proper duct connect-ing the secretory portions to theoutside, and an aberrantly shapedlumen. The SrcGFP marker isgreen in C and a single channelin C9; phalloidin is red in C and asingle channel in C$. (D–D$)The same phenotype as in C is ob-served when a UAS-rhomboid1 con-struct is expressed in the glandsusing fkhGal4. The GFP-EFGas2marker is green in D and a singlechannel in D9; Crumbs is red inD and a single channel in D$.(E–F9) Already at stage 13 the in-vaginated portion of the glandshows aberrant morphology (ec-topic lumen indicated by the ar-row in E9), and the amount ofcells remaining at the surface ap-pears too large (bracket in E9and F9). The SrcGFP marker isgreen in E and F (and a singlechannel in E9 and F9); crumbs isin red. (G–P$) Analysis of dCreb-Aand Eyegone expression: markersof secretory and duct fate, respec-tively. (G–I$) Control glands ex-
pressing only the GFP-EFGas2 marker labeled with antibodies against dCreb-A and Eyg at stage 11 (G–G$), stage 14 (H–H$),and stage 15 (I–I$). (K–M$) Glands expressing UAS-rhomboid1 in the salivary glands using fkhGal4 labeled for dCreb-A and Eyg atstage 11 (K–K$), stage 14 (L–L$), and stage 15 (M–M$). Note that, despite the irregular shape and ectopic cells (bracket in L),dCreb-A is strongly expressed in the early invaginated part of the glands (L9 and M9). Eyg is expressed in the most anterior cellsof the invaginated glands (L$ and M$), as in the control, and also in the ectopic cell bulge on the surface of the embryo (L$; bracketin L denotes the bulge, and the dashed line indicates the ventral midline). (N–P$) Glands expressing UAS-rhomboid1 using armGal4.(N–N$)dCreb-AandEygexpression intheplacodeat stage11. (OandO9)Ventral viewof theremaining placode(O)andinvaginatedglands (O9) at stage 13. More ectopic cells expressing Eyg are found on the ventral surface (bracket in O). Small stubby glands haveinvaginated and expressed dCreb-A (arrows in O9). (P–P$) Lateral view of glands at stage 14. More cells have invaginated and ex-pressed dCreb-A (arrow in P), and the most ventral cells on the surface still express Eyg (P$). (G–P) GFP makers, green; dCreb-A,red; and Eyg, blue. (G9–N9 and P9) dCreb-A as a single channel. (G$–N$ and P$) Eyg as a single channel. All panels are projections ofconfocal stacks that cover the entire thickness either of the invaginated glands or of the placode at earlier stages.
554 V. Maybeck and K. Roper
Figure 5.—EGFR signaling is necessary but not sufficient to induce salivary duct fate, and overactivation leads to ectopic celldivisions. Analysis of components of the EGFR signaling pathway in the glands. It has been reported previously that salivary glandsin rhomboid/spitz-group mutant embryos do not specify any ductal portion of the glands (Kuo et al. 1996). (A–B9) This phenotypewas confirmed in a rhoPD5 embryos, a null allele of rho (Freeman et al. 1992). (A and A9) At stage 13, most of the secretory cells of thegland have invaginated, leaving two large holes visible at the surface of the embryo (arrows in A). (A) The surface of the embryo.(A9) An internal confocal stack to reveal the shape and location of the invaginated glands. (B and B9) At stage 14, the glands havefully invaginated and detached from the surface of the embryo, leaving no ductal connection to the outside and a large hole onthe surface. (B) The surface of the embryo. (B9) An internal confocal stack to show the blunt-ended gland. The arrow in B9 pointsto the blunt end, and the arrow in B points to the hole. (A–B9) Labeling for Crumbs. (C) An amended gland-fate specificationscheme as introduced in Figure 4 to illustrate the altered signaling in rho mutant embryos, where absence of EGFR signalinginduces the entire salivary gland primordium to adopt secretory fate. (D–F) Overexpression of argos, an extracellular inhibitorof EGFR signaling (Schweitzer et al. 1995), using fkhGal4 caused a phenotype similar to that seen in rho mutants: (D–D$) At stage13, the invaginated secretory portion of the glands detaches from the surface of the embryo (arrow in D9), and no duct is formed.(E and E9) The salivary gland primordium at stage 11 appears normal. (F) Scheme showing that downregulation of EGFR sig-naling leads to conversion of presumptive duct cells into secretory cells. (G–I) Overexpression of an activated form of the EGFreceptor (CA-EGFR) using fkhGal4 leads to glands with highly disorganized and aberrant lumen from stage 12 onward. (G–G$) Aconfocal stack of stage 13 embryo; the arrow in G points to the lumen marked by Crumbs. (H and H9) Cell shapes marked byCrumbs in the salivary gland primordium at stage 11 appear normal. (I) Scheme showing that elevated EGFR signaling through-out the primordium does not induce duct fate in all cells. (K–I) Overexpression of a secreted and active form of the ligand Spitzusing fkhGal4 leads to glands that very much resemble those seen upon overexpression of rho (compare with Figure 4). (K–K$) Aconfocal stack of a stage 15 embryo. No ductal structures are formed, and the shape of the secretory cells and the lumen is ab-errant. (L and L9) The salivary gland primordium at stage 11 appears disrupted with irregular and too large apexes of the in-
Salivary Gland Morphogenesis in Drosophila 555
promoter was expressed in the glands (Figure 4, D–D$).At earlier stages during the invagination process, whenrhomboid1 was overexpressed in the gland primordiumusing fkhGal4 and either the EP(3)3704 or UAS-rhomboid1,aberrantly shaped lumena could be observed (Figure 4,E and E9), but most prominently a large bulge of fkh-expressing ectopic cells seemed to arise between thealready invaginated secretory gland portions at theposition where the individual ducts usually would form(Figure 4, F and F9; similar ectopic cells could also beobserved in other experimental situations; see below).Although the analysis of rhomboid1 mutant embryos hasshown that EGFR signaling is necessary to induce ductfate in the most ventral cells of the placode (and thusloss of EGFR signaling leads to loss of ducts; see Figure 5,A–B9), this could indicate that activation of EGFR sig-naling throughout the placode was not sufficient to in-duce duct fate in all cells. To test this hypothesis, welabeled embryos overexpressing rhomboid1 with markersfor duct [Eye gone (Eyg); Jones et al. 1998] and se-cretory (dCreb-A; Abrams and Andrew 2005) glandfate and compared the expression to wild-type embryos.Both markers labeled the salivary placode (Figure 4, Gand K) and the glands at stage 14 (Figure 4, H and L)and stage 15 (Figure 4, I and M) in a comparable patternto wild-type placodes and glands. In addition, the bulgeof potentially ectopic cells at the ventral surface that wasobserved at stages 13 and 14 (bracket in Figure 4L)strongly expressed the duct marker Eyg, indicating thatcells fated to become duct cells have overproliferated. Atstages 15 and 16, a large group of cells at the very anteriortip of the embryo that completely failed to invaginateexpressed Eyg (Figure 4, M and M$). These data suggestthat, at least when the EGFR pathway is ectopicallyactivated throughout the whole placode in a time framethat mimicked the expression of fkh, this was not enoughto convert secretory cells into duct cells. To test whetherthis failure in fate conversion could be due to the timingof the overexpression, we also expressed rhomboid1under the control of armadilloGal4 (armGal4) through-out the whole epidermis of the embryo and withexpression starting at much earlier stages (Figure 4,N–P$). The overactivation of EGFR signaling through-out the epidermis led to embryos with varying degrees
of overall affected morphology (in many cases head invo-lution failed, and general appearance of the epidermiswas less organized compared to wild type, althoughepithelial integrity/polarity appeared unperturbed asjudged by UAS-a-cateninGFP labeling that was alsodriven under the control of armGal4). In these embryos,dCreb-A and Eyg were expressed in the placode at stage11, although the dCReb-A expression domain appearedto extend beyond the placode area into the more an-terior hemi-segments (Figure 4, N and N9). Salivarygland invagination was strongly affected in that only veryshort glands invaginated into the embryo (see Figure 4,O and P, for stages 13 and 14, respectively). Nonetheless,these stubby glands expressed dCreb-A, the secretoryfate marker, in the invaginated portion of the glands(Figure 4, O and P, arrows), and expressed Eyg in a fewcells that had invaginated but were still close to thesurface of the embryo (Figure 4P$). In addition, a largebulge of Eyg-expressing cells could be found at thesurface of the embryo between the two invagination sides(Figure 4O, bracket), similar to what we observed whenrhomboid1 was overexpressed under the control of fkhGal4.
