Developmental Cell
Article
Alternative Germ Cell Death Pathway
in Drosophila Involves HtrA2/Omi,
Lysosomes, and a Caspase-9 Counterpart
Keren Yacobi-Sharon,1 Yuval Namdar,1 and Eli Arama1,*
1Department of Molecular Genetics, Weizmann Institute of Science, Rehovot 76100, Israel
*Correspondence: [email protected]
http://dx.doi.org/10.1016/j.devcel.2013.02.002
SUMMARY
In both flies and mammals, almost one-third of the
newly emerging male germ cells are spontaneously
eliminated before entering meiosis. Here, we show
that in Drosophila, germ cell death (GCD) involves
the initiator caspase Dronc independently of the
apoptosome and the main executioner caspases.
Electron microscopy of dying germ cells revealed
mixed morphologies of apoptosis and necrosis. We
further show that the lysosomes and their catabolic
enzymes, but not macroautophagy, are involved in
the execution of GCD. We then identified, in a screen,
the Parkinson’s disease-associated mitochondrial
protease, HtrA2/Omi, as an important mediator of
GCD, acting mainly through its catalytic activity
rather than by antagonizing inhibitor of apoptosis
proteins. Concomitantly, other mitochondrial-associated factors were also implicated in GCD, including
Pink1 (but not Parkin), the Bcl-2-related proteins, and
endonuclease G, which establish the mitochondria
as central mediators of GCD. These findings uncover
an alternative developmental cell death pathway in
metazoans.
INTRODUCTION
Virtually all metazoan cells have the ability to commit suicide by
activating a genetic program called programmed cell death
(PCD) (Metzstein et al., 1998; Tittel and Steller, 2000). PCD
functions to eliminate unwanted or potentially dangerous cells
during the development and homeostasis of the organism, and
the malfunction of this process is associated with the pathogenesis of a variety of diseases, including cancer and neurodegenerative disorders (White, 2006; Bredesen, 2008; Fuchs and
Steller, 2011). Apoptosis, also known as type I cell death, is the
most abundant and investigated form of PCD, characterized
by a conserved sequence of cytological and morphological
events, including cell detachment and shrinkage, nuclear
condensation and segmentation, intact organelles, membrane
blebbing, and disassembly into apoptotic bodies that are often
engulfed by other cells. Biochemically, apoptosis is manifested
by the activation of a unique family of cysteine proteases called
caspases (Degterev et al., 2003). Caspases play a key role in the
signaling and execution of apoptosis, and their activation is
tightly controlled by activating and inhibitory proteins (Budihardjo et al., 1999; Salvesen and Dixit, 1999; Bader and Steller,
2009). Activation of caspase-9, the initiator caspase of the
intrinsic apoptotic pathway, involves its recruitment to the apoptosome by Apaf-1-Cyt c complex, whereas the apical caspase
of the extrinsic apoptotic pathway, caspase-8, is activated
within the death-inducing signaling complex (DISC) (Zou et al.,
1997; Ashkenazi and Dixit, 1998; Varfolomeev et al., 1998; Rodriguez and Lazebnik, 1999). The full activation of caspases often
involves the inactivation of a conserved protein family of endogenous caspase inhibitors called the inhibitor of apoptosis
proteins (IAPs) (O’Riordan et al., 2008; Bader and Steller,
2009). Active initiator caspases cleave and activate effector caspases (Riedl and Shi, 2004), which in turn orchestrate cell death
through the proteolytic cleavage of hundreds of cellular proteins
(Lüthi and Martin, 2007).
Despite their established role in apoptosis, the activation of
effector caspases does not always lead to apoptosis, as some
cells utilize restricted levels of caspase activity to promote a
variety of nonapoptotic vital cellular processes (Feinstein-Rotkopf and Arama, 2009). Furthermore, there is emerging evidence
from different experimental systems suggesting that some cells
can undergo cell death in the absence of active caspases (Yuan
and Kroemer, 2010). In addition to apoptosis, two other types
of cell death pathways have been defined based on cytological, morphological, and biochemical/molecular characteristics
(Edinger and Thompson, 2004). The type II autophagic cell death
is mainly characterized by the accumulation of autophagyrelated double-membrane-enclosed vesicles, which function
to self-digest the cell by sequestering cellular materials to
lysosomes, and many of the key components of the classical
macroautophagy pathway are also important in this cell death
pathway. Type III cell death, or necrosis, is defined by early
plasma membrane rupture, dilatation of cytoplasmic organelles
(in particular mitochondria), and in some cells, an increase in
cytoplasmic calcium, reactive oxygen species (ROS) accumulation, and/or lysosomal membrane permeabilization (LMP), and
the concomitant actions of some lysosomal hydrolases in the
cytoplasm. In addition, a specific necrotic pathway stimulated
by the Fas/TNFR family of death-domain receptors signals the
formation of an alternative multiprotein complex, which includes
the death domain-containing kinase, RIP1, its interactor kinase,
RIP3, and the further downstream kinase, MLKL (Degterev et al.,
2008; He et al., 2009; Sun et al., 2012). Cell death pathways
Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc. 29
Developmental Cell
Omi and Lysosomes Mediate Alternative Cell Death
exhibiting mixed morphologies were also reported, and recent
analyses of molecular networks may point to some crosstalk
between the different pathways (Rubinstein et al., 2011).
