Control of mitochondrial structure
and function by the Yorkie/YAP
Raghavendra Nagaraj,1,2,6Shubha Gururaja-Rao,1,2,6Kevin T. Jones,1,2Matthew Slattery,3
Nicolas Negre,4Daniel Braas,5Heather Christofk,5Kevin P. White,4Richard Mann,3
and Utpal Banerjee1,2,7
1Department of Molecular, Cell, and Developmental Biology,2Department of Biological Chemistry, Molecular Biology Institute,
Broad Stem Cell Research Center, University of California at Los Angeles, Los Angeles, California 90095, USA;3Department of
Biochemistry and Molecular Biophysics, Columbia University, New York, New York 10032, USA;4Institute for Genomics and
Systems Biology, University of Chicago, Chicago, Illinois 60637, USA;5Institute for Molecular Medicine, David Geffen School of
Medicine, University of California at Los Angeles, Los Angeles, California 90095, USA
Mitochondrial structure and function are highly dynamic, but the potential roles for cell signaling pathways in
influencing these properties are not fully understood. Reduced mitochondrial function has been shown to cause
cell cycle arrest, and a direct role of signaling pathways in controlling mitochondrial function during development
and disease is an active area of investigation. Here, we show that the conserved Yorkie/YAP signaling pathway
implicated in the control of organ size also functions in the regulation of mitochondria in Drosophila as well as
human cells. In Drosophila, activation of Yorkie causes direct transcriptional up-regulation of genes that regulate
mitochondrial fusion, such as opa1-like (opa1) and mitochondria assembly regulatory factor (Marf), and results in
fused mitochondria with dramatic reduction in reactive oxygen species (ROS) levels. When mitochondrial fusion
is genetically attenuated, the Yorkie-induced cell proliferation and tissue overgrowth are significantly suppressed.
The function of Yorkie is conserved across evolution, as activation of YAP2 in human cell lines causes increased
mitochondrial fusion. Thus, mitochondrial fusion is an essential and direct target of the Yorkie/YAP pathway in
the regulation of organ size control during development and could play a similar role in the genesis of cancer.
[Keywords: Hippo pathway; ROS; Yap; Yorkie; marf; mitochondrial fusion; opa]
Supplemental material is available for this article.
Received November 10, 2011; revised version accepted July 27, 2012.
The Mst/Lats/YAP pathway functions in the process of
contact inhibition to regulate organ size inmammals, and
loss in this pathway’s function is implicated in multiple
cancer types (Zhao et al. 2008). In Drosophila, where this
pathway, commonly referred to as the ‘‘Hippo pathway,’’
was first identified and best understood, the correspond-
ing proteins are Hippo/Warts/Yorkie (Hariharan 2006;
Pan 2007; Zhang et al. 2009). Loss-of-function mutations
in any upstream component of the Hippo pathway will
cause Yorkie (Yki) to remain unphosphorylated and
actively signal within the nucleus, causing an overgrowth
phenotype (Dong et al. 2007; Oh and Irvine 2008). YAP/
Yki functions as a transcriptional coactivator, interacting
with a DNA-binding partner such as mammalian TEAD
proteins/Drosophila Scalloped (Sd) (Huang et al. 2005;
Goulev et al. 2008; Wu et al. 2008; Zhang et al. 2008). In
Drosophila, the Hippo pathway promotes proliferation by
positively regulating cyclin E and the microRNA bantam
(Harvey et al. 2003; Wu et al. 2003; Huang et al. 2005;
Nolo et al. 2006; Thompson and Cohen 2006). This
pathway also inhibits apoptosis by controlling DIAP1
expression (Harvey et al. 2003; Huang et al. 2005; Dong
et al. 2007; Oh and Irvine 2008). Overexpression of Yki in
the late larval and pupal eye disc cells causes an increased
adult eye size (Huang et al. 2005; Dong et al. 2007). This
phenotype is further enhanced by coexpression of Sd
(Goulev et al. 2008; Wu et al. 2008; Zhang et al. 2008).
The extreme effect of this pathway on growth and its
prominent role in cancer progression prompted us to
investigate a possible link to cellular metabolism.
Mitochondrial phenotype upon Yki pathway activation
GFP targeted to the mitochondrial matrix (mitoGFP) is
trapped and stabilized and allows visualization of the
6These authors contributed equally to this work.
Article published online ahead of print. Article and publication date are
online at http://www.genesdev.org/cgi/doi/10.1101/gad.183061.111.
GENES & DEVELOPMENT 26:2027–2037 ? 2012 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/12; www.genesdev.org 2027
mitochondrial network (Clark et al. 2006; Goyal et al.
2007; Deng et al. 2008). This reagent was used to assess
the effects of activating the yki/sd pathway on mitochon-
dria. In the wild-type pupal eye disc, the majority of
mitochondria are oval in shape and scattered around the
cell with no visible signs of an interconnected network
(Fig. 1A; Supplemental Fig. S1). Upon overexpression of
Sd, the DNA-binding partner of Yki, only a modest in-
crease in mitochondrial size seems evident (Fig. 1B). How-
ever, overexpression of Yki (GMR-Gal4, UAS-yki) causes
Fig. S1). This phenotype is further enhanced upon a simul-
taneous coexpression of Yki and Sd (Fig. 1D). Inactivation
of sd in a GMR-Gal4, UAS-yki mutant background by
either a single-copy loss of sd or a coexpression of sdRNAi
suppresses the Yki-induced mitochondrial expansion
(Fig. 1E,F). In contrast, knocking down Homothorax (hth),
another known transcriptional partner of Yorkie (Peng et al.
2009), in a Yorkie overexpression background did not
modify the mitochondrial phenotypes of Yorkie (data not
shown), suggesting that the control of mitochondria by this
pathway involves the Yki/Sd and not the Yki/Hth complex.
The phenotype of an enhanced mitochondrial network
upon Yki/Sd expression is not specific to the eye, as it is
also seen in the wing and leg imaginal discs (Fig. 1G–L;
additional controls in Supplemental Figs. S2, S3).
To determine whether Yki functions autonomously in
mitochondrial control, we generated mutant clones of the
canonical Hippo pathway components: hippo, which
encodes an upstream Ste-20 class kinase (Harvey et al.
2003; Wu et al. 2003); warts (wts), which encodes the ter-
minal kinase that phosphorylates Yki (Justice et al. 1995;
Harvey et al. 2003); and fat (ft), which encodes a receptor for
the Hippo pathway (Bennett and Harvey 2006; Silva et al.