We also tested this hypothesis further by overexpress-ing additional components of the EGF pathway in thesalivary glands: a constitutively active version of theEGFR, UAS-CA-EGFR; a secreted version of the ligandSpitz, UAS-sspi; and the negative regulator Argos, UAS-argos. Overexpression of argos using fkhGal4 led to ahigh proportion of glands that lost a ductal connectionto the embryo surface, similar to the rhomboid1 mutantembryos (Figure 5, A–D$), although less penetrant(which is probably due to timing and/or expressionlevels of the transgene). When a secreted version of theEGFR ligand Spitz was expressed using fkhGal4, thephenotypes observed appeared very similar to the onesseen in the rhomboid1 overexpressing embryos, namelyectopic cells and glands positioned too anteriorlywithout a discernible duct (Figure 5, K–L9). Overex-pression of a constitutively active form of the EGFR,UAS-CA-EGFR, in the salivary glands led to invaginationof cells with slightly aberrant shapes and to an in-vagination hole that was too large. This led to fullyinvaginated glands with a too large and aberrantlyshaped lumen, although the individual and common
vaginating cells. The arrow in L points to a group of cells that show midbodies left by mitotic divisions marked by the GFP-EFGas2microtubule marker. Ectopic mitoses can also be observed when CA-EGFR is expressed using fkhGal4. (M) Scheme showing thatelevated EGFR signaling through overexpression of secreted Spitz throughout the primordium does not induce duct fate in allcells. The SrcGFP or GFP-EFGas2 markers are green in D, E, G, H, K, and L and shown as a single channel in D9, G9, and K9.Crumbs labeling is red in D, E, G, H, K, and L and shown as a single channel in D$, E9, G$, H9, K$, L9. (M–O$9) Analysis ofectopic cell divisions induced by activation of EGF signaling using phospho-histone H3 (p-HisH3) as a marker of mitosis. Inthe marker-expressing control, p-His3 labeling is restricted to the area outside the placode (M) and invaginated gland (P). WhenUAS-secreted spitz (N and Q) or UAS-rhomboid1 (O and R) are expressed in the glands, many mitotic cells can be found in the placodeand invaginated gland. (M–R and O9) GFP markers, green; p-HisH3, red. (M–Q and R) Crumbs, blue. (O and O9) DAPI, blue. O9–O$$ show a higher magnification of the dividing cells in O. GFP-EFGas2 is shown as a single channel in O$, p-HisH3 as a singlechannel in O$9, and DAPI as a single channel in O$$. The dashed line in O9–O$$ highlights a cell in anaphase (note the spindle inO$) and the arrow in O9–O$$ points to a midbody in telophase, similar to the ones indicated in L.
556 V. Maybeck and K. Roper
ducts appeared normal (Figure 5, G–H9). The ectopicventral cells observed upon expression of rhomboid1 orsecreted spitz in the glands could be a result of over-proliferation if EGFR signaling in the placodal cellsworks not only as a fate switch but also as a mitogenicsignal. We thus analyzed the amount of cell division inthe placode at stages 12 and 13 in embryos overexpress-ing rhomboid1 or secreted spitz under the control offkhGal4 compared to wild-type placodes using an anti-body against phosphorylated histone H3 (p-HisH3), achromatin mark of mitotic cells (Wei et al. 1999). Incontrol embryos, salivary placode cell nuclei at stages12 and 13 did not contain nuclei showing p-HisH3labeling (Figure 5, M and P), whereas many placodecells overexpressing secreted spitz (Figure 5, N and Q) orrhomboid1 (Figure 5, O and R) showed the p-HisH3 markand were thus actively dividing. Also, labeling of micro-tubules with GFP-EFGas2 revealed mitotic figures (Fig-ure 5, O9 and O$, dotted lines) and remnant spindlemidbodies (Figure 5, O9 and O$, arrows).
The results presented above suggest the following:first, that EGFR signaling, although necessary for ductfate, is not sufficient to induce duct fate in all salivaryplacode cells, even though absence of EGFR signalingturns all cells into secretory cells. Second, an increase ofEGFR signaling in the placode cells can induce exces-sive proliferation in a part of these cells, probablyresulting in the mishapen glands observed upon rhom-boid1 or spitz overexpression. Third, through modula-tion of fkh levels and downstream components withinthe placode, EGFR signaling might also impinge on thecell-shape changes that invaginating cells undergo.
Egalitarian (Egl) and BicaudalD (BicD) are twoproteins that act together with cytoplasmic Dynein inthe localization of mRNAs in Drosophila embryos andthe oocyte, with Egl interacting directly with Dyneinlight chain (Bullock and Ish-Horowicz 2001; Navarro
et al. 2004). Overexpression of egl using EP(2)938 led tosalivary glands that were C-shaped or shortened (Figure6, B and C). Shortened glands could also be observedwhen egl was expressed in the glands using a UAS-eglconstruct (Figure 6, D and D9). GFP-positive cells thatappeared to lose contact with the gland (arrow in Figure6D) could be observed. Because both BicD and Egl haveessential functions during oogenesis, an analysis of egl orBicD null embryos is not possible. We therefore analyzedembryos from mothers carrying two hypomorphicalleles of egl that were mated to heterozygous fathers(see materials and methods for the exact genotypes).Embryos with reduced Egl function often showed adisrupted epidermis, with large patches that appearedto completely lack apical Crumbs labeling (compareFigure 6, F and F9, to Figure 6, G and G9). Thisphenotype was variable, however, and an example ofan embryo with a less disrupted epidermis is shown inFigure 6H. Also, during later stages of invagination atstage 13, the placode area was often disrupted and
lacked apical Crumbs (Figure 6K). Salivary gland morpho-genesis was disrupted in egl mutant embryos in thatthe cells of the placode often did not change their apicesin a coordinated way (although crumbs still accumu-lated apically in the placode; see Figure 6H9), the invagi-nation hole appeared too large and extended (Figure6H), and the invaginated portion of the glands oftenhad an irregular shape (Figure 6, I and K9). In theinvaginated portion of a gland, Crumbs was not concen-trated near the apical cell junctions as in the wild type,where this accumulation appears as a honeycomb lattice(compare Figure 6L9 to Figure 6M9). Instead, Crumbswas delocalized all over the apical surface and largeaccumulations could also be found intracellularly (ar-row in Figure 6L9).