Nonapoptotic cell death pathways have received much attention recently because of their potential roles in the pathology of
diseases and the promise for possible new therapies (Galluzzi
et al., 2011). Despite recent growing interest in unraveling the
components and mechanisms underlying alternative death
pathways, progress in this area has been relatively slow. One
reason for this is that ‘‘clean’’ alternative cell death pathways
are often manifested only following nonphysiological conditions/manipulations in certain cell lines, and thus their relevance
in normal development and homeostasis and in the pathology
of diseases remained largely unknown. Although fascinating,
thus far, only very few developmental paradigms of alternative
cell death pathways have been described in organisms with
conventional apoptotic caspases, such as Caenorhabditis
elegans and Drosophila; the connections, if any, between these
paradigms and the nonphysiological cell death pathways are
still unclear (Berry and Baehrecke, 2007; Denton et al., 2009;
Blum et al., 2012).
Here, we show that during normal spermatogenesis in the
adult Drosophila testis, between 20%–30% of the spermatogonial cysts undergo spontaneous cell death, dubbed germ cell
death (GCD). Immunofluorescence and genetic analyses of
core components in the conventional apoptotic machinery
reveal that GCD can proceed in the absence of the apoptosome
and the main effector caspases, although it involves an apoptosome-independent function of the initiator caspase. Ultrastructural analysis demonstrates distinct morphological features of
GCD reminiscent of both apoptosis and necrosis. Examinations
of other alternative cell death pathway promotors indicate no
role for macroautophagy in GCD, whereas the lysosomes and
their catabolic enzymes are active and involved in this cell death
pathway, which is also associated with ROS accumulation and
cellular acidification. In a screen for mutants that dominantly
attenuate GCD, we identified the fly ortholog of the Parkinson’s
disease-associated mitochondrial protease, HtrA2/Omi. Functional studies imply that it is the catalytic activity of this protease
and not the IAP-antagonizing function that is important for
GCD and male fertility. Epistasis analyses place HtrA2/Omi
upstream and in parallel to the lysosomal pathway. Finally, the
mitochondrial-associated factors, Pink1, the Bcl-2-related proteins, and endonuclease G were all implicated in GCD, revealing
a central role of the mitochondria in mediating this alternative
cell death pathway.
RESULTS
Premeiotic Male Germ Cells Undergo Spontaneous Cell
Death in Drosophila
In the course of our studies of late spermatogenesis in the adult
Drosophila testis, we noticed that in nearly all adult testes, germ
cells in some premeiotic spermatogonial cysts spontaneously
and synchronously assume disintegrated morphology suggestive of cell death (the apical tip of a Drosophila testis is illustrated
in Figure 1A). To start characterizing this phenomenon, testes
were first stained with DAPI, to visualize the nuclei, and terminal
deoxynucleotidyl transferase-mediated dUTP nick-end labeling
30 Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc.
(TUNEL), which labels fragmented DNA in dying cells. Indeed,
dying spermatogonial cysts could be identified by virtue of their
rounded and packed nuclei and abundant TUNEL labeling (Figure 1B, insets). In more advanced stages of the cellular demolition, the DAPI signal disappeared, while the TUNEL signal
further increased (Figure 1C, insets). Subsequent analyses of
numerous testes from 1- to 7-day-old flies indicate an average
of 1.65 ± 0.05 (mean ± SE, n = 600) dying spermatogonial cysts
per testis. Most of this germ cell death (GCD) occurs at the
eight-cell stage, albeit a low percentage of dying 4- and 16-cell
spermatogonial cysts were also recorded (Figures 1B and 1C;
data not shown). Finally, because the adult testis normally
contains 5–8 spermatogonial cysts at each developmental
stage (Lindsley and Tokuyasu, 1980), we estimate that between
20%–30% of the newly emerging spermatogonial cysts undergo
spontaneous cell death in the adult testis.
GCD Is Independent of Effector Caspases
A key feature of apoptotic cell death is the activation of
caspases, a process commonly visualized in situ using the
anti-cleaved caspase-3 antibody, which specifically detects
the cleaved active forms of the main effector caspases in
Drosophila, Drice and Dcp-1 (Fan and Bergmann, 2010; Florentin and Arama, 2012). However, in contrast to the pronounced
activation of caspases in terminally differentiating (individualizing) spermatids, no caspase activation was detected at
earlier stages and, in particular, during GCD (Figure 1D). Conversely, marked expression of activated caspases was detected
in ectopically dying spermatogonial cells following targeted
knockdown of the major fly IAP, Diap1 (Figure 1E; Figure S1A
available online). Collectively, these findings indicate no association of spontaneous GCD with detectable levels of activated
caspases.
To directly test this hypothesis, we first examined the effect
of complete loss-of-function of Drice (driceD1) and Dcp-1 (dcp1Prev) on GCD levels. Significantly, not only were the levels of
GCD not reduced in these mutants, a mild increase was recorded in both mutants (Figures 2A–2C). We then tested for
a possible additive effect by specifically expressing the paneffector caspase inhibitor protein from baculovirus, p35, and
the Drosophila caspase inhibitor, Diap1, in spermatogonial cells.
Indeed, a marked, 50%–80%, increase in GCD levels was detected when effector caspase activity was globally inhibited
(Figures 2D–2G and S1B). Therefore, GCD can proceed in the
absence of effector caspase activity, which in turn acts to inhibit
this cell death pathway.