2006). We alsogenerated clones thatoverexpress Yki. In all
cases, mutant cells show an autonomous increase in the
expression of mitochondrial markers (Fig. 2A–J). Further-
more, down-regulation of wts and ft using RNAi in the
dorsal compartment of the wing disc also causes an
increase in mitochondrial staining within the mutant
tissue (Fig. 2K,L; Supplemental Fig. S4). Reduction of Yki
by RNAi suppresses the increased mitochondrial staining
observed in wts mutant cells, showing that yki functions
downstream from wts to regulate mitochondrial expan-
sion (Supplemental Fig. S5). The microRNA bantam is a
transcriptional target of Yki, and its overexpression causes
increased cell proliferation and organ size overgrowth
(Nolo et al. 2006; Thompson and Cohen 2006). However,
overexpression of bantam does not result in a mitochon-
drial phenotype (Supplemental Fig. S6A–D). Furthermore,
although bantam overexpression enhances tissue over-
growth due to Yki, no enhancement of the Yki mitochon-
drial phenotype by bantam overexpression is apparent in
the eye disc (Supplemental Fig. S6E–H).
The Yki mitochondrial phenotype is independent
of cell proliferation
The use of GMR-Gal4 as a driver to express Yki causes
cells to proliferate, making it difficult to assess whether
the observed change in mitochondrial morphology is a
primary result of the activation of this pathway or a sec-
ondary consequence of increased proliferation. We there-
cone cells (green in Fig. 2M,O) in the eye. The spa-Gal4,
UAS-yki combination does not cause overgrowth, and yet
a robust increase in mitochondrial staining is readily ap-
parent (Fig. 2M–P), suggesting that even in the absence of
matrix. ELAV (red) marks photoreceptor nuclei. Bar, 5 mm. (A) Control (GMR-Gal4, UAS-mitoGFP) mitochondria (green) are seen as
individual dots. (B) Overexpression of Scalloped (GMR-Gal4, UAS-mitoGFP UAS-sd) causes a modest increase in mitochondria. (C)
Overexpression of Yki (GMR-Gal4, UAS-mitoGFP UAS-yki) results in extensive mitochondrial (green) expansion. (D) Combined
overexpression of Yki and Sd (GMR-Gal4, UAS-mitoGFP UAS-yki, UAS-sd) causes further enhancement of the mitochondrial
expansion phenotype. (E) In the background of Yki activation, a reduction in sd function (GMR-Gal4, UAS-mitoGFP, UAS-yki, UAS-
sdRNAi) suppresses the Yki-mediated expansion of mitochondria. (F) Single-copy loss of sd in a Yki-activated background (GMR-Gal4,
UAS-mitoGFP, UAS-yki, sd47M/+) causes a suppression in mitochondrial expansion compared with that seen upon Yki overexpression
(cf. C). (G–L) Third instar wing (G–I) and leg (J–L) imaginal discs. mitoGFP (green) marks the mitochondria. (G) Control (dpp-Gal4, UAS-
mitoGFP) wing disc; the mitochondria (green) are seen as individual dots. (H) Overexpression of Sd (dpp-Gal4, UAS-mitoGFP, UAS-sd)
does not significantly affect mitochondria. (I) A combined overexpression of Yki and Sd in the wing disc (dpp-Gal4, UAS-mitoGFP,
UAS-yki, UAS-sd) results in extensive increase in mitochondria. (J) Control (dpp-Gal4, UAS-mitoGFP) leg disc. (K) Overexpression of
Sd (dpp-Gal4, UAS-mitoGFP, UAS-sd) is similar to control. (L) Combined overexpression of Yki and Sd (dpp-Gal4, UAS-mitoGFP, UAS-
yki, UAS-sd) causes extensive expansion of mitochondria.
Yki activation causes mitochondrial phenotypes. (A–F) Mid-pupal eye discs. mitoGFP (green) marks the mitochondrial
Nagaraj et al.
2028 GENES & DEVELOPMENT
cell proliferation, this pathway is capable of promoting
mitochondrial biogenesis. This phenotype is specific to
Yki, since overexpression of other growth-promoting
factors such as activation of the EGFR or the Wingless
pathway using the spa-Gal4 driver does not cause
a similar increase in mitochondrial biogenesis (Supple-
mental Fig. S7). Furthermore, loss of two other tumor
suppressor genes (scribble and avalanche), each of which
shows an extensive growth phenotype similar to that
seen upon Yki activation, also doesnotresultinincreased
mitochondrial staining (Supplemental Fig. S8).
Finally, we asked whether the increased mitochondrial
staining seen with mitoGFP and ATP-synthase is also
reflected as higher mitochondrial membrane potential by
staining with the dye MitoTracker (Baltzer et al. 2009). As
evidenced by clones generated in the post-mitotic cells of
the fat body, Yki activation causes an increase in mito-
chondrial markers and in the intensity of MitoTracker
staining (Fig. 2Q–T).
The mitochondrial phenotype of Yki is conserved
To determine whether the role of the Hippo pathway in
the control of mitochondrial network formation is con-
served across species, we stably overexpressed a Flag-
tagged form of YAP2 (one of the human homologs of
Yorkie) in several human cancer cell lines, which endog-
enously express YAP2, as assessed by protein blots (data
not shown), and thus might have functional cofactors.
Mitochondriawere analyzedwith MitoTracker Red stain-
ing and an antibody against the mitochondrial protein
ATP synthase-a. In control cells from the breast cancer
cell lines MDAMB453 and SUM159PT, we observed a
small number of mitochondria predominantly localized
in perinuclear regions (Fig. 3A–D). Overexpression of
YAP2 causes an increase in MitoTracker and mitochon-
drial ATP synthase-a staining (Fig. 3A9–D9). We also ex-
amined three additional cell lines derived from different
human cancers that express varying levels of endogenous
YAP2. In two cases, we observed a correlation between
increased YAP2 expression and enhanced mitochondrial
mass, but this correlation was not seen in the third cell
line (data not shown). Thus, YAP expression may be
correlative but is not the sole determinant of mitochon-
drial morphology in human cancer cells. However, the
phenotypes that we observed upon YAP2 overexpression
are remarkably similar to those seen in Drosophila and
suggest that regulation of mitochondria by the Yki path-
way is conserved in humans.
Ultrastructural and metabolic analysis
of mitochondria upon Yki activation
To characterize the mitochondrial phenotype seen above
at the ultrastructural level, we carried out an electron
of wts (green) generated using MARCM and stained for ATP-syn (red in B) show increased staining in the mutant (green) cells. (C–H) Third
instar wing discs containing mutant clones (nongreen) for warts (wts) (C,D), fat (ft) (E,F), and hippo (hpo) (G,H) show increased
mitochondrial staining using ATP synthase-a antibody (red) in the mutant cells (nongreen) compared with adjacent wild-type tissue
(green). (B,D,F,H) Red channel alone is shown for clarity. Bar, 50 mm. (I,J) Clones of cells in a wing disc expressing Yki (green) using the Ay-
Gal4 system (see the Materials and Methods). Increased expression of the mitochondrial marker ATP synthase-a (red) is seen in Yki
overexpressing cells (green) as compared with the adjacent wild-type cells (nongreen). (J) Red channel alone is shown for clarity. Bar, 5 mm.
