LETTER
doi:10.1038/nature12335
Epithelial junctions maintain tissue architecture by
directing planar spindle orientation
Yu-ichiro Nakajima1, Emily J. Meyer1, Amanda Kroesen1, Sean A. McKinney1 & Matthew C. Gibson1,2
During epithelial cell proliferation, planar alignment of the mitotic
spindle coordinates the local process of symmetric cell cleavage with
the global maintenance of polarized tissue architecture1,2. Although
the disruption of planar spindle alignment is proposed to cause epithelial to mesenchymal transition and cancer3–6, the in vivo mechanisms regulating mitotic spindle orientation remain elusive. Here we
demonstrate that the actomyosin cortex and the junction-localized
neoplastic tumour suppressors Scribbled and Discs large 1 have
essential roles in planar spindle alignment and thus the control of
epithelial integrity in the Drosophila imaginal disc. We show that
defective alignment of the mitotic spindle correlates with cell delamination and apoptotic death, and that blocking the death of misaligned cells is sufficient to drive the formation of basally localized
tumour-like masses. These findings indicate a key role for junctionmediated spindle alignment in the maintenance of epithelial integrity, and also reveal a previously unknown cell-death-mediated
tumour-suppressor function inherent in the polarized architecture
of epithelia.
Throughout the animal kingdom, mitotic division typically initiates
with cell rounding driven by Rho-kinase (ROK)-dependent actomyosin contractility at the cell cortex7,8. In polarized epithelia, this process
is generally followed by planar alignment of the mitotic spindle and cell
cleavage orthogonal to the plane of the epithelium1,7,9–14. These events
effectively coordinate the geometry of chromosome segregation with
the architecture of polarized cell–cell junctions, thereby ensuring continuous epithelial integrity. Nevertheless, despite the critical nature of
CNN MUD PH3
c
Prophase
xy
b
xz
MUD
d
CNN PH3 DNA
DLG CNN–GFP PH3 DNA
Interphase
a
Control
this function, the mechanisms that restrict spindle orientation to the
plane of the epithelium remain poorly understood, as do the precise
consequences of spindle misalignment for epithelial organization.
In order to uncover the mechanism guiding planar orientation of the
mitotic spindle in the Drosophila wing disc epithelium, we first analysed spindle pole movements with a fluorescent centriolar marker,
Asterless–yellow fluorescent protein (ASL–YFP), or a Centrosomin–
green fluorescent protein (CNN–GFP) fusion protein (Fig. 1a and
Supplementary Fig. 1b, c). In interphase cells, centrioles marked with
ASL–YFP consistently localized in a super-apical position, well above
the plane of the adherens junctions or basolateral septate junctions
(Supplementary Fig. 1a, b). At prophase entry, centrosomes moved
basally (Fig. 1b and Supplementary Fig. 1c), and the mitotic spindle
aligned with the plane of the epithelium during metaphase (averaging
6.5 6 4.9u from planar) and anaphase (4.6 6 3.7u from planar; Fig. 1a,
b and Supplementary Fig. 1d, e). Indeed, mitotic centrosomes were
almost always constrained to the region of the cell cortex defined by
NRG–GFP accumulation at the septate junctions (n 5 97 out of 100
metaphase figures; Supplementary Fig. 1d). Accordingly, metaphase
and anaphase spindle orientations were consistently within the maximal
limits set by the septate junction-delimited mitotic zone (estimated at
25.6 6 1.3u, n 5 91).
The mitotic spindle could achieve planar orientation through either
an active mechanism involving spatial cues or, more simply, alignment
with the cellular long axis (for example, Hertwig’s rule). Inconsistent
with a long-axis orientation mechanism, however, both interphase and
Metaphase
(%)90o
80 70
50
40
30
20
60
Telophase
Anaphase
e nub-GAL4>mudRNAi (%)90o
n = 75
50
50
40
40
30
30
20
20
80 70
60
n = 88
50
40
30
20
10
10
10
10
0
0
0
0
Figure 1 | Planar orientation of the mitotic spindle during wing disc
development. a, b, Images of mitotic spindles in xy (a) and xz (b) images. c, In
metaphase, MUD localized to the cell cortex and was enriched on the spindle
poles. Left panel is a merged image of CNN, PH3 to label mitotic chromosomes,
and MUD. Right panel shows MUD alone. d, e, Quantification of mitotic
spindle alignment in control (d) and mud RNAi (e) wing discs. n indicates the
number of spindles observed. Scale bars, 5 mm.