What could be the mechanism leading to a loss ofapical surface identity or constituents? Egl and BicDtogether with Dynein act as minus-end-directed micro-tubule motors, and as in most epithelial cells, the minusends of microtubules are located near the apical surfacein the salivary glands (Myat and Andrew 2002). hairymRNA is one of the best-understood cargoes of Egl- andBicD-mediated transport (Bullock et al. 2003, 2006),and Hairy has been shown to be important for theregulation of apical membrane growth during salivarygland formation, in part through modulation of Crumbs(Myat and Andrew 2002). Thus, affecting hairy tran-script localization through lowered levels of Egl andBicD could in turn affect the maintenance of apicalmembrane identity in the secretory cells. Alternatively,recent reports have shown that crumbs mRNA itself,and also the RNA of the Crumbs-associated proteinStardust (Std), are apically localized and this apical local-ization is important for function (Horne-Badovinac
and Bilder 2008; Li et al. 2008). Thus, if crumbs mRNAlocalization were dependent upon Egl and BicD, thenreduction of Egl and BicD would result in a loss offunctional Crumbs at the apical surface, leading to a lossof epithelial characteristics. The apical localization atleast of std mRNA appears developmentally regulatedin the embryo (Horne-Badovinac and Bilder 2008).Thus, it is possible to envision that salivary gland cellapical maintenance is Egl and BicD dependent andespecially sensitive to levels of Egl and thus to Crumbs incomparison to other epithelial tissues at the same stage.We are currently investigating this issue in more detail.
A gene endogenously expressed in glands identifiedthrough antisense expression: Bitesize (Btsz) is the soleDrosophila synaptotagmin-like protein. Its mRNA is ex-pressed in the salivary glands and also in other epithelialtissues, with a strong apical enrichment (Serano andRubin 2003). Btsz has recently been shown to controlthe organization of actin at adherens junctions in earlyembryos, although it might be dispensable in adult flyepithelia (Pilot et al. 2006). Recruitment of Btsz in earlyembryos to the apical junctional region is not depen-dent on E-Cadherin but on phosphatidylinositol-(4,5)-
Salivary Gland Morphogenesis in Drosophila 557
bisphosphate (PIP2) and Par-3/Bazooka, a proteinof the Par-3/Par-6/aPKC apical complex (Pilot et al.2006). Two mutant btsz alleles have been described:btszK13-4 that deletes a portion of the N terminus of someBtsz protein isoforms and btszJ5-2 that introduces a stopcodon due to a frameshift in the N-terminal portion ofbtsz (Serano and Rubin 2003). Expression driven byfkhGal4 from the EP element identified in our screen,
EP(3)3567, should lead to production of an antisenseRNA in most of the btsz coding sequence and thus coulddownregulate endogenous btsz mRNA levels (see schemein Figure 7A and supplemental Figure 2). In embryoswhere EP(3)3567 is driven by fkhGal4 at stage 13, whenmost secretory cells have invaginated from the placode,the epidermis in the region where the antisense RNAwas expressed was disrupted and had lost apical Crumbs
Figure 6.—egalitarian overexpression reveals a potential role for egalitarian and BicD in salivary gland morphogenesis. (A)Scheme of the egalitarian (egl) locus indicating the gene structure and the position of the EP identified in the screen. (B andC) The two phenotypes observed in the screen upon overexpression of egl using EP(2)938: bent (B) and shortened (C) glands.The GFP-EFGas2 marker is in green, and phalloidin labeling is in red; both images are wide-field fluorescent. (D and D) Over-expression of a UAS-egl construct using fkhGal4 frequently led to short glands at stage 14, with some GFP-positive cells losing con-tact with the glands (arrow in D). (E and E9) A comparable wild-type embryo. Crumbs is in red in D and E and as a single channelin D9 and E9, SrcGFP is in green in D, and Shot is in green in E. (F–G9) Dorsolateral views of stage 14 embryos. (F and F9) egl mutantembryos often show a disrupted epidermis with mislocalized Crumbs labeling (arrows in F9 point to areas where Crumbs is com-pletely absent), whereas in wild-type embryos Crumbs is localized apically and circumferentially in all epithelial cells (G and G9).(H and H9) egl mutant embryo showing a disorganized salivary gland placode, with a too-large and extended invagination hole.(I and I9) A stage 13 egl mutant embryo with a gland that appears too wide and short, showing mislocalized Crumbs labeling. (Kand K9) Ventral view of a stage 13 egl mutant embryo. (K) A surface confocal stack, showing two disrupted areas in the epidermiswhere the glands have invaginated (arrows). (K9) An internal confocal stack of the same embryo, with a too-wide and aberrantgland (the red dashed line traces the outline of the gland). (L and L$) Higher magnification of the gland shown in K9 in a smallerconfocal stack. Note the mislocalized Crumbs protein at the lateral sides of cells and internally (arrow in L9) that cannot be seen inwild-type glands (compare to M9). (H–L$) Crumbs labeling is red in H, I, K, and L and a single channel in H9, I9, K9, and L9;phalloidin labeling is green in H, I, K, and L and a single channel in L$. (M and M9) Magnification of a section through a stage13 wild-type gland; SrcGFP is green in M, Crumbs is red in M and a single channel in M9.
558 V. Maybeck and K. Roper
Figure 7.—bitesize overex-pression reveals a potential rolefor bitesize in salivary gland mor-phogenesis. (A) Scheme of the bi-tesize (btsz) locus indicating thegene structure and the positionand orientation of the EP identi-fied in the screen. (B–C9) Potentialknockdown of Bitesize throughoverexpression of antisense RNAusing EP(3)3567 leads to epithelialdefects in the overexpressing cells(arrow in B9) and glands that invag-inatewithaberrantmorphology.(Band B9) A surface confocal stack ofa stage 13 embryo. (C and C9) Thecorresponding internal stack thatreveals the glands. Note the ab-sence of Crumbs labeling in thearea that shows GFP-EFGas2marker expression (arrow in B9).GFP-EFGas2 is green in B and C,Crumbs is red in B and C and a sin-gle channel in B9 and C9. (D–E9)Ventral views of btszK13-4 mutant vs.wild-type embryos at stage 13. Thearrow in D points to the disruptedepidermis in the mutant embryo.Note thedisorganizationof thepla-code area compared to wild type(indicated by the brackets in Dand E) and the failure to accumu-late Shot apically (arrow in E$ indi-cates the accumulation in the wildtype). Crumbs is red in D and Eand a single channel in D9 and E9.Shot is green in D and E and a sin-gle channel in D$ and E$. (F–I$)Examples of lateral views of btszK13-4
mutant vs. wild-type embryos atstage 14. (F–F$) Highly disruptedand disorganized epidermis in thebtszK13-4 mutant (arrows in F pointto areas lacking apical circumfer-ential Crumbs labeling; compareto the wild-type epidermis in I–I$). (G–G$) An internal stack ofthe same embryo as in F (the corre-sponding internal stack for thewild-type embryo in I is shown inK–K$).Note that thesalivaryglandof the btszK13-4 mutant embryo is los-ing apical Crumbs accumulation(G9) compared to the wild type(K9); the phalloidin labeling inG$ still shows cell outlines, butthese also lack apical actin accu-mulation as seen in the wild type(K$). Crumbs labeling is greenin F, G, I, and K and a single chan-nel in F9, G9, I9, and K9, andphalloi-din is red in F, G, I, and K and asingle channel in F$, G$, I$, and
K$. (H and L) Confocal stacks of the embryos in F and I at the level where the salivary duct reaches the epidermis labeled for Crumbs.(H) The btszK13-4 mutant. (L) The wild type. Note that the duct in H has lost apical Crumbs accumulation (the arrow points to theremnants of Crumbs labeling in the duct) and that the epidermis at the point from which the glands invaginated is disruptedand lacks apical Crumbs (indicated by the bar in H). (M–N$) The btsz J5-2 mutant at stage 14 also shows disrupted epidermis and lossof Crumbs (M$) and also DE-Cadherin (M9) in the area where the placode was previously located. Crumbs labeling in the invaginated
Salivary Gland Morphogenesis in Drosophila 559
accumulation (Figure 7, B and B9). The glands them-selves showed an irregular lumen (Figure 7, C and C9).btszK13-4 mutant embryos that manage to cellularize andcomplete gastrulation showed a somewhat disruptedepidermis, with loss of apical Crumbs accumulation inpatches at later stages (see Figure 7D for a stage 13 andFigure 7F for a stage 14 embryo). Nonetheless, manysalivary gland placode cells still showed enhancedCrumbs labeling, although the shapes of the apicalcircumferences of invaginating cells were irregular(compare Figure 7, D and D9, to wild type in Figure 7,E and E9). Also, the apical accumulation of the flyspectraplakin protein Shot could not be observed inbtszK13-4 embryos in contrast to wild type (compare Figure7D$ to Figure 7E$). At stage 14, the secretory portion ofthe glands in btszK13-4 embryos was often found to loseapical localization of Crumbs (and also to show reducedapical actin enrichment; Figure 7, G and G$). Similar towhat we observed upon expression of EP(3)3567 withfkhGal4, the epidermis in the region where the placodalcells were located previously was disrupted and lostapical Crumbs and also DE-Cadherin labeling com-pletely (compare Figure 7H to wild type in Figure 7L;Figure 7M). The disruption of the epidermis and thefailure of the proper apical localization of Crumbs is tosome extent reminiscent of the phenotypes observed inegl mutant embryos. As btsz mRNA is another RNA that islocalized apically in various epithelial cells (and thelocalization signal has been identified; Serano andRubin 2003), one could speculate that its localizationcould also be dependent on Egl and BicD, thus explain-ing some overlap in the phenotypes.