(K,L) Knockdown of wts function in the dorsal compartment of the wing disc (K; green, ap-Gal4 UAS-GFP, UAS-wtsRNAi) causes an
increase in mitochondrial marker ATP synthase-a staining (L; red). The white dotted line marks the dorsal–ventral boundary. Bar, 50 mm.
(M–P) Pupal eye discs. (M,N) Control, pupal eye disc from spa-Gal4, UAS-GFP background. The cone cells are marked with GFP (M; green)
and stained for mitochondrial marker ATP-syn (N; red). (O,P) Overexpression of Yki in cone cells (spa-Gal4, UAS-GFP, UAS-yki) (O;
green) causes a significant increase in mitochondrial marker ATP-syn staining (P; red). Bar, 5 mm. (Q–T) Clones of cells in the fat body
expressing Yki (shown in Q and S in green) using the Ay-Gal4 system (see the Materials and Methods). Increased staining with ATP
synthase-a antibody (R; red) and increased uptake of the mitochondrial membrane potential-sensitive dye MitoTracker Red (S; red) can be
seen in Yki overexpressing cells (S; green) at the perinuclear regions when compared with adjacent wild-type cells (S; nongreen). Bars, 5 mm.
Mitochondrial phenotypes in Hippo pathway mutant backgrounds. (A–L) Third instar wing discs. (A,B) Isolated mutant clones
Yorkie/YAP and mitochondrial function
GENES & DEVELOPMENT 2029
microscopic (EM) analysis in cultured human cells over-
expressing YAP2 (Fig. 3E,F) as well as in Drosophila
tissues in which Yki is activated (Fig. 3G–J). Consistent
with the light microscopic immunohistochemical re-
sults, at the EM level, the mitochondria are elongated
and enlarged (Fig. 3E–J), suggesting that the observed
phenotypes are due to mitochondrial fusion. The elon-
gated mitochondria continue to maintain normally struc-
tured cristae. Quantitation of the fusion phenotype
revealed an average twofold increase in the length of
mitochondria upon Yki/Sd activation (P = 0.0009) (Supple-
mental Fig S9A). Mitochondrial numbers are moderately
MDAMB453 and SUM159PT cells expressing empty vector using MitoTracker Red (A,C) and ATP-synthase-a antibody (B,D). Stable
expression of Flag-tagged YAP2 in MDAMB453 and SUM159PTcells causes an increase in mitochondria monitored using MitoTracker
Red (A9,C9) and ATP-synthase-a antibody (B9,D9). Bar, 5 mm. (E–J) Ultrastructural analysis of mitochondria (highlighted in red) upon
Yki/YAP2 overexpression. (E,F) Electron micrographs of mitochondria from human cells. YAP2 overexpressing MDAMT453 cells
(F) have elongated and enlarged mitochondria compared with their vector-transformed controls (E). Bar, 1 mm. (G,H) Electron micrographs
of mitochondria from Drosophila salivary glands. The mitochondria are elongated upon combined overexpression of Yki and Sd (H)
compared with wild-type control (G). Bar, 1 mm. (I,J) Electron micrographs of mitochondria from Drosophila pupal eye discs. High-
magnification view of mitochondria from control (I) and Yki-activated (J) pupal eye tissue. Bar, 0.5 mm. (K–M) Reduction of ROS in YAP2
overexpressing cells. Flow cytometric analysis of ROS-labeled control cells (red trace) and YAP2-overexpressing cells (blue trace) in
HS578T (K), SUM159PT (L), and MDAMB231 (M) cell lines. The ROS levels show a dramatic decrease of ;100-fold in all three cell lines
upon YAP2 expression.
Mitochondrial phenotypes of Yki/YAP2-activated human and Drosophila cells. (A–D) Controls. Mitochondrial staining in
Nagaraj et al.
2030 GENES & DEVELOPMENT
increased (52%) (Supplemental Fig. S9B), which could be
due to either increased biogenesis or decreased turnover,
which is reported to result from increased fusion (Twig
et al. 2008).
To assess the functional effects of Yki/YAP2 on mito-
chondria, we measured several metabolic outputs in
human and Drosophila cells. ATP levels are not signifi-
cantly altered upon Yki/YAP2 overexpression in either
Drosophila or human cells, and consumption of glucose,
oxygen, and glutamine and production of lactate and
glutamate were also unaffected in several human cell
lines (Supplemental Fig. S10). However, levels of reactive
oxygen species (ROS) show a dramatic, two orders of
magnitude decrease when YAP2 is overexpressed in the
three independent cell lines tested (Fig. 3K–M). These
results demonstrate that one consequence of the mam-
malian YAP2 pathway activation is a reduction of ROS,
which is likely to increase cellular resistance to oxida-
Mitochondrial targets of Yki activation
To probe mechanistic determinants of Yki function in
mitochondrial expansion, we conducted a genome-wide
microarray experiment and specifically compared expres-
sion patterns of mitochondria-related transcripts from
control (GMR-Gal4) and Yorkie overexpressing (GMR-
Gal4; UAS-yki) pupal eye discs. Genes that have pre-
viously been implicated in basal mitochondrial biogene-
sis, such as Spargel (PGC-1a) and DELG (NRF2) (Baltzer
et al. 2009), are not altered upon overexpression of Yki.
However, a number of other genes associated with mito-
chondrial function are up-regulated by Yki, including the
two mitochondrial fusion genes opa1-like (opa1) and mi-
tochondria assembly regulatory factor (Marf) (Supplemen-
tal Fig S11A; Chan 2006). These could potentially provide
a link between Yki activation and enhanced mitochondrial
fusion evident from the ultrastructural analysis.
To determine whether Yki and Sd are directly bound
to DNA corresponding to mitochondrial genes, we
used chromatin immunoprecipitation (ChIP) and whole-
genome tiling arrays (ChIP–chip) to identify regions
bound bythese factors in eye–antennal and wingimaginal
discs. We found that Yki and Sd together bind to enhancer
elements corresponding to several genes related to mito-
chondrial fusion, including opa1 and Marf (Fig. 4A,B;
Supplemental S11B). These genes include a consensus
motif (AGGAATGT) in their upstream sequence that
matches with the published Sd-binding consensus site.
Importantly, most of the Sd-bound regions around these
genes also contain strong matches to the Sd consensus
site that we defined using in vitro selection followed by
next-generation sequencing (Fig. 4C). Gel shift assays
established that the Sd consensus sites within opa1 and
Marf are indeed bound by Sd (Fig. 4D). Luciferase reporter
assays in S2 cells using the Sd-binding domain containing
enhancer regions of opa1 and Marf showed a 15-fold and
fivefold induction of opa1 and Marf, respectively, upon
activation with Yki/Sd (Supplemental Fig. S12). In situ
hybridization studies further show that in wts mutant
clones, opa1 expression is up-regulated (Fig. 4E–G). Based
on the genetic, microarray, and ChIP data and reporter
assays, we conclude that the Yki/Sd complex controls
expression of the mitochondrial fusion genes opa1 and
Marf that lead to the formation of the extensively fused
mitochondria observed upon Yki activation.