1
Stowers Institute for Medical Research, 1000 East 50th Street, Kansas City, Missouri 64110, USA. 2Department of Anatomy and Cell Biology, University of Kansas Medical Center, 3901 Rainbow Boulevard,
Kansas City, Kansas 66160, USA.
0 0 M O N T H 2 0 1 3 | VO L 0 0 0 | N AT U R E | 1
©2013 Macmillan Publishers Limited. All rights reserved
RESEARCH LETTER
technique and observed similar abnormalities (Supplementary Fig. 3e).
Consistent with these effects, 28.6% of spindles from nub-GAL4.MoeRNAi
were strongly misoriented and 17.9% had milder abnormalities (n 5 112;
Fig. 2h, j and Supplementary Fig. 3c). These phenotypes were confirmed
with independent constructs (Supplementary Fig. 3d), indicating that
ROK and MOE function to control planar spindle orientation during
mitosis.
We next investigated how the plane of the epithelium is interpreted
by mitotic cells. Septate junctions, which define the miotic zone and
correlate with the positioning of the spindle poles (Fig. 1), correspond
to vertebrate tight junctions, and are composed of core components
including Coracle, Neurexin IV and Neuroglian, as well as the associated neoplastic tumour suppressors Scribbled (SCRIB) and Discs large
1 (DLG, also known as DLG1) (Fig. 3a and Supplementary Fig. 4a–d)22.
Intriguingly, mitotic spindle poles localized within 100 nm of septate
junctions in intact wing discs (Fig. 3b, c and Supplementary Fig. 4e).
Further, extensive epithelial disorganization is observed in scrib and dlg
mutant imaginal discs23, raising the possibility that their neoplastic phenotypes partly reflect defects in the control of planar spindle alignment.
To determine the requirements for SCRIB and DLG in mitotic spindle
orientation, we first used RNAi to knock down scrib in developing wing
discs (nub-GAL4.UAS-scribRNAi; Supplementary Fig. 4c, d). Unexpectedly,
given its role in the establishment of epithelial polarity24, cells expressing scrib RNAi exhibited no obvious apicobasal polarity defects in the
late third instar (Fig. 3d and Supplementary Fig. 5). At the same
experimental time point, however, more than 40% of spindles showed
abnormal planar orientation (Fig. 3d–f and Supplementary Fig. 6a).
The frequency of spindle misorientation progressively increased during an abnormal developmental delay (Supplementary Fig. 6b, c).
Consistent with cooperative activity of SCRIB and DLG, DLG localization became diffuse after scrib knockdown (Supplementary Fig. 7a),
and we observed spindle orientation defects in discs expressing dlg
RNAi (Fig. 3g and Supplementary Fig. 7b, c). To confirm these results,
we generated mutant clones using the mosaic analysis with a repressible cell marker (MARCM) technique. Spindle orientation was nearly
randomized in scrib1 mutant cells undergoing mitosis at the apical
epithelial surface (Fig. 3h, i and Supplementary Fig. 8a–c). This contrasts with the tendency for the spindle to reorient perpendicular to the
plane of the epithelium following disruption of actin or ROK (Fig. 2e–j),
CNN F-actin PH3
γ-Tub PH3 DNA
e
γ-Tub PH3 DNA
h
F-actin
xz
PBS
b
c
p-MOE F-actin PH3
xy
a
MOE–GFP DNA
mitotic cells were more extended along their apicobasal axes (data not
shown; Supplementary Fig. 1f), suggesting that molecular cues guide
planar spindle alignment. MUD, a Drosophila orthologue of the vertebrate spindle regulator nuclear mitotic apparatus protein (NuMA), controls spindle orientation in neuroblasts and embryonic epithelia15–17.
In wing discs, MUD was strongly enriched on the spindle poles during
mitosis (Fig. 1c and Supplementary Fig. 2a). To determine the role of
MUD in planar spindle orientation, we expressed an RNA interference
(RNAi) knockdown construct under the control of the wing-specific
driver nubbin-GAL4 (nub-GAL4.UAS-mudRNAi). In mud RNAi discs,
36.4% of the spindles exhibited aberrant planar orientation (n 5 88;
Fig. 1d, e and Supplementary Fig. 2b, c), indicating that MUD functions to position the spindle in the plane of the epithelium.