The Initiator Caspase Dronc Promotes GCD in an
Apoptosome-Independent Manner
The main activator of the effector caspases during Drosophila
apoptosis is the initiator caspase-9 ortholog, Dronc (Chew
et al., 2004; Xu et al., 2005). As in vertebrates, apoptosome
assembly and Dronc activation is usually triggered by binding
of pro-Dronc to the Apaf-1-like adaptor protein, Ark, and in
accordance, inactivation of these proteins lead to almost identical phenotypes (Xu et al., 2005; Srivastava et al., 2007). To
test the effect of the apoptosome on GCD, we examined
several independent loss-of-function alleles of dronc and ark.
Unexpectedly, however, whereas a marked increase in GCD
Developmental Cell
Omi and Lysosomes Mediate Alternative Cell Death
Figure 1. Spontaneous GCD Is Associated with DNA Fragmentation but Not Effector Caspase Activation
(A) Schematic representation of the Drosophila testis apex. At the very tip of the testis are the nondividing somatic hub cells (orange), which constitute a niche for
the germline stem cells (GSC; blue). Asymmetric division of a GSC yields a stem cell and a gonialblast (dark purple). The latter then undergoes four mitotic
divisions with incomplete cytokinesis producing interconnected 2-, 4-, 8-, and 16-cell spermatogonial bundles called cysts (light purple), which, after an extensive
growth phase, enter meiosis as primary spermatocytes. Germ cell cysts remain enveloped by a pair of somatic ‘‘cyst cells’’ (light blue), originating from cyst
progenitor cells (CPC; green), throughout spermatogenesis. Note the illustrated dying eight-cell spermatogonial cyst (red).
(B and C) Apical tips of wild-type testes stained to visualize the nuclei (DAPI; blue) and fragmented DNA (TUNEL; red). The insets present channel separations and
enlargements of the squared areas.
(D) A wild-type testis labeled as in (B) and additionally stained to visualize activated effector caspases (cCasp.3; green). IS, individualizing spermatids; WBs,
waste bags.
(E) A testis expressing a diap1 RNAi transgene in spermatogonial cells by the bam-Gal4 driver line was labeled as in (D). Dying cysts are either active caspasepositive but TUNEL-negative (early cell death stage; arrows) or TUNEL-positive with a faint active caspase signal (late stage; white arrowheads). A spontaneously
dying spermatogonial cyst in a region excluded from the bam-Gal4 expression domain is indicated by a yellow arrowhead.
Scale bars, 20 mm. See also Figure S1.
levels was recorded in the ark mutants, the dronc mutants displayed a significant 40%–60% decrease in GCD levels (Figures
2H–2J). There was no effect on GCD levels in mutants of the
two other initiator-like caspases, Strica and Dredd (Figure 2J;
data not shown). Altogether, these results suggest that Dronc
promotes GCD in an apoptosome-independent manner.
GCD Displays Mixed Morphologies
Apoptosis has been originally characterized by virtue of its
unique cellular morphologies (Wyllie et al., 1980). To explore
the morphological traits of spontaneous GCD, we used transmission electron microscopy (TEM) to look at ultrathin sections
of the apical tips of testes. These traits were subsequently
contrasted with morphological features of apoptotic spermatogonia triggered in the diap1 RNAi-expressing flies (Figure 1E).
Spontaneously dying spermatogonial cysts could be distin-
guished from their healthy counterparts based on their unusually
dense and disintegrated morphologies (red and white arrows,
Figures 3A, 3D, and 3G). In addition to significant cellular
shrinkage, these cells also exhibited chromatin condensation
and chromosomal DNA fragmentation, both features that are
reminiscent of apoptosis (a purple asterisk in Figure 3B and
green arrowheads in Figure 3C, respectively), and which were
also associated with apoptotic spermatogonia (Figure S2C0 ).
Unlike apoptosis, however, the nuclei in spontaneously dying
spermatogonia remained largely intact and were enclosed by
membranes even during late degradative stages (Figures 3B,
3C, and 3E). Of note, the crenellated morphology of the nuclei
is not unique to GCD, as healthy spermatogonial cells normally
contain unusually crenellated nuclei at these stages. Nonetheless, apoptotic spermatogonia exhibited increased nuclear
crenellation and were sometimes also mildly fragmented
Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc. 31
Developmental Cell
Omi and Lysosomes Mediate Alternative Cell Death
Figure 2. GCD Proceeds in the Absence of the Effector Caspases and the Apoptosome but Involves Dronc
(A and B) Testes from drice and dcp-1 homozygous mutants, respectively, were stained as in Figure 1B. Scale bars, 20 mm.
(C) Quantification of GCD levels in the testes described in (A) and (B). The presented values are relative to wild-type. The number of counted testes (n) is indicated
within each column. Error bars represent standard errors. p values of the level differences between wild-type and mutants are indicated above the bars. NS,
nonsignificant.
(D–F) Testes ectopically expressing either (D) GFP, (E) p35, or (F) Diap1 in the spermatogonial cells (using the nos-Gal4 driver; see also Figure S1B) were stained
as in Figure 1B.
(G) Quantification of GCD levels in the testes described in (D)–(F). All calculations were performed as in (C).