A gene not endogenously expressed in the glandsidentified through overexpression: Capricious (Caps)belongs to the class of leucine-reach-repeat (LRR)transmembrane proteins, together with its close pa-ralogue Tartan (Trn). Both proteins have been impli-cated in the formation of compartments of cells withdifferent affinities in the wing disc (Milan et al. 2001),modulation of epithelial integrity within the wing disc(Mao et al. 2008), correct targeting of a subset ofphotoreceptor axons to the correct layer within theoptic lobe (Shinza-Kameda et al. 2006), and the joiningof tracheal branches over segment boundaries (Krause
et al. 2006). One study also showed that in tissue cultureCapricious and Tartan are able to mediate homophiliccell adhesion, a molecular function that could explaintheir roles discussed above (Shinza-Kameda et al. 2006).Overexpression of capricious using EP(3)552 led toglands with an enlarged lumen and very aberrant shapesat stage 15 (Figure 8, B and C). The same phenotype was
observed when a transgene of capricious was expressedunder UAS control (Figure 8, D–E9). During early stagesof invagination, the invagination hole appeared en-larged compared to wild type and extended along theanterior–posterior axis (compare Figure 8, F–G9, toFigure 8, H and H9), suggesting problems in the shapeof invagination cells and the order of invagination. Thisdisorganization at later stages could lead to the aberrantshapes of the secretory part of the glands observed. Wethen analyzed capricious and tartan single and capriciousand tartan double-mutant embryos (Mao et al. 2008), assome previous studies have indicated redundancy be-tween both molecules in some tissues (Mao et al. 2008).In all mutant situations, invaginating glands oftenshowed irregular lumens (Figure 8, I–M), indicatingthat both proteins might work together during salivarygland invagination. We next analyzed if and wherecapricious and tartan are expressed during salivary glandmorphogenesis using P-element insertions in each genethat carry a lacZ gene leading to b-galactosidase expres-sion under the endogenous expression control of eachof the genes. Capricious was expressed in the embryo inthe region of the salivary glands from stages 12–15, butappeared to be mostly excluded from the salivary glandsthemselves, although it was always expressed in cells inclose contact with the glands (Figure 8, N–S). Incontrast, tartan was expressed strongly in the salivaryglands from placode stage onward (Figure 8, T and U).Similar expression patterns could also be observed in insitu hybridization for capricious and tartan mRNAs (seesupplemental Figures 3 and 4). Thus, in analogy to thesituation in the developing trachae where both proteinsare expressed in complementary tissues to allow forproper dorsal branch fusion (Krause et al. 2006), thereciprocal expression of capricious and tartan in andaround the salivary glands could play a part in thecorrect invagination and later positioning of the glandswith respect to the surrounding tissues.
DISCUSSION
Here we show that a gain-of-function screen lookingfor factors affecting a process of epithelial morphogen-esis, that is, the formation of the salivary glands in theDrosophila embryos, was efficient in identifying a rangeof known and new players. The screen was designed toidentify genes that are endogenously expressed in theglands and where overexpression or antisense expres-sion by an EP element interfered with endogenousfunction. The screen could also identify genes notendogenously expressed in the glands but with an
gland is aberrant (N$) whereas DE-Cadherin still appears to be apical (N9). (M–M$) A projection of a 5-mm-thick confocal surfacestack. (N–N$) The projection of a 20-mm-thick internal stack covering the whole gland. DE-Cadherin is green in M and N and a singlechannel in M9 and N9, and Crumbs is red in M and N and a single channel in M$ and N$.
560 V. Maybeck and K. Roper
Figure 8.—capricious overexpression reveals a potential role for capricious and tartan in salivary gland morphogenesis. (A) Scheme ofthe capricious (caps) locus indicating the gene structure and the position of the EP identified in the screen. (B and C) Live images of theGFP-EFGas2 marker of the caps overexpression phenotype using EP(3)552 observed in stage 15 embryos in the screen. (D–E9) Confocalstacks of two examples of aberrantly shaped lumen of salivary glands at stage 15 upon overexpression of a UAS-caps construct usingfkhGal4. The lumen is highlighted by Crumbs labeling in D9 and E9 and very much resembles the defects observed in the screen.(F–H9) Invaginating glands at stage 12. (F and F9) A surface stack of the primordium upon UAS-caps overexpression. Note that the holeat the invagination point is too extended and not positioned completely within the primordium (as highlighted by the GFP marker)comparedtothewild-typeprimordiuminHandH9. (GandG9)Acompletestackof theglandsat stage12uponUAS-capsoverexpression.Note that the size of the invagination hole (marked by the red dotted lines in G9) is again too large and irregular compared to wild type,and the invaginated portion of the glands shows a too-wide and irregular lumen. The GFP-EFGas2 marker is green and Crumbs labelingis red in D, E, F, G, and H and Crumbs is shown as a single channel in D9, E9, F9, and G9. (I–M) capsPB1 single, trn28.4 single, and capsDel1 trn28.4
doublemutants(allarenullmutations;Mao etal.2008)oftenshowdefects insalivaryglandmorphology, i.e., irregular lumenatdifferentstages of invagination.Staining for theflyspectraplakinShot isgreen and Crumbs is red inI,K,L,and M, and Crumbs is shownasa singlechannel in I9 and K9. (N–S) A lacZ-containing P-element insertion into the caps locus reveals that caps is not expressed inmostcells of thesalivary glands. The outline of the glands is marked by a white dashed line, b-gal labeling is in green and Crumbs in red. N and O showb-galactosidase (b-gal) labeling at stage 12 and stage 15, respectively. (P–S) Cross sections of a gland at stage 14. Note that the glands aresurrounded by cells expressing caps. (T–U) A lacZ-containing P-element insertion in the trn locus reveals that trn is expressed in salivarygland cells at all stages. (Tand T9) Most cells of the salivary gland placode at stage 11 express trn at varying levels (border of the placode ismarked by dashed lines). (U) At stage 14, trn is still expressed strongly in all salivary gland cells, including the duct. The outline of thegland is indicated by a dashed line. b-gal labeling is in green and Crumbs in red in T and U; b-gal is shown as a single channol in T9.