The ChIP–chip array showed that 261 mitochondrial
genes were bound by both Sd and Yki, of which 36 are also
up-regulated at least 1.2-fold in the microarray analysis
(Fig. 4H). We confirmed the up-regulation of opa1 and
Marf byquantitative PCR (Supplemental Fig. S13). The 36
genes include various mitochondrial enzymes and trans-
porters, including complex I genes, antioxidants, and the
two mitochondrial fusion genes opa1 and Marf, which
implies that the Yki/Sd complex directly binds to the
upstream regions of these genes to regulate their tran-
scription. In addition to these fusion proteins, the two
major classes of up-regulated proteins whose relevance
seems evident from our genetic analysis include the anti-
oxidant proteins and members of complex I of the electron
transport chain. Both of these protein classes have dem-
onstrated functions related to ROS levels (Finkel 2003;
Owusu-Ansah et al. 2008), and their up-regulation by the
Yki/YAP pathway helps explain the dramatic reduction in
ROS levels upon YAP2 activation (Fig. 3K–M) and suggests
a conserved function for Yki/YAP2 signaling in the mod-
ulation of oxidative stress. The best-characterized target of
the Yki/Sd pathway, Diap1, is also up-regulated at levels
similar to the mitochondrial genes (1.2-fold) in our tran-
opa1 and Marf suppress Yki growth and mitochondrial
The role of opa1 and Marf as downstream targets of Yki
was further dissected through genetic analysis. MARCM
clones of wts in wing discs in which opa1 function is
attenuated using UAS- opa1RNAicauses significant sup-
pression in the growth of mutant clones (Fig. 5A–E). Using
the dpp-Gal4 driver, we found that combined inactivation
of opa1 and Marf significantly suppresses the increased
tissue growth observed upon overexpression of Yki (Fig.
5F–L). This effect is Yki-specific, as overgrowth due to loss
of other tumor suppressor backgrounds is not suppressed
by opa1 (Supplemental Fig. S14).
The attenuation in Yki-mediated tissue growth can be
traced to a reduction in cell proliferation. In the larval
wing disc, knockdown of opa1 and Marf using RNAi
suppresses the enhanced EdU incorporation observed
upon Yki overexpression (Fig. 5F–K,M). Likewise, over-
expression of Yki causes increased proliferation of cells in
the third instar eye imaginal disc, and this phenotype is
significantly suppressed upon reduction of either opa1 or
Marf as measured by phospho-Histone H3 staining (Fig.
5N; additional controls in Supplemental Figs. S15, S16) in
the eye. We analyzed Caspase-3 staining in multiple
RNAi backgrounds for both opa1 and Marf used in this
study and did not detect an increase in cleaved Caspase-3
(Supplemental Fig. S17). The suppression of Yki-mediated
proliferation by the loss of fusion genes correlates with the
Yorkie/YAP and mitochondrial function
GENES & DEVELOPMENT 2031
mitochondrial morphogenesis genes. Sd and Yki ChIP–chip binding profiles at opa1 (A) and Marf (B). Regions designated as bound by
a given factor (5% FDR [false discovery rate]) are indicated by solid rectangles below a given peak. (C) Sd consensus motif identified by
in vitro selection and next-generation sequencing in mitochondrial fusion genes. (D) Gel mobility shift assay. Purified TEA domain of
Sd protein used for mobility shift assay shows binding to wild-type but not mutated consensus Sd-binding sites as indicated. opa1- and
Marf-specific DNA sequences containing wild-type but not mutated putative Sd-binding sites can compete and block Sd binding to the
probe. (E–G) Fluorescence in situ hybridization to visualize opa1 transcripts in wts mutant clones. wts mutant clones (nongreen,
marked by arrows in E) were generated in third instar eye discs using the Flp/FRT system. In situ hybridization for GFP (green) marks
the wild-type tissue. (F) opa1 expression (red) is up-regulated in the mutant (nongreen) tissue. The merged image is shown in G. Note
that the level of expression of opa1 in the wild-type eye disc is below the levels of detection. However, in these experiments, we were
able to detect wild-type opa1 expression in the wing disc pouch, ring glands, and eye disc upon overexpression of opa1 using the
combination GMR-Gal4, UAS-opa1. (H) List of categories of mitochondrial genes that are up-regulated with their fold change (judged
by microarray analysis) that are also bound by both Yki and Sd (based on ChIP–chip data set generated for Yki and Sd).
Direct regulation of mitochondria-related genes by the Yorkie/Scalloped complex. (A,B) Sd and Yki are bound near the loci of
wing imaginal discs. (A) Mock clones (green) generated with wild-type FRT82B chromosome. (B) Mock clones (green) expressing
opa1RNAi. (C) wts clones (green) showing an extensive overgrowth phenotype. (D) wts clones (green) expressing opa1RNAisuppress the
overgrowth phenotype seen upon loss of wts function (cf. C and D). (E) A quantitative analysis of clone sizes in different genotypes as
indicated at the bottom of the graph. The clone sizes in the wild type and opa1RNAiare similar with no significant differences, while
expression of opa1RNAiin wts clones results in a significant (P = 0.004) suppression of clone size. Bars, 50 mm. (F–K) Representative
images of third instar wing discs labeled with EdU to mark S-phase cells. dpp-Gal4, UAS-GFP is used to drive appropriate RNAi/ UAS-
Yki overexpression. Genotypes are indicated (RNAi constructs used are opa1KK105706and MarfKK105261). Bars, 50 mm. (L) Quantification
of the area of the dpp-Gal4, UAS-GFP-expressing tissue. RNAi targeting opa1, Marf, or both does not significantly alter growth in the
wing disc, as demonstrated by measuring the area of dpp-expressing tissue normalized to control (dpp-Gal4 UAS-GFP crossed to w1118,
opa1 RNAi, Marf RNAi, or opa1 RNAi/Marf RNAi). Overexpression of yorkie (dpp-Gal4 UAS-GFP, UAS-yki) causes a robust increase in
tissue size (P = 1.2 3 10?4), and this is significantly reduced by combined inactivation of opa1 and Marf (P = 0.037). Average sizes
relative to control and error bars indicating standard deviation are shown. (M) Quantification of EdU-positive cells within the dpp-Gal4,
UAS-GFP-expressing tissue. Genotypes are as indicated. Compared with control (dpp-Gal4, UAS-GFP), the reduction of opa1 or Marf
function by RNAi, separately or in combination, does not significantly reduce the number of EdU-labeled cells. Overexpression of
yorkie (dpp-Gal4, UAS-GFP, UAS-yki) causes a dramatic increase in EdU-positive cells (P = 0.008), and coexpression of opa1 and Marf
RNAis with yorkie (dpp-Gal4, UAS-GFP, UAS-yki, UAS-opa1 RNAi, UAS-Marf RNAi) leads to a significant suppression of EdU-
positive cell number (P = 0.023). Average number of EdU spots in the dpp(+) area is plotted relative to control, and error bars indicate
standard deviation. (N) Quantitative analysis of phospho-Histone H3 (P-H3)-positive cells upon Yki activation and in combination with
a reduction of opa1 and Marf function. Genotypes are as indicated. Overexpression of Yki in the third instar eye discs causes
a significant increase in the number of P-H3-positive cells (P = 2.7 3 10?6). Knockdown of opa1 alone, Marf alone, or both in a Yki
overexpression background significantly suppresses the number of P-H3-positive cells (P = 0.01, 0.001, and 8.9 3 10?5, respectively).