During asymmetric cell divisions in Drosophila and mammalian
tissue culture, cues embedded in the actomyosin cortex direct spindle
orientation through interaction with astral microtubules18,19. In wing
discs, we observed a pronounced accumulation of F-actin, phosphomyosin and the actin-binding protein moesin (MOE–GFP) at the
cortex of mitotic cells (Fig. 2a, b)7. MOE is the sole Drosophila ezrin/
radixin/moesin (ERM) protein, and upon phospho-activation it crosslinks the plasma membrane to the actomyosin cortex to modulate cell
shape20. Consistent with a potential role for MOE in both mitotic
rounding and planar spindle orientation in the wing disc, phosphoMOE was enriched in mitotic cells, where it localized in a polarized
manner to the basal cortex (Fig. 2c, d and Supplementary Fig. 3a).
To test the role of the cell cortex in planar spindle alignment, we
used cytochalasin D (CytoD) to disrupt actin polymerization and
Y-27632 to disrupt ROK, which phospho-activates both myosin and
MOE20,21. Notably, following a 30-min incubation of live wing discs
with CytoD, mitotic spindles rotated orthogonal to the plane of the
epithelium (78.4% strongly misoriented, n 5 111; Fig. 2e, f and Supplementary Fig. 3b). Inhibition of ROK with Y-27632 caused similar
defects, with 80.8% of spindles becoming severely misoriented (n 5 26;
Fig. 2e, g and Supplementary Fig. 3b). To confirm these results genetically, we knocked down rok and Moe by RNAi. In discs expressing
rok RNAi (nub-GAL4.UAS-rokRNAi), 58.6% of spindles were misoriented orthogonal to the plane of the epithelium, and another 19.5%
exhibited milder defects (n 5 133; Fig. 2h, i and Supplementary Fig. 3c).
We also expressed rok RNAi in cell clones induced by the flp-out/GAL4
(%)
90° 80 70
n = 70
70
60
60
50
50
40
40
30
30
20
20
10
10
0
0
nub-GAL4>GFP (%)
90° 80
70
50
60 n = 90
50
40
30
f
i
CytoD
xz
d
p-MOE
F-actin
xy
(i)
(i)
(ii)
(ii)
(iii)
(iii)
(%)
90° 80 70
n = 111
70
60
60
50
50
40
40
30
30
20
20
10
10
0
0
nub-GAL4>rokRNAi (%)
90° 80
70
50
60 n = 133
50
40
30
g
Y-27632
(%)
90° 80 70
70
60 n = 26
60
50
50
40
40
30
30
20
20
10
10
0
0
j nub-GAL4>MoeRNAi (%)90° 80
50
70
30
60 n = 112
50
40
30
20
20
20
20
20
20
10
10
10
10
10
10
0
0
0
0
0
0
40
30
Figure 2 | The actomyosin cortex is required for planar spindle orientation.
a, F-actin accumulated at the cortex of mitotic cells. b, MOE–GFP localized to
the cortex of mitotic cells in the presumptive wing pouch. c, d, Phospho-MOE
(p-MOE) was specifically enriched in the basal cortex of mitotic cells. Panels in
d show different planes according to the z-positions indicated in
40
30
40
c. e–g, Quantification of mitotic spindle alignments in discs treated with PBS
(control) (e), CytoD (f) and Y-27632 (g) for 30 min. h–j, Quantification of
mitotic spindle alignment in wing discs expressing GFP (control) (h), rokRNAi
(i) and MoeRNAi (j). n indicates the number of spindles observed. c-Tub,
c-tubulin. Scale bars, 5 mm.
2 | N AT U R E | VO L 0 0 0 | 0 0 M O N T H 2 0 1 3
©2013 Macmillan Publishers Limited. All rights reserved
LETTER RESEARCH
a
SCRIB–GFP γ-Tub PH3 DNA
Metaphase
c
b
h
f
SJ
nub-GAL4>scribRNAi (%)
90° 80
50
Control
70
40
30
Telophase
CNN F-actin PH3
1 μm
ECAD F-actin PH3 DNA
e
ECAD
30
n = 91
50
40
30
20
20
10
10
10
0
0
0
0
20
GFP γ-Tub PH3
Control
i
g
nub-GAL4>dlgRNAi (%)
90° 80
70
40
30
60
scrib1
(%) 90°
80 70
50
n = 59
50
40
30
20
40
30
20
20
60
n = 129
50
40
30
20
10
10
10
10
0
0
0
0
γ-Tub F-actin PH3
GFP γ-Tub PH3
localization of ECAD. e, Magnification of the box in d. f, g, Quantification of
mitotic spindle alignments in scrib RNAi (f) and dlg RNAi (g) wing discs
compared with control data from Fig. 1d. h, i, Quantification of metaphase
spindle alignments in control and scrib1 MARCM clones. The red line shows
the median angular deviation. n indicates the number of spindles observed.