Salivary Gland Morphogenesis in Drosophila 561
apparent overexpression phenotype that uncovered apotential function for this gene in cell shape or cyto-skeletal regulation elsewhere. Genes falling into all ofthese classes have been identified in the screen. Weshow that factors with a variety of proposed functionscan affect salivary gland invagination—from cytoskele-tal components via signaling factors (some of whichimpinge on the cytoskeleton themselves) to microRNAsand novel uncharacterized proteins. Most of the factorsidentified in this screen should affect salivary glandmorphogenesis in a gland-autonomous fashion due tothe restriction of overexpression to the salivary glandsonly, the only exception being the overexpression ofsecreted signaling factors.
We could observe a variety of phenotypes from ab-errantly shaped glands to irregular lumena and wronglypositioned glands. In half of the cases, upon over-expression of a gene (or, in a few cases, potentially anantisense RNA) several different phenotypes could beobserved, as opposed to a single dominant phenotypethat was found in the other half. At the level of detail atwhich we analyzed the phenotypes (GFP markers of cellshape or microtubules plus phalloidin labeling of actin),these phenotypes seemed to fall into a limited numberof classes, suggesting that several different perturbancesof the system might lead to similar phenotypes. We alsofound no case that consistently led to a complete failurein salivary gland invagination. This is not completelysurprising, as dominant effects or knockdowns that wewould expect to see in our screen might not perturb thesystem enough to lead to a complete failure in all aspectsof the cell-shape changes required for the invagination.Also, when analyzing various mutants in the initialfollow-up of a subset of the hits identified, even insituations where the embryonic epidermis seemed to bevery disrupted and cell shapes of invaginating cells werehighly irregular, as, for example, in btsz or egl mutants,the glands nonetheless managed to invaginate. Theseobservations suggest that there is a strong drive for theinvagination of the cells of the salivary gland primor-dium, with many different factors contributing at theeffector level. Elimination or perturbance of only one ofthese factors will not prevent invagination completely,but will rather lead to a slightly disordered invaginationprocess that in the end might result in the aberrantshapes and phenotypes that we observed. This situationappears similar to the process of mesoderm invaginationduring gastrulation in the Drosophila embryo. Manyfactors contribute to this process, but loss of none—apartfrom the most upstream transcription factor initiatingthe whole mesoderm invagination program (twist)—willabolish invagination completely. In all other downstreammutants analyzed, the mesoderm will nonetheless man-age to invaginate, albeit in an uncoordinated and/ordelayed fashion (Leptin 2005). Similarly, during salivarygland invagiantion, fkh appears to be the most upstreamtranscription factor initiating the invagination program
for both the secretory and the ductal part of the glands.In the absence of Fkh, invagination fails completely(Myat and Andrew 2000a). Several direct targets ofFkh have been identified, including the transcriptionfactors senseless (Chandrasekaran and Beckendorf
2003) and sage (Abrams et al. 2006); PH4aSG2, a prolyl-4-hydroxylase (Abrams et al. 2006); and crebA, which inturns control the expression of secretory genes in theglands (Abrams and Andrew 2005).
Only one recent study has so far addressed the di-rect targets of fkh in a genomewide manner (Liu andLehmann 2008). In this study, whole-genome expres-sion levels in control pupae and pupae with forcedexpression of fkh were compared . At the beginning ofpupariation, fkh controls both the expression of thesalivary gland secretion proteins (Roth et al. 1999) andthe cell death that occurs during pupariation in thistissue (Myat and Andrew 2000a). Downstream targetsof fkh identified in this study included cell death genesand genes involved in autophagy, in phospholipidmetabolism, in glucose and fatty acid metabolism, inhormone-dependent signaling pathways, and others.These other factors regulated by fkh included severalproteins that we also identified in our screen: capricious,bitesize, gliotactin, ptpmeg, rhomboid1, and spitz. This over-lap of fkh-dependent factors and genes identified by us,which are also potentially fkh dependent at differentstages of development, i.e., at embryo and pupa stages,could suggest that these overlapping factors are regu-lated by fkh independently of stage-specific cofactors.
The set of genes identified in our screen encodesproteins with a wide range of potential functions: cyto-skeleton or cytoskeleton associated, signaling, nuclearor transcription factor, protein synthesis and degrada-tion, membrane traffic, cell surface and extracellular,enzymes, mitosis/meiosis germline, and uncharacter-ized genes and microRNAs. Nonetheless, many of thesehave been implicated in some aspect of epithelial mor-phogenetic function, some even within the salivaryglands themselves. Thus, we are confident that manyof the factors identified in this gain-of-function screenwill turn out to have a function in epithelial morpho-genesis. The initial folllow-up of the subset of genesdescribed in detail above also confirms that gain-of-function phenotypes can point to new factors involvedin a process and can also reveal new aspects of a functionof a protein that were not previously appreciated, as inthe case of rhomboid1.
The overexpression phenotype of rhomboid1 in thesalivary glands showed an intriguing phenotype: ectopicEGFR signaling throughout the part of the primordiumthat will constitute the secretory part of the gland doesnot simply induce these cells to switch to ductal fate.Instead, it suggests either that other permissive factorsthat are absent from the secretory primordium areexpressed in the duct primordium independently ofEGFR signaling or that further inhibitory factors that
562 V. Maybeck and K. Roper
prevent ductal fate are at work in the secretory primor-dium in addition to EGF-induced fkh. The timing of theoverexpression of rhomboid1 in the screen, using partof the fkh promoter in the fkhGal4 line to drive theexpression of the EP elements, did not seem to becrucial to the fate decision, as expression of rhomboid1throughout the whole epidermis using an earlier Gal4-driver (armGal4) still led to invagination of glands withidentifiable secretory and ductal cells (Figure 4, N–P).Thus it does not appear that the cells in the placode areresponsive to the EGFR signal in terms of fate assignmentin only a narrow window of time, but supports the notionthat other factors are involved. It will be interesting todetermine in the future what these factors are.
The analysis of caps and tartan mutant phenotypessuggests a role for these genes in salivary gland mor-phogenesis. The molecular function of both Caps andTrn proteins is still unclear. They appear important inmediating interaction and distinctiveness betweengroups of cells: neurons finding appropriate targets inthe brain (Shishido et al. 1998; Shinza-Kameda et al.2006), separation of ventral and dorsal compartmentcells in the wing disc (Milan et al. 2001), and trachealmorphogenesis across segment boundaries (Krause
et al. 2006). It has been suggested that Caps and Trnact as homophilic or heterophilic adhesion receptors orserve another unidentified function during adhesion.Our results indicate that salivary gland morphogenesismight be a useful system for addressing their molecularfunction in more detail. It is also interesting to note thatnot only Caps and Trn, but also Slit and the proteinencoded by CG14351, which were also both identifiedas hits in the screen, belong to the family of LRR pro-teins, suggesting a general role for this class of surfacereceptors in salivary gland morphogenesis.