(O–Q) Suppression of Yki-induced mitochondrial phenotype upon loss of opa1 function. Mitochondrial morphology is monitored using
mitoGFP (green). Cell nuclei are marked with Topro (blue). Bar, 5 mm. (R–V) Suppression of Yki-induced mitochondrial phenotype upon
loss of opa1 and Marf function. Mitochondrial morphology is monitored using ATP synthase-a staining (green). Photoreceptor nuclei
are marked with ELAV (red). Bar, 5 mm. Combined reduction of opa1 and Marf in a Yki-activated background causes strong suppression
of mitochondrial expansion.
Mutations in mitochondrial fusion genes suppress phenotypes caused by Yki activation. (A–E) MARCM clones of wts in the
suppression in Yki-induced mitochondrial fusion observed
in the pupal eye disc (Fig. 5O–V; further controls in
Supplemental Fig. S18). The relevance of opa1-mediated
mitochondrial fusion to the Yki-mediated mitochondrial
phenotype is underscored by the dosage-dependent inter-
action between a single-copy loss of opa1 and Yki over-
expression (Supplemental Fig. S19).
Previous studies have shown that the Hippo pathway
functions in flies as well as vertebrates to control organ
size (Dong et al. 2007). Furthermore, mutations in com-
ponents of this pathway have been implicated in multiple
forms of cancer (Edgar 2006). This pathway has been
shown to directly promote cell proliferation and repress
apoptosis. Here, we show that the mitochondrion is an
important additional target of the Hippo pathway. An
increase in Yki activity causes an increase in mitochon-
drial fusion due to direct transcriptional activation of
major mitochondrial fusion genes. Increased mitochon-
drial fusion has been linked to regulation of the G1–S
checkpoint and Cyclin E activity (Mandel et al. 2005;
Mitra et al. 2009), and cells harboring fused mitochondria
are resistant to stress-induced apoptosis (Tondera et al.
2009; van der Bliek 2009). Interestingly, these are precisely
the phenotypes seen upon activation of the Hpo/Yki
pathway. It seems likely that this pathway independently
activates the cell cycle, represses apoptosis (Harvey et al.
2003; Hay and Guo 2003; Wu et al. 2003), and promotes
mitochondrial and metabolic changes described in this
study that together cause tumor growth. However, since
mitochondrial fusion plays a role in S-phase entry and
stress resistance (Mitra et al. 2009), it is attractive to
speculate that the mitochondrial effects could indirectly
affect Yki’s up-regulation of Cyclin E and DIAP. This is
supported by our observation that in hpo, opa1 double-
mutant clones, Cyclin E expression is suppressed when
compared with that seen in hpo mutant clones (Supple-
mental Fig. S21). The increase in fusion gene levels upon
Yki activation is modest. This may be important in pro-
ducing the observed phenotype of fused but functional
mitochondria. Gross overexpression of fusion genes leads
to abnormal and globular mitochondria (Supplemental
Fig. S20). Moreover, the modification of mitochondrial
structure by Yki is accompanied by an up-regulation of
antioxidant enzymes and subunits of complex I of the
electron transport chain and a dramatic reduction in
intracellular ROS. It has recently been demonstrated that
other oncogenes (Ras, Raf, and Myc) also reduce ROS, but
in our system, these oncogenes do not up-regulate mito-
chondrial fusion as seen with activation of the Yki
pathway, suggesting that different oncogenic pathways
could alter ROS by distinct mechanisms (DeNicola et al.
2011). The importance of mitochondrial fusion in Yki
signaling is further highlighted by our observation that
a reduction in mitochondrial fusion suppresses Yki-medi-
ated growth phenotypes. This observed link between the
Hpo/Yki pathway and mitochondrial fusion is of signifi-
cance to both normal development and cancer biology.
Our study reveals that increased mitochondrial fusion and
expansion by Yki is evolutionarily conserved and that this
function of the Hippo pathway is relevant for both normal
and patho–physiological situations.
Materials and methods
The following stocks were used: sd47M; UAS-sd; UAS-mitoGFP;
spa-Gal4; GMR-Gal4; ap-Gal4; Ay-Gal4; dpp-Gal4 (Bloomington
Stock Center); UAS-yki (D. Pan); Sd-GFP (A. Spradling); UAS-
opa1RNAi(KK105706 and HMS000349); UAS-MarfRNAi(GD40478,
KK105261, and JF01650) (multiple opa1 and Marf RNAis were
tested for adult eye size suppression and in other assays de-
scribed in Fig. 5); UAS-wtsRNAi(KK106174 and HMS00026);
UAS-sdRNAi(KK108264); UAS-avlRNAi(KK107264); UAS-scribRNAi
opa1 (J. Chung); and MARCM stocks (J. Martinez).
We used the following antibodies: anti-mouse ATP synthase-a
(1:100; Mitosciences), anti-rabbit Mn-SODII (1:100; Stressgen),
MitoTracker Red (Molecular Probes), anti-rat anti-phospho-his-
tone 3 (1:300), and anti-Yki (1:500; D. Pan). BrdU labeling was
done by incubating freshly dissected eye discs in 75 mg/mL BrdU
for 30 min, followed by fixation, washing in PBS, and mounting.
Antibody staining was performed as described previously (Rogge
et al. 1991).
Ay-Gal4 flip-out clones
Flip-out clones were generated in the eye and in the fat body.
hs-flp was used to flip out the stuffer cassette from act-
FRTyFRT-Gal4 to generate act Gal4, which expresses Gal4 in
all cells of the eye imaginal disc and fat body (Ito et al. 1997).
The respective crosses were maintained at 18°C. At mid-
second instar, a 10-min heat shock at 37°C was applied (Ito
et al. 1997). Tissues were dissected in the third instar, fixed,
and stained with appropriate antibodies or MitoTracker Red
EdU labeling and tissue growth measurements
Appropriate genetic crosses were allowed to lay eggs on grape
juice/agar plates, and larvae hatched over a 3-h period were
collected and transferred to normal food plates maintained at
29°C. Upon reaching the third instar, larvae were dissected in
room temperature Schneider’s Drosophila medium, imaginal
wing discs were incubated for 30 min in 10 mM EdU, and then
discs were fixed and stained according to standard protocol
(Click-iT EdU Alexa Fluor 555 imaging kit, Life Technologies).