(H and I) Testes from representative mutant allelic combinations of dronc and ark, respectively, were stained as in (A).
(J) Quantification of GCD levels in mutant testes of the indicated phenotypes. All calculations were performed as in (C).
(Figure S2B). In addition, in both cell death types, only mild
nuclear condensation was recorded, which may be secondary
to the global condensation of these dying cells (the circumference of a dying cell nucleus and the circumference of its healthy
neighboring counterpart are indicated in red and white, respectively, Figure 3E; Figure S2C0 ). As opposed to apoptosis, where
organelles remain largely intact, spontaneously dying spermatogonia, but not apoptotic spermatogonia, exhibited deformed
mitochondria with abnormally swollen cristae (yellow asterisks
in Figures 3F and 3I; compare with Figures S2A0 , S2B0 , and
S2D0 ). Also, degradative foci accompanied by clear vesicles
and/or disorganized membranes and other materials, but not
autophagosome-like double-membrane vesicles, were often
detected during GCD, indicating extensive catabolic activity
32 Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc.
(blue and yellow arrowheads, respectively, Figures 3B and 3H).
On the other hand, other typical apoptotic features, including
significant cell detachment, membrane blebbing, and disassembly into apoptotic bodies were all readily detected in the
apoptotic spermatogonia, but not in spontaneously dying spermatogonia (Figures S2D, S2D0 , and S2D00 , receptively). We
conclude that GCD is manifested by coexistence of mixed
morphologies, some of which are reminiscent of apoptosis and
necrosis, but not autophagic cell death.
We occasionally detected spermatogonial cysts at advanced
GCD degradative stages that were surrounded by common
cellular material, implying that as opposed to the piecemeal
clearance of apoptotic bodies, dying spermatogonial cysts
are engulfed as single entities by other cells (Figure S2E).
Developmental Cell
Omi and Lysosomes Mediate Alternative Cell Death
Figure 3. TEM Analysis of Spontaneously
Dying Spermatogonial Cysts
Electron micrographs of cross-sections through
the apical tip of the Drosophila adult testes. Areas
confined by white squares and marked by white
letters were magnified and presented in separated
micrographs with corresponding letters.
Scale bars, 5 mm (A, D, and G); 0.5 mm (B, F, and H);
1 mm (C); 2 mm (E); and 0.2 mm (I). See also
Figure S2.
Furthermore, TUNEL labeling of testes expressing GFP under
the control of the phagocyte-specific driver, croquemort (crq)Gal4 (Franc et al., 1996), revealed some unusually large and
often amorphic GFP-expressing cells, which contain several
separated TUNEL-positive cellular entities (arrows in Figures
S2F and S2G). Therefore, phagocytes residing in the testis
appear to clear spontaneously dying spermatogonial cysts.
GCD Is Associated with High Lysosomal Activity,
ROS Accumulation, and Cellular Acidification,
but Not Macroautophagy
Macroautophagy has been suggested to play a role in developmental cell death of certain Drosophila cells (Ryoo and Baehrecke, 2010). To investigate whether the autophagic machinery
may also be involved in GCD, we first examined flies with lossof-function mutations in two major autophagy-related genes,
atg8 and atg7. Furthermore, we overexpressed a dominantnegative form of yet another important autophagy protein,
Atg1, in early spermatogonial cells. However, no significant
effects on the levels of GCD have been recorded in any of these
mutant backgrounds (Figure S3A). Consistently, expression of
the common LC3(Atg8)-GFP autophagy reporter in these cells
revealed no typical puncta of lipidated Atg8 accumulating on
autophagosomal vesicles, but rather more homogenous cytoplasmic GFP distribution in living cells and a much dimmer signal
in dying cells (Figure S3B). Taken together, our ultrastructural,
genetic, and cytological analyses do
not support a role for macroautophagy
in GCD.
One of these Atg8-based autophagy
reporters was fused in tandem to
eGFP and mCherry, thus labeling autophagosomes in both green and red
fluorescence, whereas autolysosomes
are in red only because of the quenching
of the eGFP in the acidic environment
of the lysosomes (Nezis et al., 2010). Interestingly, despite the dimming of the eGFP
signal, the mCherry signal was still pronounced in dying spermatogonial cysts,
suggesting that the eGFP protein is not
degraded but rather quenched because
of global acidification of the cellular environment during GCD (Figure S3C).
The necrosis-like morphological features and the evidence of cellular acidification prompted us to further explore
other possible mechanistic similarities, such as ROS production
or lysosomal activity. Staining testes with the oxidant-sensitive
dye, dihydroethidium (DHE), revealed specific accumulation of
ROS at the time when dying spermatogonial cysts became
round and compact and the DNA was still visible (mid-GCD
stage; Figure 4A). Furthermore, using the lysosomal probe,
LysoTracker, which selectively accumulates in cellular compartments with low internal pH, revealed a profound signal in
what appeared to be dying spermatogonial cysts during midand late-GCD stages based on nuclear morphology (Figure 4B).
Indeed, when testes were costained with LysoTracker and TUNEL, the dying cysts were positive for both (Figure 4C).