Another gene with a highly penetrant phenotype(severely shortened glands) identified in the screen isTNF-receptor-associated factor-4 (traf-4, previously anno-tated as traf-1 in FlyBase). Traf-4 has previously beenshown to induce apoptosis via activation of JNK-kinasewhen overexpressed in other tissues (Kuranaga et al.2002). Traf-4 has also been linked to the Ste20 kinaseMishapen (another gene identified in our screen),which in turn has been shown to be important forcoordinated cell-shape changes occurring, for example,during dorsal closure in the fly embryo and duringepiboly in the zebrafish embryo (Koppen et al. 2006). Italso appears to have a role in mesoderm invagination(M. Leptin, personal communication). Overexpressionof traf-4 using EP(2)578 led to a severe reduction in thenumber of cells in the salivary glands at embryonic stage15 [when counting cell numbers in fluorescence imagestaken through the middle of wild-type glands, these had39.9 6 3.9 cells (n ¼ 40) around the perimeter of thegland, whereas traf-4 overexpressing glands had 18.7 6
5.0 cells (n ¼ 45) around the perimeter]. We have notdirectly tested whether the missing cells died through
induction of apoptosis, but would expect this to be thecase in agreement with earlier studies. To addresswhether traf-4 has a function linked to mishapen andcell-shape changes in the glands, we analyzed fly em-bryos lacking Traf-4 (this mutant was a kind gift of MariaLeptin) for any problems in the early cell-shape changesoccurring during salivary gland invagination, but couldnot find any strong defects (data not shown). Thus,despite many similarities between the epithelial mor-phogenetic processes of mesoderm invagination (whichalso starts with the invagination of an epithelial sheet)and salivary gland invagination, downstream effectorsvary between the two systems.
Another interesting group of hits identified in thescreen are the EP elements potentially driving overex-pression of the mir-310 microRNA cluster. This clustercontains the microRNA genes mir-310, mir-311, mir-312,and mir-313. Overexpression of the cluster from threedifferent EP elements located just upstream of thecluster, EP(2)2536, EP(2)2586, and EP(2)2587, led ineach case to glands with widened and irregular lumena,although with varying penetrance (data not shown).Antisense-mediated depletion of each microRNA fromthis cluster has previously been shown to perturb dorsalclosure and head involution in the embryo, indicatingthat the inhibition of downstream targets of this clustermight be important for various epithelial morphoge-netic events (Leaman et al. 2005). Thus, the microRNAsin this cluster could be important in regulating targetsthat require downregulation to facilitate invaginationduring salivary gland morphogenesis. Two other micro-RNAs have been shown to be expressed in the salivarygland in the embryo—mir-8 and mir-375—with mir-8showing a dynamic expression pattern (Aboobaker
et al. 2005); expression patterns for the mir-310 clusterhave not been analyzed yet. These data together withour screen results suggest that microRNA-dependentcontrol of gene expression might be an important factorin salivary gland morphogenesis.
In summary, the gain-of-function screen for factorsaffecting cell shape during salivary gland morphogenesisin the Drosophila embryo presented here was successfulin identifying a range of candidates. These candidatesrepresent, on the one hand, genes that are endoge-nously expressed in the glands and thus are likely toserve a role during salivary gland morphogenesis/tubulogenesis. On the other hand, we identified genesthat are not endogenously expressed in the glands butnonetheless interfered with their invagination, poten-tially through effects on cell shape or the cytoskeleton.These genes might therefore also be important for theregulation of cell shape in other tissues. It will beinteresting to determine which of the candidate genesof the first class serve a function only in the salivaryglands and which are required for tubulogenesis eventsin general and in other species. For the second group ofcandidates, an analysis of their role during cell-shape
Salivary Gland Morphogenesis in Drosophila 563
changes in other morphogenetic events will be key tounderstanding how they affected salivary gland mor-phogenesis in our screen. Because the screen identifiedseveral uncharacterized genes whose expression gavestrong phenotypes in the glands and that have closeorthologs in mammals, the salivary glands appear to be agood model system for analyzing the function of suchgenes. Finally, our follow-up investigation of a selectedset of candidate genes through loss-of-function mutantsdemonstrates that the combination of functional screen-ing and phenotypic loss-of-function analysis provides auseful approach to identifying downstream effectors in amorphogenetic process.
The authors thank Nick Brown, Matthew Freeman, Simon Bullock,Deborah Andrew, Maria Leptin, and the Bloomington and SzegedStock Centers for fly stocks; Matthew Freeman, Simon Bullock, SarahBray, Deborah Andrew, and the Developmental Studies HybridomaBank at the University of Iowa for antibodies; Maria Leptin forcommunication of results prior to publication; Nick Brown for useof his confocal microscope; and Sean Munro and Nick Brown forhelpful comments on the manuscript. This work was supported by theBiotechnology and Biological Sciences Research Council (grant no.BB/B501798/1) and the Royal Society.
LITERATURE CITED
Aboobaker, A. A., P. Tomancak, N. Patel, G. M. Rubin and E. C. Lai,2005 Drosophila microRNAs exhibit diverse spatial expressionpatterns during embryonic development. Proc. Natl. Acad. Sci.USA 102: 18017–18022.
Abrams, E. W., and D. J. Andrew, 2005 CrebA regulates secretoryactivity in the Drosophila salivary gland and epidermis. Develop-ment 132: 2743–2758.
Abrams, E. W., W. K. Mihoulides and D. J. Andrew, 2006 Forkhead and Sage maintain a uniform and patent salivary glandlumen through regulation of two downstream target genes,PH4alphaSG1 and PH4alphaSG2. Development 133: 3517–3527.
Andrew, D. J., A. Baig, P. Bhanot, S. M. Smolik and K. D. Henderson,1997 The Drosophila dCREB-A gene is required for dorsal/ventralpatterningof the larval cuticle.Development 124:181–193.
Barros, T. P., K. Kinoshita, A. A. Hyman and J. W. Raff, 2005 AuroraA activates D-TACC-Msps complexes exclusively at centrosomesto stabilize centrosomal microtubules. J. Cell Biol. 170: 1039–1046.
Bate, M., and A. Martinez Arias, 1993 The Development of Drosophilamelanogaster, Vol. II. Cold Spring Harbor Laboratory Press, ColdSpring Harbor, NY.
Bejarano, F., C. M. Luque, H. Herranz, G. Sorrosal, N. Rafel et al.,2008 A gain-of-function suppressor screen for genes involved indorsal-ventral boundary formation in the Drosophila wing. Ge-netics 178: 307–323.
Benlali, A., I. Draskovic, D. J. Hazelett and J. E. Treisman,2000 act up controls actin polymerization to alter cell shapeand restrict Hedgehog signaling in the Drosophila eye disc. Cell101: 271–281.
Bornemann, D. J., J. E. Duncan, W. Staatz, S. Selleck and R.Warrior, 2004 Abrogation of heparan sulfate synthesis in Dro-sophila disrupts the Wingless, Hedgehog and Decapentaplegicsignaling pathways. Development 131: 1927–1938.
Bradley, P. L., M. M. Myat, C. A. Comeaux and D. J. Andrew,2003 Posterior migration of the salivary gland requires an intactvisceral mesoderm and integrin function. Dev. Biol. 257: 249–262.