For visualization of GFP, tissue was stained with a monoclonal
GFP antibody (Millipore) at a dilution of 1:100. To quantify the
number of EdU-positive cells, images were analyzed with ImageJ
software. Briefly, automatic thresholds were applied to green and
red channels, and the green channel was inverted and subtracted
from the red channel to select only EdU spots within the dpp-
expressing area. Watershed segmentation was then performed to
separate overlapping EdU spots, and particles 10 mm2in size or
greater were counted. To measure tissue sizes, ImageJ was used
to automatically threshold the fluorescent channel, and area was
Nagaraj et al.
2034GENES & DEVELOPMENT
MARCM clones for wts (third chromosome) and its suppression
by opa1RNAi(second chromosome) were done using egg collec-
tion in vials for 3 h. Larvae from the crosses were heat-shocked
for 45 min at 37°C at 48 h after egg laying. Mid-third instar larvae
were collected for dissection, and the discs were fixed and
mounted in VectaShield for confocal microscopy. The area under
the clone was measured using ImageJ software, and statistical
data analysis was performed in Excel.
In situ hybridization
wts mutant clones were generated using the Flp/FRT system by
a 30-min heat-shock pulse at 37°C during the mid-second instar.
Plasmids for the generation of GFP and Opa1 probes were
obtained from Kathy Ngo and the Drosophila Genomics Re-
source Center (DGRC), respectively. The opa1 probe was labeled
with digoxygenin-tagged alkaline phosphatase, and the GFP
probe was labeled with fluorescein-tagged HA. For tissue in situ
hybridization experiments, third instar larvae from the cross
were washed in DEPC-treated water, dissected in Schneider’s
medium, fixed in 3.7% paraformaldehyde in 100 mM PIPES
buffer for 45 min on ice, and subsequently dehydrated in ethanol
for 12 h. The following day, discs were subjected to Proteinase K
treatment (10 mg/mL) for 4 min, followed by washes in glycine
buffer (20 mg/mL). The discs were then incubated in prehybrid-
ization buffer (2 h) and then incubated in hybridization buffer for
2 h at 55°C. The probe was prepared in 200 mL of hybridization
buffer and heated at 90°C, followed by coolingfor 5 minon ice for
denaturation. The discs were incubated with probe in hybridiza-
tion buffer for 18–22 h at 55°C. The following day, the discs were
washed multiple times in PBTat 55°C. The signal was amplified
with fluorescent-labeled antibodies (fl-HA for GFP and Dig-AP
for opa1), and the discs were then mounted in VectaShield.
Images were captured using confocal microscopy.
Samples were imaged using a Bio-Rad Radiance 2000 confocal
with LaserSharp 2000 acquisition software and a Zeiss LSM700
confocal with Zen 2009 acquisition software. Fluorescent in-
tensity quantifications were analyzed by ImageJ software.
Eye discs (100 for each genotype) were dissected from GMR-Gal4
(control) and GMR-Gal4, UAS-yki genotypes ;40 h after pupar-
iation. Total RNA was isolated from the dissected discs (Trizol
method and Qiagen RNAeasy kit) and used to generate micro-
array probes that were hybridized to Drosophila genome 2 arrays
(Affymetrix). The GeneChip Operating system (Affymetrix) and
dCHIP program (Harvard University) were used to generate pair-
wise comparisons between the transcription profiles of control
and Yorkie overexpressing discs.
ChIP, ChIP–chip, and motif analysis
Wandering third instar larvae were dissected and fixed. Imaginal
discs were collected in PBS on ice. Discs were fixed with 1.8%
formaldehyde, and cross-linked chromatin was sonicated to an
average size of 500 base pairs (bp). Chromatin preparation and
ChIPs were performed as described (Estella et al. 2008). Both
rabbit anti-Yki (D. Pan) and rabbit anti-GFP (Abcam, ab290) were
used at final dilutions of 1:300. Immunoprecipitated DNA and
input DNA were amplified for array hybridization using the
GenomePlex WGA4 whole-genome amplification kit (Sigma).
The samples were then labeled according to Affymetrix pro-
tocols and hybridized on Affymetrix GeneChip Drosophila
Tiling 2.0R arrays. ChIP–chip for each factor and tissue was
performed in biological triplicate. All raw and processed ChIP–
chip data are publicly available through the Gene Expression
Omnibus (GEO; accession no. GSE26678).
Tiling array data were processed with MAT (model-based
analysis of tiling arrays), and peaks were called at a 5% FDR
(false discovery rate) using MAT. For a given tissue, peaks were
called as shared if the Yki and Sd peaks overlapped by 1 bp or
more. Target genes were called based on the transcription start
site nearest a given peak. At the 5% FDR threshold, 6713 Sd+Yki
peaks were called in the wing disc and 5170 Sd+Yki peaks were
called in the eye–antenna disc. Altogether, 5203 genes are called
as Sd+Yki targets in either the wing or eye–antenna. Sd+Yki
target gene GO (gene ontology) overlap probabilities were
calculated from the hypergeometric distribution based on the
number of FlyBase GO annotated Sd+Yki target genes in the
given category and the total number of Drosophila melanogaster
genes in the annotation category.
In vitro selection (or SELEX) was performed with GST-Sd
immobilized on glutathione agarose. The input SELEX library
sequence was 59-GTTCAGAGTTCTACAGTCCGACGATCTG
CTTCTGCTTG-39, where ‘‘N’’ represents A/G/T/C. Eluted DNA
was PCR-amplified and prepared for Illumina sequencing by three
rounds of PCR with the following primers: 59-CAAGCAGAA
GACGGCATACGA-39 and 59-AATGATACGGCGACCACCGA
CAGGTTCAGAGTTCTACAGTCCGA-39. Illumina sequencing
(once for 36 cycles) was performed according to the manufac-
turer’s protocol for small RNA sequencing, as this is optimized
for sequencing short DNA fragments. The Sd-binding motif
was constructed from the top 300 16-mers selected after two
rounds of in vitro selection with full-length GST-tagged Sd. A
position weight matrix (PWM) was generated using MEME
(Bailey and Elkan 1994) and the following parameters: mini-
mum motif width of 6, maximum motif width of 8, one per
Gel shift assays were performed essentially as described pre-
viously (Estella et al. 2008), except the oligonucleotide probe was
dual-labeled with 59 IRDye 700 (Integrated DNA Technologies),
and gels were scanned using the Odyssey infrared imaging
system (Li-Cor Biosciences). The TEA domain of Sd, described
previously (Halder and Carroll 2001), was expressed in BL21 and
purified using Ni-NTA agarose (Qiagen). The following probes
were used: Consensus, 59-TTCGATACACTTGTGGAATGTGT
TTGATTTGTTAGCCCCG-39; Consensus-mutated, 59-TTCGA
GTA-39; Opa1-mutated, 59-GTTTCTTTTTAATATTcatAATaaA
CTTTGCAAAGAAGTA-39; Marf, 59-AATCATTGTCAGGGTA
AATTCTATTTTTAAATTAGACAAA-39; and Marf-mutated, 59-
MitoTracker and antibody staining of human cells
Flag-tagged YAP2 was cloned into the pLJM1 lentiviral vector,
and breast cancer cells stably expressing Flag-YAP2 were se-
lected with puromycin. SUM159PT and MDAMB453 cells as
well as culture conditions are described previously (Neve et al.