Next, to directly investigate the involvement of lysosomes
in GCD, we examined the effects of mutations in lysosomal
catabolic or biogenesis genes. Significantly, mutant flies
homozygous for either a null allele of the abundant aspartyl
protease, cathepsin D (cathD1), a hypomorphic allele of the
acid Deoxyribonuclease (DNase) II (dnaseIIlo), or two hypomorphic alleles of the genetically linked lysosomal biogenesis
proteins, Deep-orange and Carnation (dor4 and car1, respectively), all displayed a marked, 40%–60%, decrease in the
levels of GCD (Figures 4D–4F). Of note, in addition to the
reduction in GCD levels, TUNEL labeling often appeared
diffused and/or incomplete in the dnaseII mutants, further
demonstrating the role of this nuclease in DNA fragmentation
during GCD (Figures S3D and S3E). Collectively, these findings
Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc. 33
Developmental Cell
Omi and Lysosomes Mediate Alternative Cell Death
Figure 4. ROS Accumulation, High Lysosomal Activity, and Some Lysosomal Hydrolases Are Involved in GCD
(A–C) Apical tips of wild-type testes stained with (A and B) Hochest or (C) DAPI to reveal the nuclei (blue) and either (A) DHE to detect ROS (red) or (B and C)
LysoTracker to detect lysosomal activity (red). (C) To reveal the dying cysts, testes were also labeled with TUNEL (green). (B) A yellow circle outlines a dying cyst
with visible packed nuclei (an early-mid GCD stage), whereas a white circle outlines a dying cyst with faint Hochest signal of the nuclei (an advanced GCD stage).
(C) Arrowheads indicate two dying cysts at either early-mid GCD stage (the lower cyst) or advanced stage (the upper cyst).
(D and E) Representative testes from the indicated genotypes were stained as in Figure 1B.
(F) Quantification of GCD levels in testes from the indicated genotypes. All calculations were performed as in Figure 2C.
Scale bars, 20 mm (A–E). See also Figure S3.
34 Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc.
Developmental Cell
Omi and Lysosomes Mediate Alternative Cell Death
demonstrate that catabolic activity originating in lysosomes is
central to GCD.
A Genetic Screen Identifies the Mitochondrial Serine
Protease HtrA2/Omi as an Important Mediator of GCD
To gain more insight into the regulation of GCD, we conducted
a deficiency screen for chromosomal deletions that dominantly
suppress GCD, as visualized by TUNEL labeling. After narrowing down one of the suppressive regions to 35 genes, we
used a candidate approach to prioritize genes based on high
expression in the testis (according to the FlyAtlas database)
and known functions. Importantly, one of these genes encodes
the Drosophila ortholog of the mitochondrial serine protease,
HtrA2/Omi, which has been previously implicated in both cell
death and the pathology of Parkinson’s disease and is thus
a strong candidate for the recorded effect on GCD. To explore
this possibility, we first generated two partially overlapping
small deficiency lines, omiDf1 and omiDf2, covering the entire
htrA2/omi gene as well as several adjacent genes (Figure 5A).
In parallel, using the fly-TILL service, we identified two mutant
alleles of htrA2/omi, omiIFD (containing a 123 bp in-frame
deletion), which lacks 41 amino acids (aa) in the protease
domain, including the catalytic serine, and omiV110E (contains
a T-to-A transversion), where a conserved valine was replaced
with glutamic acid within the homotrimerization domain
(Figure 5B). Finally, in the course of this study, an htrA2/omi
null allele, omiD1, was also generated (Figure 5A) (Tain et al.,
2009).
Initial analysis of these htrA2/omi mutants revealed that all are
viable and female fertile but male sterile, either as homozygotes
or in trans to the small deficiencies or to one another. This is in
agreement with previous studies reporting that inactivation of
htrA2/omi caused male sterility in flies (Yun et al., 2008; Tain
et al., 2009). We then examined whether GCD levels may be
attenuated in the htrA2/omi mutants. Critically, in accordance
to the dominant effect on GCD recorded for the original deficiency from the screen, flies heterozygous for either omiDf1,
omiDf2, or omiIFD displayed 50%–60% reduction in the levels of
GCD, uncovering a dosage-sensitive effect (haploinsufficiency)
of the htrA2/omi gene (Figure 5C). This effect may be attributed
to the fact that HtrA2/Omi is functional in cell death as a homotrimer rather than a monomer (Li et al., 2002). Consistent with
this idea, suppression of GCD was not significantly enhanced
in homozygotes or transheterozygous allelic combinations of
the htrA2/omi mutants, including the null omiD1 allele (Figures
5C, 5D, and 5G). Of note, flies heterozygous for the H99 deficiency, which deletes the four primary proapoptotic genes,
reaper, grim, hid, and sickle, displayed no significant effect on
GCD levels (Figure 5C).