Bray, S., H. Musisi and M. Bienz, 2005 Bre1 is required for Notchsignaling and histone modification. Dev. Cell 8: 279–286.
Bullock, S. L., and D. Ish-Horowicz, 2001 Conserved signals andmachinery for RNA transport in Drosophila oogenesis and em-bryogenesis. Nature 414: 611–616.
Bullock, S. L., D. Zicha and D. Ish-Horowicz, 2003 The Drosoph-ila hairy RNA localization signal modulates the kinetics of cyto-plasmic mRNA transport. EMBO J. 22: 2484–2494.
Bullock, S. L., A. Nicol, S. P. Gross and D. Zicha, 2006 Guidanceof bidirectional motor complexes by mRNA cargoes throughcontrol of dynein number and activity. Curr. Biol. 16: 1447–1452.
Campos-Ortega, J. A., and V. Hartenstein, 1985 The Embryonic De-velopment of Drosophila melanogaster. Springer-Verlag, Berlin.
Cha, G. H., K. S. Cho, J. H. Lee, M. Kim, E. Kim et al., 2003 Discretefunctions of TRAF1 and TRAF2 in Drosophila melanogastermediated by c-Jun N-terminal kinase and NF-kappaB-dependentsignaling pathways. Mol. Cell. Biol. 23: 7982–7991.
Chandrasekaran, V., and S. K. Beckendorf, 2003 senseless is nec-essary for the survival of embryonic salivary glands in Drosophila.Development 130: 4719–4728.
Chandrasekaran, V., and S. K. Beckendorf, 2005 Tec29 controlsactin remodeling and endoreplication during invagination ofthe Drosophila embryonic salivary glands. Development 132:3515–3524.
Colas, J. F., and G. C. Schoenwolf, 2001 Towards a cellular andmolecular understanding of neurulation. Dev. Dyn. 221: 117–145.
Daubresse, G., R. Deuring, L. Moore, O. Papoulas, I. Zakrajsek
et al., 1999 The Drosophila kismet gene is related to chromatin-remodeling factors and is required for both segmentation andsegment identity. Development 126: 1175–1187.
Florence, B., A. Guichet, A. Ephrussi and A. Laughon, 1997 Ftz-F1is a cofactor in Ftz activation of the Drosophila engrailed gene.Development 124: 839–847.
Freeman, M., B. E. Kimmel and G. M. Rubin, 1992 Identifying tar-gets of the rough homeobox gene of Drosophila: evidence thatrhomboid functions in eye development. Development 116: 335–346.
Haberman, A. S., D. D. Isaac and D. J. Andrew, 2003 Specificationof cell fates within the salivary gland primordium. Dev. Biol. 258:443–453.
Harris, K. E., and S. K. Beckendorf, 2007 Different Wnt signals actthrough the Frizzled and RYK receptors during Drosophila sali-vary gland migration. Development 134: 2017–2025.
Harris, K. E., N. Schnittke and S. K. Beckendorf, 2007 Twoligands signal through the Drosophila PDGF/VEGF receptorto ensure proper salivary gland positioning. Mech. Dev. 124:441–448.
Henderson, K. D., and D. J. Andrew, 2000 Regulation and func-tion of Scr, exd, and hth in the Drosophila salivary gland. Dev.Biol. 217: 362–374.
Henderson, K. D., D. D. Isaac and D. J. Andrew, 1999 Cell fatespecification in the Drosophila salivary gland: the integrationof homeotic gene function with the DPP signaling cascade.Dev. Biol. 205: 10–21.
Hogan, B. L., and P. A. Kolodziej, 2002 Organogenesis: molecularmechanisms of tubulogenesis. Nat. Rev. Genet. 3: 513–523.
Horne-Badovinac, S., and D. Bilder, 2008 Dynein regulates epi-thelial polarity and the apical localization of stardust A mRNA.PLoS Genet. 4: e8.
Houalla, T., D. Hien Vuong, W. Ruan, B. Suter and Y. Rao,2005 The Ste20-like kinase misshapen functions together withBicaudal-D and dynein in driving nuclear migration in the devel-oping Drosophila eye. Mech. Dev. 122: 97–108.
Jones, N. A., Y. M. Kuo, Y. H. Sun and S. K. Beckendorf, 1998 TheDrosophila Pax gene eye gone is required for embryonic salivaryduct development. Development 125: 4163–4174.
Kaltschmidt, J. A., C. M. Davidson, N. H. Brown and A. H. Brand,2000 Rotation and asymmetry of the mitotic spindle directasymmetric cell division in the developing central nervous sys-tem. Nat. Cell Biol. 2: 7–12.
Kolesnikov, T., and S. K. Beckendorf, 2005 NETRIN and SLITguide salivary gland migration. Dev. Biol. 284: 102–111.
Koppen, M., B. G. Fernandez, L. Carvalho, A. Jacinto and C. P.Heisenberg, 2006 Coordinated cell-shape changes control ep-ithelial movement in zebrafish and Drosophila. Development133: 2671–2681.
Krause, C., C. Wolf, J. Hemphala, C. Samakovlis and R. Schuh,2006 Distinct functions of the leucine-rich repeat transmem-brane proteins capricious and tartan in the Drosophila trachealmorphogenesis. Dev. Biol. 296: 253–264.
564 V. Maybeck and K. Roper
Kuo, Y. M., N. Jones, B. Zhou, S. Panzer, V. Larson et al., 1996 Sali-vary duct determination in Drosophila: roles of the EGF receptorsignalling pathway and the transcription factors fork head andtrachealess. Development 122: 1909–1917.
Kuranaga, E., H. Kanuka, T. Igaki, K. Sawamoto, H. Ichijo et al.,2002 Reaper-mediated inhibition of DIAP1-induced DTRAF1degradation results in activation of JNK in Drosophila. Nat. CellBiol. 4: 705–710.
Kusano, A., C. Staber and B. Ganetzky, 2001 Nuclear mislocaliza-tion of enzymatically active RanGAP causes segregation distor-tion in Drosophila. Dev. Cell 1: 351–361.
Leaman, D., P. Y. Chen, J. Fak, A. Yalcin, M. Pearce et al.,2005 Antisense-mediated depletionreveals essential andspecificfunctions of microRNAs in Drosophila development. Cell 121:1097–1108.
Leptin, M., 2005 Gastrulation movements: the logic and the nutsand bolts. Dev. Cell 8: 305–320.
Li, Z., L. Wang, T. S. Hays and Y. Cai, 2008 Dynein-mediated apicallocalization of crumbs transcripts is required for Crumbs activityin epithelial polarity. J. Cell Biol. 180: 31–38.
Liu, Y., and M. Lehmann, 2008 Genes and biological processes con-trolled by the Drosophila FOXA orthologue Fork head. InsectMol. Biol. 17: 91–101.
Lubarsky, B., and M. A. Krasnow, 2003 Tube morphogenesis: mak-ing and shaping biological tubes. Cell 112: 19–28.
Mao, Y., M. Kerr and M. Freeman, 2008 Modulation of Drosophilaretinal epithelial integrity by the adhesion proteins capriciousand tartan. PLoS ONE 3: e1827.
Milan, M., U. Weihe, L. Perez and S. M. Cohen, 2001 The LRRproteins capricious and Tartan mediate cell interactions duringDVboundary formation in the Drosophilawing.Cell106:785–794.