Scale bars, 5 mm (a, e–i) and 20 mm (d).
nub-GAL4>rokRNAi
nub-GAL4
a
d
ARM
CASP3
DNA
e
nub-GAL4>p35+rokRNAi
nub-GAL4>p35
c
b
ARM
F-actin
DNA
p35
rokRNAi+p35
f
g
mudRNAi+p35
GFP
F-actin
DNA
h
F-actin
mudRNAi+p35
MMP1
GFP F-actin MMP1 DNA
perhaps owing to defects in mitotic cell rounding in those manipulations
(Supplementary Fig. 9). Taken together, these results show that SCRIB
and DLG function as determinants of planar spindle orientation and
raise the possibility that the neoplastic phenotypes associated with
lesions in scrib or dlg partly result from spindle orientation defects
(Fig. 3d, e and Supplementary Fig. 6). Hypothetically, a primary defect
in spindle orientation could cause a secondary breakdown of junctional
architecture, leading to what is effectively an epithelial-to-mesenchymal
transition (EMT).
Loss of tissue architecture is a hallmark of epithelial cancer25, and
our results indicate that defects in planar spindle orientation could be a
critical initiating event in neoplastic overgrowth of scrib and dlg mutant
discs (Supplementary Fig. 6c). However, misorientation of the spindle
may not be sufficient to initiate tumorigenesis, as neoplastic phenotypes
were not observed after mud or rok knockdown. Instead, in nubGAL4.rokRNAi discs and in clones expressing rok RNAi, basally localized cells lost polarity markers, underwent apoptotic cell death, and
were removed from the epithelium, as previously observed in Moe
mutants (Fig. 4a, b and Supplementary Fig. 10a)26. To determine the
consequence of spindle misorientation, we established a custom selective
plane illumination microscopy (SPIM) system. In nub-GAL4.rokRNAi
wing discs, the basal daughters of severely misoriented divisions exhibited rapid basal movements, consistent with their eventual delamination
from the epithelium (Supplementary Fig. 10d–f and Supplementary
Videos 1, 2). Further, basally delaminated cells frequently exhibited
centrosomal accumulation of CNN–GFP in the absence of mitotic
markers (Supplementary Fig. 10g–i), suggesting that these cells may
have undergone abnormal divisions and delaminated from the epithelium before their programmed death.
To block apoptosis of misaligned daughter cells in nub-GAL4.rokRNAi
discs, we co-expressed the caspase inhibitor p35. This resulted in a highly
disorganized epithelial structure when compared to nub-GAL4.rokRNAi
alone (Fig. 4c, d). Intriguingly, increased numbers of apoptotic cells were
also observed following scrib knockdown (Supplementary Fig. 10b),
and the neoplastic phenotype was again strongly enhanced by caspase
inhibition (Supplementary Fig. 11a). Combined, these results suggest
that abnormal epithelial cells arising from spindle misalignment are
normally cleared by apoptotic cell death and thereby prevented from
disrupting epithelial organization.
To extend these findings as a model for EMT by aberrant spindle
orientation, we knocked down rok in cell clones co-expressing UAS-p35.
60
10
50
Figure 3 | SCRIB and DLG determine planar orientation of the mitotic
spindle. a, SCRIB–GFP localized to septate junctions, correlating with spindle
position. b, A rounded anaphase cell (yellow line) exhibits chromosome
segregation (arrows). c, Centrioles (arrows) localized in close proximity to
septate junctions (SJ, bracket) in a magnification of the red box from
b. d, e, nub-GAL4.UAS-scribRNAi wing discs exhibit misoriented spindles
(dotted line in e) despite normal epithelial architecture and normal apical
40
20
200 nm
nub-GAL4>scribRNAi
d
(%) 90°
80 70
50
60 n = 138
50
40
30
i
Apoptosis
Planar orientation
SJs
Misorientation
Delamination
EMT
Centrosome
Figure 4 | In the absence of apoptosis, spindle misorientation induces EMTlike effects. a, b, rok knockdown induced basal cell delamination, loss of the
Armadillo polarity marker (ARM), and frequent apoptosis. c, d, Suppression of
cell death with p35-enhanced tissue disorganization in rok RNAi wing discs.
e, Control clones expressing p35 were normally integrated in the epithelial
layer. By contrast, GFP1 cell clones expressing p35 with rok RNAi (f) or mud
RNAi (g, h) exhibited morphological and molecular characteristics of EMT.
h, Strong MMP1 expression and actin remodelling observed in a basal tumourlike mass. i, Conceptual model for the contribution of spindle misorientation to
aberrant EMT. Scale bars, 20 mm.