To further validate that the loss of htrA2/omi is the cause of
the attenuation in GCD, we generated two ‘‘rescue’’ transgenes
inserted at a defined site of the Drosophila genome. The first
transgene, omiR-BAC, is a large (bacterial artificial chromosome;
BAC) genomic clone encompassing htrA2/omi and three
additional flanking genes, whereas the second transgene,
omiR-2, is a smaller genomic segment containing the htrA2/
omi locus only (red bars in Figure 5A). A third ‘‘rescue’’ transgene, omiR-1, which moderately expands beyond the 50 and
30 sequences of omiR-2, but is still confined to only the
htrA2/omi locus, was obtained from another group (Figure 5A)
(Tain et al., 2009). All the ‘‘rescue’’ transgenes were then
crossed to the omiDf1/IFD transheterozygous mutants and
analyzed for their ability to restore normal levels of GCD. Two
copies of each of the three transgenes restored GCD levels
to about 80% of the levels in wild-type testes, confirming the
important role of HtrA2/Omi in promoting GCD (Figures 5E
and 5F; Figures 6B and 6C). It is noteworthy that when
tested for its ability to restore normal levels of GCD in a yet
another mutant background, (i.e., the omiD1 homozygotes),
the omiR-1 transgene elevated the levels of GCD to 120% of
that in wild-type, indicating some alterations in the ‘‘rescue’’
efficiency of the transgenes in the different mutant backgrounds (Figure 5G).
The Catalytic Activity of HtrA2/Omi, and Not the IAPDependent Function, Is Important for Both GCD and
Male Fertility
After being processed in the mitochondria, the mature form of
HtrA2/Omi exposes an N-terminal IAP-binding motif (IBM),
related to the tetrapeptide motifs found in IAP antagonists of
the Reaper-family proteins in Drosophila and Smac/DIABLO in
mammals (Suzuki et al., 2001). Various apoptotic insults in cell
culture were reported to trigger translocation of the mature
HtrA2/Omi to the cytosol, where it can promote either
caspase-dependent or -independent cell death by binding to
IAPs or through its self-protease activity, respectively (Vande
Walle et al., 2008). To gain more insight into the function of
HtrA2/Omi in promoting GCD, we performed functional studies
in vivo by mutating either the catalytic serine or the three
suggested IBMs of HtrA2/Omi within the omiR-2 ‘‘rescue’’ transgene and testing whether they can still rescue GCD levels in
the omiDf1/IFD transheterozygous mutants (Figures 6A and 6E).
Importantly, the elimination of the catalytic serine, not the
IBMs, significantly (albeit not completely) deteriorated the
ability of the omiR-2 transgene to induce GCD in the mutant
(Figures 6B and 6C). Therefore, it is the catalytic activity of
HtrA2/Omi, and not the IAP-dependent function, that may be
important for its function in GCD.
Fertility tests revealed that omiDf1/IFD mutant males produce
only 2% of the number of progeny generated by wild-type (Figure 6D). Whereas the cause of this sterility is unclear as these
mutants still produce mature and motile sperm cells (data not
shown), it may be, at least in part, unrelated to the role of
HtrA2/Omi in GCD, because normally the majority of spermatogonial cysts (70%–80%) are not destined to die. Nevertheless, the function of HtrA2/Omi in male fertility appears to
depend on its catalytic activity rather than its ability to interact
with IAPs, because expression of the IBM-less form of HtrA2/
Omi, but not the catalytic mutant, rescued the sterility of
omiDf1/IFD mutant males to similar levels as those obtained
with the intact, omiR-2, transgene (about 75% of wild-type
levels; Figure 6D). Interestingly, although the mutation in the
catalytic serine strongly deteriorated the ability of the omiR-2
transgene to rescue GCD and sterility, it may still preserve
some residual activity because it was sufficient to cause
a 25% increase in GCD levels, as well as a 3-fold increase
in the number of progeny produced by mutant males (Figures
6C and 6D).
Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc. 35
Developmental Cell
Omi and Lysosomes Mediate Alternative Cell Death
Figure 5. The Mitochondrial Serine Protease HtrA2/Omi Is an Important Mediator of GCD
(A) An in-scale diagram of the genomic organization of htrA2/omi (red arrow trapezoid) and other flanking genes (black arrow trapezoids) is highlighted in pink. The
directions of the arrows correspond to the relative 50 -30 directions of the genes. Chromosomal bands (gray) are indicated above. Green bars denote the deleted
areas in the small deficiencies and the omiD1 mutant, whereas dashed gray lines depict their distal and proximal breakpoints. The genomic segments used in the
‘‘rescue’’ constructs are indicated by red bars.
(B) A schematic structure of the Drosophila HtrA2/Omi protein. The relative locations of the different domains and the fly-TILL mutants are indicated. MTS,
mitochondrial targeting sequence.
(C) Quantification of GCD levels in testes from the indicated genotypes. All calculations were performed as in Figure 2C.
(D and E) Representative testes from the indicated genotypes were stained as in Figure 1B. Scale bar, 20 mm.
(F and G) Quantification of GCD levels in testes from the indicated genotypes. The presence of a specific ‘‘rescue’’ transgene is depicted in red letters. All
calculations were performed as in Figure 2C.
See also Figure S4.
36 Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc.
Developmental Cell
Omi and Lysosomes Mediate Alternative Cell Death
Figure 6. In Vivo Analyses of IBM-Deficient
and Catalytically Inactive Forms of HtrA2/
Omi
(A) A schematic structure of the coding region in
the HtrA2/Omi rescue constructs. The introduced
mutations are denoted by red asterisks.
(B) Representative testes from the indicated
genotypes were stained as in Figure 1B. Scale bar,
20 mm.
(C) Quantification of GCD levels in testes from the
indicated genotypes. ‘‘Rescue’’ transgenes are
depicted in red letters. All calculations were performed as in Figure 2C.