Minakhina, S., R. Myers, M. Druzhinina and R. Steward,2005 Crosstalk between the actin cytoskeleton and Ran-medi-ated nuclear transport. BMC Cell Biol. 6: 32.
Molnar, C., A. Lopez-Varea, R. Hernandez and J. F. de Celis,2006 A gain-of-function screen identifying genes required forvein formation in the Drosophila melanogaster wing. Genetics174: 1635–1659.
Myat, M. M., and D. J. Andrew, 2000a Fork head prevents apopto-sis and promotes cell shape change during formation of the Dro-sophila salivary glands. Development 127: 4217–4226.
Myat, M. M., and D. J. Andrew, 2000b Organ shape in the Dro-sophila salivary gland is controlled by regulated, sequential inter-nalization of the primordia. Development 127: 679–691.
Myat, M. M., and D. J. Andrew, 2002 Epithelial tube morphology isdetermined by the polarized growth and delivery of apical mem-brane. Cell 111: 879–891.
Navarro, C., H. Puthalakath, J. M. Adams, A. Strasser and R.Lehmann, 2004 Egalitarian binds dynein light chain to establishoocyte polarity and maintain oocyte fate. Nat. Cell Biol. 6: 427–435.
Panzer, S., D. Weigel and S. K. Beckendorf, 1992 Organogenesisin Drosophila melanogaster: embryonic salivary gland determi-nation is controlled by homeotic and dorsoventral patterninggenes. Development 114: 49–57.
Paul, S. M., M. Ternet, P. M. Salvaterra and G. J. Beitel,2003 The Na1/K1 ATPase is required for septate junctionfunction and epithelial tube-size control in the Drosophila tra-cheal system. Development 130: 4963–4974.
Pilot, F., J. M. Philippe, C. Lemmers and T. Lecuit, 2006 Spatialcontrol of actin organization at adherens junctions by a synapto-tagmin-like protein Btsz. Nature 442: 580–584.
Pirraglia, C., R. Jattani and M. M. Myat, 2006 Rac function inepithelial tube morphogenesis. Dev. Biol. 290: 435–446.
Rørth, P., K. Szabo, A. Bailey, T. Laverty, J. Rehm et al.,1998 Systematic gain-of-function genetics in Drosophila. Devel-opment 125: 1049–1057.
Roth, G. E., S. Wattler, H. Bornschein, M. Lehmann and G.Korge, 1999 Structure and regulation of the salivary gland se-cretion protein gene Sgs-1 of Drosophila melanogaster. Genetics153: 753–762.
Schenkel, H., S. Hanke, C. De Lorenzo, R. Schmitt and B. M.Mechler, 2002 P elements inserted in the vicinity of or withinthe Drosophila snRNP SmD3 gene nested in the first intron ofthe ornithine decarboxylase antizyme gene affect only the ex-pression of SmD3. Genetics 161: 763–772.
Schweitzer, R., R. Howes, R. Smith, B. Z. Shilo and M. Freeman,1995 Inhibition of Drosophila EGF receptor activation by thesecreted protein Argos. Nature 376: 699–702.
Serano, J., and G. M. Rubin, 2003 The Drosophila synaptotagmin-like protein bitesize is required for growth and has mRNA local-ization sequences within its open reading frame. Proc. Natl.Acad. Sci. USA 100: 13368–13373.
Shilo, B. Z., 2005 Regulating the dynamics of EGF receptor signal-ing in space and time. Development 132: 4017–4027.
Shinza-Kameda, M., E. Takasu, K. Sakurai, S. Hayashi and A. Nose,2006 Regulation of layer-specific targeting by reciprocal expres-sion of a cell adhesion molecule, capricious. Neuron 49: 205–213.
Shishido, E., M. Takeichi and A. Nose, 1998 Drosophila synapseformation: regulation by transmembrane protein with Leu-richrepeats, CAPRICIOUS. Science 280: 2118–2121.
Steigemann,P.,A.Molitor,S.Fellert,H.JackleandG.Vorbruggen,2004 Heparan sulfate proteoglycan syndecan promotes axonaland myotube guidance by slit/robo signaling. Curr. Biol. 14:225–230.
Strumpf, D., and T. Volk, 1998 Kakapo, a novel cytoskeletal-associated protein is essential for the restricted localization ofthe neuregulin-like factor, vein, at the muscle-tendon junctionsite. J. Cell Biol. 143: 1259–1270.
Subramanian, A., A. Prokop, M. Yamamoto, K. Sugimura, T.Uemura et al., 2003 Shortstop recruits EB1/APC1 and pro-motes microtubule assembly at the muscle-tendon junction.Curr. Biol. 13: 1086–1095.
Svendsen, P. C., S. D. Marshall, M. Kyba and W. J. Brook,2000 The combgap locus encodes a zinc-finger protein that reg-ulates cubitus interruptus during limb development in Drosoph-ila melanogaster. Development 127: 4083–4093.
Szul, T., R. Grabski, S. Lyons, Y. Morohashi, S. Shestopal et al.,2007 Dissecting the role of the ARF guanine nucleotide ex-change factor GBF1 in Golgi biogenesis and protein trafficking.J. Cell Sci. 120: 3929–3940.
Tautz, D., and C. Pfeifle, 1989 A non-radioactive in situ hybridiza-tion method for the localization of specific RNAs in Drosophilaembryos reveals translational control of the segmentation genehunchback. Chromosoma 98: 81–85.
The, I., Y. Bellaiche and N. Perrimon, 1999 Hedgehog movementis regulated through tout velu-dependent synthesis of a heparansulfate proteoglycan. Mol. Cell 4: 633–639.
Townsley, F. M., and M. Bienz, 2000 Actin-dependent membraneassociation of a Drosophila epithelial APC protein and its effecton junctional Armadillo. Curr. Biol. 10: 1339–1348.
Verheyen, E. M., and L. Cooley, 1994 Profilin mutations disruptmultiple actin-dependent processes during Drosophila develop-ment. Development 120: 717–728.
Vining, M. S., P. L. Bradley, C. A. Comeaux and D. J. Andrew,2005 Organ positioning in Drosophila requires complex tis-sue-tissue interactions. Dev. Biol. 287: 19–34.
Wei, Y., L. Yu, J. Bowen, M. A. Gorovsky and C. D. Allis,1999 Phosphorylation of histone H3 is required for properchromosome condensation and segregation. Cell 97: 99–109.
Xu, H., J. A. Brill, J. Hsien, R. McBride, G. L. Boulianne et al.,2002 Syntaxin 5 is required for cytokinesis and spermatid differ-entiation in Drosophila. Dev. Biol. 251: 294–306.
Xu, N., B. Keung and M. M. Myat, 2008 Rho GTPase controls in-vagination and cohesive migration of the Drosophila salivarygland through Crumbs and Rho-kinase. Dev. Biol. 321: 88–100.
Yarmola, E. G., and M. R. Bubb, 2006 Profilin: emerging conceptsand lingering misconceptions. Trends Biochem. Sci. 31: 197–205.
Zhou, B., A. Bagri and S. K. Beckendorf, 2001 Salivary glanddetermination in Drosophila: a salivary-specific, fork head en-hancer integrates spatial pattern and allows fork head autoregu-lation. Dev. Biol. 237: 54–67.
Zhu, S., R. Perez, M. Pan and T. Lee, 2005 Requirement of Cul3 foraxonal arborization and dendritic elaboration in Drosophilamushroom body neurons. J. Neurosci. 25: 4189–4197.
Communicating editor: T. Schupbach
Salivary Gland Morphogenesis in Drosophila 565