2006). For MitoTracker staining, human cells plated on cover-
slips were treated with 250 nm of MitoTracker Red (Invitrogen)
in culture medium for 30 min at 37°C. The cells were fixed,
washed, and mounted in VectaShield (Vector Laboratories) for
Yorkie/YAP and mitochondrial function
GENES & DEVELOPMENT2035
For antibody staining, cells were fixed in 4% paraformalde-
hyde for 5 min, washed for 2 min (PBS + 0.05% Triton X-100),
permeabilized with PBS + 0.1% TX-100 PBS for 5 min, and
washed twice for 5 min each. The cells were blocked with 10%
normal goat serum for 30 min, followed by two washes for 5 min
each. Cells were then incubated with primary antibody for 1 h at
room temperature, followed by two 5-min washes and incuba-
tion with the secondary antibody for 1 h at room temperature.
Finally, the cells were washed three times for 5 min each and
mounted on glass slides with VectaShield (Vector Laboratories).
TO-PRO-3 (Invitrogen) was applied in the second wash.
Drosophila imaginal discs from 50 larvae expressing Yorkie with
a T80-Gal4 driver were dissected and collected in lysis buffer,
and ATP was extracted according to the manufacturer’s instruc-
tions using an ATP Bioluminescence assay kit HS II (Roche). ATP
was measured by using a luminometer, and the values were
normalized with protein amounts measured from the same lysed
extracts using a Pierce BCA protein assay kit (Thermo Scien-
tific). ATP measurements from human cells were carried out
using the same procedure.
Cells (106) of empty vector or YAP2 transfected cells were grown
in DMEM supplemented with 10% FBS and 1% penicillin/
streptomycin, and the medium was replaced 24 h prior to
analysis. Cell numbers were counted using a Beckman Coulter
Z1 particle counter, and glucose, lactate, glutamine, and gluta-
mate concentrations were simultaneously analyzed using
a NOVA Basic 4 Bioanalyzer. Metabolite consumption/produc-
tion was expressed as moles per 106cells per hour.
Drosophila tissues were dissected in PBS and fixed in 2%
glutaraldehyde and 2.5% formaldehyde in PBS, and human cells
were fixed in 2% glutaraldehyde in PBS. Embedding, thin
sectioning, and staining were carried out according to standard
protocol (Afzelius and Maunsbach 2004). The sections were
stained with uranyl acetate and lead citrate and examined on
a JEOL 100CX electron microscope at 80 kV.
The ROS dye (CM-H2DCFDA, Invitrogen) was dissolved in
DMSO and added to serum-containing medium to a final con-
centration of 5 mM. Cells were incubated with the dye-contain-
ing medium for 20 min at 37°C. Cells were rinsed twice with
PBS, trypsinized, and recovered in full medium followed by a
5-min spin down at 200g. Cells were resuspended in 400 mL of PBS
and used for flow cytometric analysis using the FITC channel.
Putative enhancers upstream of opa1-like (chromosome 2R,
10123270–10123692) and Marf (chromosome X, 6259520–
6260412) were inserted between the MluI and XhoI sites of the
pGL2 vector to generate luciferase reporter constructs. The
expanded and 3xSd reporter constructs, also pGL-based, and
pIE7-Flag-Sd and pAc-HA-Yki expression constructs were de-
scribed previously (Nicolay et al. 2011). S2 cells were transfected
with an individual reporter plasmid, Sd and/or Yki expression
plasmids, and a Renilla luciferase plasmid (pAC-Rluc) to control
for transfection efficiency. Transfections were performed with
Effectene transfection reagent (Qiagen), and reporter assays were
carried out in triplicate 48 h post-transfection. Reporter assays
were performed using the Dual-Glo Luciferase assay system
We thank K. Ngo, D. Pan, I. Hariharan, A. Spradling, J. Chung, S.
Cohen, and the stock centers of Bloomington, the National
Institute of Genetics (Tokyo), and the Vienna Drosophila RNAi
Center (Vienna) for providing Drosophila stocks and reagents.
We thank H. Richardson, H. McNeill, and T. Orr-Weaver for
Cyclin E antibody. We gratefully acknowledge Xiaoyue Wang’s
help in the luciferase reporter assays. We thank and acknowledge
Derek Cheung’s contribution in analyzing data. We thank W.
Freije for help with microarray analysis, and Genaro Villa for
help with creating YAP2 constructs. We acknowledge the help
from the microarray core facility in the Department of Pathology
at the University of California at Los Angeles. We also thank
Sirus Kohan and the Brain Research Institute EM facility for
assistance with electron microscopy. K.T.J. is supported by a
post-doctoral fellowship (#PF-10-130-01-DDC) from the American
Cancer Society. This work is supported by National Institutes of
Health Grant RO1EY008152 to U.B.
Afzelius BA, Maunsbach AB. 2004. Biological ultrastructure
research; the first 50 years. Tissue Cell 36: 83–94.
Bailey TL, Elkan C. 1994. Fitting a mixture model by expecta-
tion maximization to discover motifs in biopolymers. Proc
Int Conf Intell Syst Mol Biol 2: 28–36.
Baltzer C, Tiefenbock SK, Marti M, Frei C. 2009. Nutrition
controls mitochondrial biogenesis in the Drosophila adipose
tissue through Delg and cyclin D/Cdk4. PLoS ONE 4: e6935.
Bennett FC, Harvey KF. 2006. Fat cadherin modulates organ size
in Drosophila via the Salvador/Warts/Hippo signaling path-
way. Curr Biol 16: 2101–2110.
Chan DC. 2006. Mitochondrial fusion and fission in mammals.
Annu Rev Cell Dev Biol 22: 79–99.
Clark IE, Dodson MW, Jiang C, Cao JH, Huh JR, Seol JH, Yoo SJ,
Hay BA, Guo M. 2006. Drosophila pink1 is required for
mitochondrial function and interacts genetically with parkin.
Nature 441: 1162–1166.
Deng H, Dodson MW, Huang H, Guo M. 2008. The Parkinson’s
disease genes pink1 and parkin promote mitochondrial
fission and/or inhibit fusion in Drosophila. Proc Natl Acad
Sci 105: 14503–14508.
DeNicola GM, Karreth FA, Humpton TJ, Gopinathan A, Wei C,
Frese K, Mangal D, Yu KH, Yeo CJ, Calhoun ES, et al. 2011.
Oncogene-induced Nrf2 transcription promotes ROS de-
toxification and tumorigenesis. Nature 475: 106–109.