0 0 M O N T H 2 0 1 3 | VO L 0 0 0 | N AT U R E | 3
©2013 Macmillan Publishers Limited. All rights reserved
RESEARCH LETTER
These clones formed large disorganized masses of aberrant cells on the
basal side of the disc (n 5 55 out of 55 discs, Fig. 4e, f). Intriguingly, the
abnormal cells exhibited molecular characteristics of EMT27, including
loss of E-cadherin, induction of Matrix metalloproteinase 1 (MMP1)
and actin remodelling (Supplementary Fig. 12a, c, e). A similar EMTlike effect was observed in scrib RNAi clones co-expressing p35 (n 5 40
out of 61 discs; Supplementary Fig. 11b), as well as in scrib1 mutant cells
expressing UAS-p35 (Supplementary Fig. 11c). Finally, to more directly
implicate spindle misorientation in this process, we used UAS-p35 to
suppress the cell death normally observed in cells expressing mud
RNAi (Supplementary Fig. 10c). This manipulation phenocopied the
effects of disrupting rok or scrib (n 5 33 out of 38 discs; Fig. 4g, h and
Supplementary Fig. 12b, d, f), demonstrating that in the absence of
corrective cell death, spindle misorientation is alone sufficient to induce
many key features of EMT. Crucially, the inhibition of apoptosis in
dying cells did not lead to the formation of basal tumour-like masses,
indicating that the EMT-like effect cannot simply be attributed to
forced cell survival (n 5 1 out of 54 discs; Supplementary Fig. 11d).
Combined, these results show that misorientation of the mitotic
spindle is typically corrected by cell death, but can result in EMT under
conditions where normal apoptotic mechanisms are compromised
(Fig. 4i).
Here we demonstrate that planar alignment of the mitotic spindle
requires interactions between the mitotic apparatus, the actomyosin
cortex and the septate-junction-localized scaffolding proteins SCRIB
and DLG. Although SCRIB and DLG have long been implicated in
the suppression of neoplastic tumours in Drosophila and humans23,28,29,
our results uncover a novel requirement for both proteins during
alignment of the mitotic spindle to the plane of the epithelium
(Fig. 3). By contrast, other potential tumour suppressors linked to cell
polarity did not affect planar spindle orientation in the wing disc
(Supplementary Table 1). On the basis of these results, we propose
that aberrant cleavage plane orientation could be a significant contributor to the loss of epithelial integrity in scrib or dlg mutant cells (Fig. 4i).
In addition to the above findings, our results demonstrate a critical
function for planar orientation of the mitotic spindle in the maintenance of epithelial architecture. Consistent with this view, disruption of
spindle orientation with rok or mud RNAi was sufficient to trigger
delamination of basal daughter cells (Supplementary Fig. 10 and Supplementary Videos 1, 2). When combined with the suppression of cell
death, these manipulations produced delaminated mesenchyme-like
masses exhibiting some molecular characteristics of EMT (Fig. 4h and
Supplementary Fig. 12). During the early steps of tumour progression
in humans, sporadic mutant cells are thought to acquire a potential to
outgrow normal tissue by escaping from the epithelial sheet through
EMT or EMT-like events5,30. On the basis of our findings, we propose
that the deleterious effects of aberrant spindle alignment are typically
corrected by apoptosis, and that suppression of this corrective mechanism could be a common initial driver of epithelial dysplasia and
tumorigenesis in vivo.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
Pease, J. C. & Tirnauer, J. S. Mitotic spindle misorientation in cancer–out of
alignment and into the fire. J. Cell Sci. 124, 1007–1016 (2011).
Vasiliev, J. M., Omelchenko, T., Gelfand, I. M., Feder, H. H. & Bonder, E. M. Rho
overexpression leads to mitosis-associated detachment of cells from epithelial
sheets: a link to the mechanism of cancer dissemination. Proc. Natl Acad. Sci. USA
101, 12526–12530 (2004).
Fleming, E. S., Temchin, M., Wu, Q., Maggio-Price, L. & Tirnauer, J. S. Spindle
misorientation in tumors from APCmin/1 mice. Mol. Carcinog. 48, 592–598 (2009).