(D) Quantification of pupal progeny of male parents
from the indicated genotypes mated to wild-type
females. ‘‘Rescue’’ transgenes are depicted in red
letters. The number of repeats (n) is indicated in the
vicinity of each column. Error bars represent
standard errors. p values of the level differences
are indicated above the bars. NS, nonsignificant.
(E) The omi ‘‘rescue’’ transgenes are expressed at
similar levels. Transcriptional expression from the
transgenes was confirmed by RT-PCR analyses
on RNA from testes of the indicated genotypes.
RT+Taq and Taq indicate reactions with or without
reverse transcriptase, respectively. WT, wild-type
cDNA; omitran, transgene cDNAs; omiIFD, omiDf1/IFD
cDNA; omigen, omi genomic DNA.
HtrA2/Omi and the Lysosomes Act in Both Common
and Parallel Pathways during GCD
Lysosomes have been implicated in both the initiation and
execution of cell death (Kroemer and Jäättelä, 2005; Golstein
and Kroemer, 2007). To explore the relationship between the
mitochondria and lysosomes in GCD, we performed epistatic
experiments by first analyzing double mutants for htrA2/omi
and cathD. Significantly, GCD levels in the double mutant were
further decreased by half, from 40% of the wild-type levels in
the single mutants to about 20% in the double mutant, suggesting the existence of some independent functions of these two
pathways in GCD (Figure S4A). However, when measuring the
levels of LysoTracker-positive spermatogonia in htrA2/omi
mutant testes, we recorded a 50% decrease as compared to
wild-type levels, indicating that lysosomes act both downstream
and in parallel to HtrA2/Omi to promote GCD (Figure S4B).
The Parkinson’s Disease-Associated Gene pink1,
Not parkin, Is Involved in GCD
Previous genetic studies in mouse and Drosophila have suggested that HtrA2/Omi may physically and genetically interact
with the Parkinson’s disease (PD)-associated mitochondrial
kinase, Pink1 (Plun-Favreau et al., 2007;
Tain et al., 2009). Pink1 has been demonstrated to act upstream of another important PD-associated gene, parkin, to
promote a mitochondrial homeostasis
pathway (Vande Walle et al., 2008). We
therefore tested whether the pink1-parkin
pathway may also be involved in GCD.
Whereas a significant 40% decrease in
GCD levels was recorded in loss-of-function mutants for pink1
(pink1B9), no significant effect was detected in a null allelic
combination of parkin (park25/Df) (Figure S5A). Previous reports
in Drosophila have similarly demonstrated common phenotypes
for pink1 and parkin mutants that are distinct from the phenotypes shared by pink1 and htrA2/omi mutants (Yun et al., 2008;
Tain et al., 2009). Furthermore, although mutations in all three
genes cause male sterility, phenotypic analyses imply distinct
roles of Parkin and HtrA2/Omi during late spermatogenesis
(data not shown; Yun et al., 2008). We therefore conclude that
the pink1-htrA2/omi pathway is involved in GCD independently
of Parkin and that both pathways act at different stages to
promote proper sperm production.
The Bcl-2 Family Proteins Debcl and Buffy
and the Mitochondrial Nuclease EndoG
Are Associated with GCD
Similar to Cyt-c, HtrA2/Omi is also released from the mitochondria into the cytosol during apoptosis in mammals (Vande Walle
et al., 2008), a process mediated by the proapoptotic members
of the Bcl-2 family (Youle and Strasser, 2008). Two members
of this protein family exist in Drosophila, Debcl and Buffy, and
Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc. 37
Developmental Cell
Omi and Lysosomes Mediate Alternative Cell Death
Figure 7. Comparison between the Models
of GCD and Conventional Apoptosis Pathways in Drosophila
(A) The diagram depicts the components involved
in GCD. Solid arrows indicate proposed connections based on interpretation of the genetic data in
this study, whereas dashed arrows describe
possible interactions based on data from the
literature. The asterisk signifies a proposed target
for Dronc based on (Gyrd-Hansen et al., 2006).
(B) A schematic model of the core apoptotic
machinery in Drosophila. Major components and
interactions are depicted by solid arrows (activation) or T-shapes (inhibition), whereas dashed lines
or arrows signify less general components and
interactions.
See also Figure S5.
their involvement in apoptosis is marginal, if any (Sevrioukov
et al., 2007; Galindo et al., 2009). We examined the levels of
GCD in strong debcl (debclE26) and buffy (buffyH37) mutants,
using homozygous, hemizygous (in trans to deficiencies), and
double-mutant allelic combinations (Sevrioukov et al., 2007).
All these mutant allelic combinations displayed a 60%–80%
reduction in GCD levels, which was not further enhanced in the
double mutants, suggesting that both proteins may act in a linear
pathway to promote GCD (Figure S5B). These findings suggest
an important role of the Drosophila Bcl-2 proteins in developmental cell death.
Endonuclease G is another prodeath enzyme, which resides
in the mitochondrial intermembrane space and is released to
the cytosol during apoptosis (Widlak and Garrard, 2005). A
loss-of-function allele of the ubiquitously expressed archetype
of this group in Drosophila, endoGMB07150, has been recently
described (DeLuca and O’Farrell, 2012). Using this mutant,
we detected a 60% decrease in GCD levels (Figure S5C). Altogether, these findings establish the mitochondria as central
regulators of GCD.