Dong J, Feldmann G, Huang J, Wu S, Zhang N, Comerford SA,
Gayyed MF, Anders RA, Maitra A, Pan D. 2007. Elucidation
of a universal size-control mechanism in Drosophila and
mammals. Cell 130: 1120–1133.
Edgar BA. 2006. From cell structure to transcription: Hippo
forges a new path. Cell 124: 267–273.
Estella C, McKay DJ, Mann RS. 2008. Molecular integration of
wingless, decapentaplegic, and autoregulatory inputs into
Distalless during Drosophila leg development. Dev Cell 14:
Finkel T. 2003. Oxidant signals and oxidative stress. Curr Opin
Cell Biol 15: 247–254.
Nagaraj et al.
2036GENES & DEVELOPMENT
Goulev Y, Fauny JD, Gonzalez-Marti B, Flagiello D, Silber J, Download full-text
Zider A. 2008. SCALLOPED interacts with YORKIE, the
nuclear effector of the hippo tumor-suppressor pathway in
Drosophila. Curr Biol 18: 435–441.
Goyal G, Fell B, Sarin A, Youle RJ, Sriram V. 2007. Role of
mitochondrial remodeling in programmed cell death in
Drosophila melanogaster. Dev Cell 12: 807–816.
Halder G, Carroll SB. 2001. Binding of the Vestigial co-factor
switches the DNA-target selectivity of the Scalloped selector
protein. Development 128: 3295–3305.
Hariharan IK. 2006. Growth regulation: A beginning for the
hippo pathway. Curr Biol 16: R1037–R1039. doi: 10.1016/
Harvey KF, Pfleger CM, Hariharan IK. 2003. The Drosophila Mst
ortholog, hippo, restricts growth and cell proliferation and
promotes apoptosis. Cell 114: 457–467.
Hay BA, Guo M. 2003. Coupling cell growth, proliferation, and
death. Hippo weighs in. Dev Cell 5: 361–363.
Huang J, Wu S, Barrera J, Matthews K, Pan D. 2005. The Hippo
signaling pathway coordinately regulates cell proliferation
and apoptosis by inactivating Yorkie, the Drosophila homo-
log of YAP. Cell 122: 421–434.
Ito K, Awano W, Suzuki K, Hiromi Y, Yamamoto D. 1997. The
Drosophila mushroom body is a quadruple structure of
clonal units each of which contains a virtually identical set
of neurones and glial cells. Development 124: 761–771.
Justice RW, Zilian O, Woods DF, Noll M, Bryant PJ. 1995. The
Drosophila tumor suppressor gene warts encodes a homolog
of human myotonic dystrophy kinase and is required for
the control of cell shape and proliferation. Genes Dev 9:
Mandal S, Guptan P, Owusu-Ansah E, Banerjee U. 2005.
Mitochondrial regulation of cell cycle progression during
development as revealed by the tenured mutation in Dro-
sophila. Dev Cell 9: 843–854.
Mitra K, Wunder C, Roysam B, Lin G, Lippincott-Schwartz J.
2009. A hyperfused mitochondrial state achieved at G1–S
regulates cyclin E buildup and entry into S phase. Proc Natl
Acad Sci 106: 11960–11965.
Neve RM, Chin K, Fridlyand J, Yeh J, Baehner FL, Fevr T, Clark
L, Bayani N, Coppe JP, Tong F, et al. 2006. A collection of
breast cancer cell lines for the study of functionally distinct
cancer subtypes. Cancer Cell 10: 515–527.
Nicolay BN, Bayarmagnai B, Islam AB, Lopez-Bigas N, Frolov
MV. 2011. Cooperation between dE2F1 and Yki/Sd defines
a distinct transcriptional program necessary to bypass cell
cycle exit. Genes Dev 25: 323–335.
Nolo R, Morrison CM, Tao C, Zhang X, Halder G. 2006. The
bantam microRNA is a target of the hippo tumor-suppressor
pathway. Curr Biol 16: 1895–1904.
Oh H, Irvine KD. 2008. In vivo regulation of Yorkie phosphor-
ylation and localization. Development 135: 1081–1088.
Owusu-Ansah E, Yavari A, Mandal S, Banerjee U. 2008. Distinct
mitochondrial retrograde signals control the G1–S cell cycle
checkpoint. Nat Genet 40: 356–361.
Pan D. 2007. Hippo signaling in organ size control. Genes Dev
Peng HW, Slattery M, Mann RS. 2009. Transcription factor
choice in the Hippo signaling pathway: homothorax and
yorkie regulation of the microRNA bantam in the progenitor
domain of the Drosophila eye imaginal disc. Genes Dev 23:
Rogge RD, Karlovich CA, Banerjee U. 1991. Genetic dissection
of a neurodevelopmental pathway: Son of sevenless func-
tions downstream of the sevenless and EGF receptor tyrosine
kinases. Cell 64: 39–48.
Silva E, Tsatskis Y, Gardano L, Tapon N, McNeill H. 2006. The
tumor-suppressor gene fat controls tissue growth upstream
of expanded in the hippo signaling pathway. Curr Biol 16:
Thompson BJ, Cohen SM. 2006. The Hippo pathway regulates
the bantam microRNA to control cell proliferation and
apoptosis in Drosophila. Cell 126: 767–774.
Tondera D, Grandemange S, Jourdain A, Karbowski M,
Mattenberger Y, Herzig S, Da Cruz S, Clerc P, Raschke I,
Merkwirth C, et al. 2009. SLP-2 is required for stress-induced
mitochondrial hyperfusion. EMBO J 28: 1589–1600.
Twig G, Elorza A, Molina AJ, Mohamed H, Wikstrom JD, Walzer
G, Stiles L, Haigh SE, Katz S, Las G, et al. 2008. Fission and
selective fusion govern mitochondrial segregation and elim-
ination by autophagy. EMBO J 27: 433–446.
van der Bliek AM. 2009. Fussy mitochondria fuse in response to
stress. EMBO J 28: 1533–1534.
Wu S, Huang J, Dong J, Pan D. 2003. hippo encodes a Ste-20
family protein kinase that restricts cell proliferation and
promotes apoptosis in conjunction with salvador and warts.
Cell 114: 445–456.
Wu S, Liu Y, Zheng Y, Dong J, Pan D. 2008. The TEAD/TEF
family protein Scalloped mediates transcriptional output of
the Hippo growth-regulatory pathway. Dev Cell 14: 388–398.
Zhang L, Ren F, Zhang Q, Chen Y, Wang B, Jiang J. 2008. The
TEAD/TEF family of transcription factor Scalloped mediates
Hippo signaling in organ size control. Dev Cell 14: 377–387.
Zhang L, Yue T, Jiang J. 2009. Hippo signaling pathway and
organ size control. Fly (Austin) 3: 68–73.
Zhao B, Lei QY, Guan KL. 2008. The Hippo-YAP pathway: New
connections between regulation of organ size and cancer.
Curr Opin Cell Biol 20: 638–646.
Yorkie/YAP and mitochondrial function
GENES & DEVELOPMENT2037