Meyer, E. J., Ikmi, A. & Gibson, M. C. Interkinetic nuclear migration is a broadly
conserved feature of cell division in pseudostratified epithelia. Curr. Biol. 21,
485–491 (2011).
Spear, P. C. & Erickson, C. A. Apical movement during interkinetic nuclear
migration is a two-step process. Dev. Biol. 370, 33–41 (2012).
Reinsch, S. & Karsenti, E. Orientation of spindle axis and distribution of plasma
membrane proteins during cell division in polarized MDCKII cells. J. Cell Biol. 126,
1509–1526 (1994).
Gibson, M. C., Patel, A. B., Nagpal, R. & Perrimon, N. The emergence of geometric
order in proliferating metazoan epithelia. Nature 442, 1038–1041 (2006).
Fleming, E. S. et al. Planar spindle orientation and asymmetric cytokinesis in the
mouse small intestine. J. Histochem. Cytochem. 55, 1173–1180 (2007).
Lu, B., Roegiers, F., Jan, L. Y. & Jan, Y. N. Adherens junctions inhibit asymmetric
division in the Drosophila epithelium. Nature 409, 522–525 (2001).
Ségalen, M. et al. The Fz-Dsh planar cell polarity pathway induces oriented cell
division via Mud/NuMA in Drosophila and zebrafish. Dev. Cell 19, 740–752 (2010).
Guilgur, L. G., Prudêncio, P., Ferreira, T., Pimenta-Marques, A. R. & Martinho, R. G.
Drosophila aPKC is required for mitotic spindle orientation during symmetric
division of epithelial cells. Development 139, 503–513 (2012).
Siller, K. H., Cabernard, C. & Doe, C. Q. The NuMA-related Mud protein binds Pins
and regulates spindle orientation in Drosophila neuroblasts. Nature Cell Biol. 8,
594–600 (2006).
Izumi, Y., Ohta, N., Hisata, K., Raabe, T. & Matsuzaki, F. Drosophila Pins-binding
protein Mud regulates spindle-polarity coupling and centrosome organization.
Nature Cell Biol. 8, 586–593 (2006).
Bowman, S. K., Neumüller, R. A., Novatchkova, M., Du, Q. & Knoblich, J. A. The
Drosophila NuMA homolog Mud regulates spindle orientation in asymmetric cell
division. Dev. Cell 10, 731–742 (2006).
Siller, K. H. & Doe, C. Q. Spindle orientation during asymmetric cell division. Nature
Cell Biol. 11, 365–374 (2009).
Sandquist, J. C., Kita, A. M. & Bement, W. M. And the dead shall rise: actin and
myosin return to the spindle. Dev. Cell 21, 410–419 (2011).
Fehon, R. G., McClatchey, A. I. & Bretscher, A. Organizing the cell cortex: the role of
ERM proteins. Nature Rev. Mol. Cell Biol. 11, 276–287 (2010).
Quintin, S., Gally, C. & Labouesse, M. Epithelial morphogenesis in embryos:
asymmetries, motors and brakes. Trends Genet. 24, 221–230 (2008).
Tepass, U., Tanentzapf, G., Ward, R. & Fehon, R. Epithelial cell polarity and cell
junctions in Drosophila. Annu. Rev. Genet. 35, 747–784 (2001).
Bilder, D., Li, M. & Perrimon, N. Cooperative regulation of cell polarity and growth
by Drosophila tumor suppressors. Science 289, 113–116 (2000).
Bilder, D. & Perrimon, N. Localization of apical epithelial determinants by the
basolateral PDZ protein Scribble. Nature 403, 676–680 (2000).
Royer, C. & Lu, X. Epithelial cell polarity: a major gatekeeper against cancer? Cell
Death Differ. 18, 1470–1477 (2011).
Speck, O., Hughes, S. C., Noren, N. K., Kulikauskas, R. M. & Fehon, R. G. Moesin
functions antagonistically to the Rho pathway to maintain epithelial integrity.
Nature 421, 83–87 (2003).
Thiery, J. P., Acloque, H., Huang, R. Y. J. & Nieto, M. A. Epithelial-mesenchymal
transitions in development and disease. Cell 139, 871–890 (2009).
Pagliarini, R. A. & Xu, T. A genetic screen in Drosophila for metastatic behavior.
Science 302, 1227–1231 (2003).
Humbert, P. O. et al. Control of tumourigenesis by the Scribble/Dlg/Lgl polarity
module. Oncogene 27, 6888–6907 (2008).
Leung, C. T. & Brugge, J. S. Outgrowth of single oncogene-expressing cells from
suppressive epithelial environments. Nature 482, 410–413 (2012).
Supplementary Information is available in the online version of the paper.
METHODS SUMMARY
Detailed information including the fly stocks, immunofluorescence, drug treatment, imaging and image analysis, and associated references, are included in
Methods.
Full Methods and any associated references are available in the online version of
the paper.
Received 24 October 2012; accepted 24 May 2013.
Published online 21 July 2013.
1.
2.
3.
Morin, X. & Bellaı̈che, Y. Mitotic spindle orientation in asymmetric and symmetric
cell divisions during animal development. Dev. Cell 21, 102–119 (2011).
Gillies, T. E. & Cabernard, C. Cell division orientation in animals. Curr. Biol. 21,
R599–R609 (2011).
Noatynska, A., Gotta, M. & Meraldi, P. Mitotic spindle (DIS)orientation and DISease:
cause or consequence? J. Cell Biol. 199, 1025–1035 (2012).
Acknowledgements We thank D. Bilder, M. Miura, Y. Yamashita, the Bloomington Stock
Center, Vienna Drosophila RNAi Center, TRiP, and FlyTrap project for fly stocks and
C. Doe, F. Matsuzaki and the Developmental Studies Hybridoma Bank for antibodies.
We thank R. Fehon for suggestions. We thank T. Akiyama, A. Fritz, A. Ikmi and
M. Szuperak for comments on the manuscript and L. Liang for technical advice. We also
thank M. Gogol for graphical assistance, F. Guo for TEM support and L. Gutchewsky for
administrative support. This work was supported by the Stowers Institute for Medical
Research and the Burroughs Wellcome Fund for Biomedical Research.
Author Contributions Y.-I.N. and M.C.G. conceived the project, designed the
experiments and wrote the manuscript. Y.-I.N. and E.J.M. performed the experiments
and analysed the data. A.K. and S.A.M. constructed the SPIM system and performed live
imaging.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial interests.
Readers are welcome to comment on the online version of the paper. Correspondence
and requests for materials should be addressed to M.C.G. (MG2@stowers.org).
4 | N AT U R E | VO L 0 0 0 | 0 0 M O N T H 2 0 1 3
©2013 Macmillan Publishers Limited. All rights reserved
LETTER RESEARCH
METHODS
Fly stocks and genetics. The following stocks were used: OreR for wild-type
control, A9-GAL4, UAS-cnn-GFP31, nub-GAL4, His2Av-mRFP (monomeric red
fluorescent protein), UAS-Moe-GFP32, UAS-p35, UAS-reaper, UAS-GFP, UASDicer-2, Asl-YFP33, Nrg-GFP34, scrib-GFP35, hs-flp UAS-mCD8-GFP; Tub-GAL4
FRT82B Tub-GAL80 for MARCM36 clones, yw; FRT82B, yw; FRT82B scrib1/
TM6C23, hs-flp; Act.y1.GAL4 UAS-GFP for FLP/FRT-mediated (FLP-out)
clones, UAS-mudRNAi (Bloomington 28074, Bloomington 35044), UAS-rokRNAi
(Bloomington 28797, Bloomington 34324), UAS-MoeRNAi (VDRC 37917, Bloomington
33936), UAS-scribRNAi (Bloomington 35748), UAS-dlgRNAi (Bloomington 33629,
Bloomington 25780) for RNAi knockdown experiments (VDRC37 and TRiP38).
Larvae were collected at 25 uC. Clones of RNAi constructs or mutants were generated with a 1-h heat shock at 48 h after egg lay.
Immunofluorescence. The following antibodies and dyes were used: rabbit antiphospho-histone H3 (1:1,000, Millipore), mouse anti-phospho-histone H3 (1:2,000,
Millipore), mouse anti-c-tubulin (1:1,000, Sigma), rat anti-a-tubulin (1:250, Serotec),
rabbit anti-phospho-MOE (1:100, Cell Signaling), rabbit anti-cleaved CASP3 (1:500,
Cell Signaling), guinea pig anti-CNN (1:1,000, F. Matsuzaki), mouse anti-MUD
(1:50, F. Matsuzaki), rabbit anti-SCRIB (1:5,000, C. Doe), rabbit anti-BAZ (1:500,
F. Matsuzaki), mouse anti-DLG (1:500, DSHB), mouse anti-ARM (1:200, DSHB),
rat anti-ECAD (1:25, DSHB), mouse anti-CORA (1:100, C615.16, DSHB), mouse
anti-FAS3 (1:100, DSHB), mouse anti-MMP1 (1:100, 1:1:1 cocktail of 5H7B11,
3A6B4, 3B8D12, DSHB), rabbit anti-aPKC-f C20 (1:1,000, Santa Cruz Biotechnology), fluorescent second antibodies (1:500, Invitrogen), Alexa Phalloidin 488,
546 (1:500, Invitrogen) and Hoechst 33342 (2 mM, Thermo Scientific).
Drug treatments. CytoD(5 mM,Calbiochem)andY-27632(ref.39)(1 mM,Calbiochem)
were diluted in PBS and wing discs were incubated for 30 min as described7. These
concentrations do not disrupt wing epithelial morphology during the assay period.
Imaging and image analysis. Confocal images were collected with a 363 glycerol
or 340 oil objective lens on a Leica SP5 AOBS confocal microscope system (Leica
Microsystems). To measure mitotic spindle orientation, images of mitotic spindles
were taken by optical cross section (xz scan). The angle between the line perpendicular to the epithelium and the line through the spindle poles was measured with
Fiji/ImageJ (Supplementary Fig. 1d). On the basis of the maximal outer limits
defined by the average length of the septate junctions (25.6 6 1.3u, n 5 91), we
scored spindles as abnormal when angular deviation from planar (h), h $ 30u. Values
60 # h # 90u were considered strong misorientation and values 30 # h # 60u were
considered mild. In Figs 1–3, prometaphase/metaphase cells are shown for the
quantification of mitotic spindle orientation. Spindle orientation in anaphase/
telophase cells are in Supplementary Information.
SPIM. Wing discs were embedded in 1% agarose in Drosophila Ringers’ solution.
Embedded wing discs were then placed in a chamber filled with fly media40 and
imaged every minute for 2–3 h at approximately 21 uC using SPIM41. Fifty Zsections with 1-mm spacing were collected with an exposure time of 27 ms using
488 nm and 561 nm laser excitation scanned sequentially and detected using a W
Plan-Apochromat 340/1.0 NA objective. Further details can be found in
Supplementary Information.
31. Megraw, T. L., Kilaru, S., Turner, F. R. & Kaufman, T. C. The centrosome is a dynamic
structure that ejects PCM flares. J. Cell Sci. 115, 4707–4718 (2002).
32. Bloor, J. W. & Kiehart, D. P. zipper nonmuscle myosin-II functions downstream of
PS2 integrin in Drosophila myogenesis and is necessary for myofibril formation.
Dev. Biol. 239, 215–228 (2001).
33. Varmark, H. et al. Asterless is a centriolar protein required for centrosome
function and embryo development in Drosophila. Curr. Biol. 17, 1735–1745
(2007).
34. Morin, X., Daneman, R., Zavortink, M. & Chia, W. A protein trap strategy to detect
GFP-tagged proteins expressed from their endogenous loci in Drosophila. Proc.
Natl Acad. Sci. USA 98, 15050–15055 (2001).
35. Buszczak, M. et al. The carnegie protein trap library: a versatile tool for Drosophila
developmental studies. Genetics 175, 1505–1531 (2007).
36. Lee, T. & Luo, L. Mosaic analysis with a repressible cell marker for studies of gene
function in neuronal morphogenesis. Neuron 22, 451–461 (1999).
37. Dietzl, G. et al. A genome-wide transgenic RNAi library for conditional gene
inactivation in Drosophila. Nature 448, 151–156 (2007).
38. Ni, J. Q. et al. A genome-scale shRNA resource for transgenic RNAi in Drosophila.
Nature Methods 8, 405–407 (2011).
39. Uehata, M. et al. Calcium sensitization of smooth muscle mediated by a Rhoassociated protein kinase in hypertension. Nature 389, 990–994 (1997).
40. Gibson, W. T. et al. Control of the mitotic cleavage plane by local epithelial topology.
Cell 144, 427–438 (2011).
41. Keller, P. J., Schmidt, A. D., Wittbrodt, J. & Stelzer, E. H. K. Reconstruction of
zebrafish early embryonic development by scanned light sheet microscopy.
Science 322, 1065–1069 (2008).
©2013 Macmillan Publishers Limited. All rights reserved