DISCUSSION
In the present study, we describe an alternative cell death
pathway in Drosophila, distinct from the conventional apoptotic
pathway, which operates to execute some of the early developing male germ cells in the adult testis (Figure 7).
Caspases and GCD
Our findings that GCD levels increase when caspase activity is
blocked is reminiscent of a phenomenon previously noted in
certain mammalian cells lacking key apoptotic components,
yet induced to undergo apoptosis; instead of maintaining
cellular viability these cells may shift the apoptotic pathway to
another cell death pathway (Vercammen et al., 1998). Although
the mechanisms and physiological relevance of this phenomenon are still vague, it has been hypothesized that some cells
may maintain baseline levels of caspase activity in order to
prevent, possibly through cleavage and inactivation of pronecrotic factors, a default induction of necrosis (He et al., 2009).
It is important to note that in addition to this phenomenon,
GCD shares other similarities with certain necrotic pathways,
including some morphological traits. In particular, Jäättelä
38 Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc.
and colleagues have described an alternative cell death
pathway induced by tumor necrosis factor (TNF) in murine
embryonic fibroblasts (MEFs), which largely proceeds in the
absence of effector caspase activity but requires the lysosomal
pathway. Strikingly, similar to GCD, which involves Dronc but
not Ark, it is caspase-9 and not Apaf-1 that is required for
this TNF-induced cell death pathway. Furthermore, caspase-9
activity in these cells was shown to trigger LMP and the
concomitant release of cathepsins into the cytosol (GyrdHansen et al., 2006). However, the signals that trigger the
lysosomal pathway in MEFs may be distinct from GCD, as
mutations in the sole fly TNF ortholog, Eiger, did not attenuate
GCD (data not shown). Furthermore, whereas in MEFs, the
apical caspase-8 is required for the apoptosome-independent
activation of caspase-9, loss-of-function mutants for the two
other initiator-like caspases in Drosophila, Dredd and Strica,
did not affect GCD levels.
The Lysosomal Pathway in GCD and the Crosstalk
with the Mitochondrial Pathway
The notion that under certain conditions the lysosomes can
serve as ‘‘suicide bags’’ is not a new one (Kirkegaard and Jäättelä, 2009). Lysosomes contain over 50 different hydrolases,
including proteases that are highly important for not only
nonspecific but also selective protein degradation (Turk et al.,
2002). Two major mechanisms for the action of the lysosomes
in cellular degradation during alternative cell death pathways
have been proposed: sequestration of cellular contents to lysosomes by the autophagic machinery or spillage of lysosomal
catabolic enzymes to the cytosol following LMP induction. Interestingly, whereas most of these enzymes function optimally in
the acidic environment of the lysosomal lumen, some of them,
including cathepsins B and D, are also functional in the cytosol
(Golstein and Kroemer, 2007; Kirkegaard and Jäättelä, 2009).
Our findings that GCD is not dependent on the autophagy
pathway, while it is still associated with cytosolic acidification
and lysosomal activity, including cathepsin D, favors an LMPrelated mode of action in GCD. This idea is also supported by
the finding that dying spermatogonia accumulate ROS, which
has been proposed to be an important inducer of LMP (Kroemer
and Jäättelä, 2005).
The role of the lysosomes in promoting GCD, as well as the
crosstalk between lysosomes and the mitochondria may be
Developmental Cell
Omi and Lysosomes Mediate Alternative Cell Death
complex. The current study suggests that the lysosomes may
act both downstream and in parallel to the mitochondrial
pathway and fulfill at least some of the execution functions of
GCD. Both the mitochondria and the lytic vacuole (the functional
equivalent of the lysosome) have been recently implicated in
a developmental program of nuclear destruction during yeast
sporogenesis under starvation, reminiscent of some aspects
of GCD, which raises the hypothesis that GCD might have
evolved before the emergence of the caspases and apoptosis
(Eastwood et al., 2012).
HtrA2/Omi and the Mitochondrial Pathway in GCD
One exciting implication of the current study concerns the
discovery of a role for HtrA2/Omi in developmental cell death.
A large body of circumstantial evidence indicates two distinct
mechanisms of action for HtrA2/Omi in mammalian cell death.
HtrA2/Omi can induce caspase activation and apoptosis by
binding to IAPs and relieving their inhibition of caspases, but
it can also trigger caspase-independent cell death through its
own serine protease activity (Suzuki et al., 2001; Li et al.,
2002). Interestingly, several lines of evidence from the current
work and other studies suggest that it is the catalytic activity
of HtrA2/Omi, and not its IAP-dependent function, that is important during physiological cell death. In accordance, although
these IBMs were shown to interact with the baculovirus IAP
repeat (BIR) domains of Diap1, cell death induced by ectopic
expression of the different mature (cleaved) forms of HtrA2/
Omi in S2 cells and the Drosophila eye could not be blocked
by coexpression of the baculoviral effector caspase inhibitor
protein, p35, indicating that at least in Drosophila, HtrA2/Omi
mainly induces caspase-independent cell death (Igaki et al.,
2007; Challa et al., 2007; Khan et al., 2008). Indeed, as opposed
to the compelling evidence for the importance of the Reaperfamily proteins in the induction of most if not all developmental
and stress-induced cell death in Drosophila (Fuchs and Steller,
2011), HtrA2/Omi has been shown to be dispensable for
apoptosis in flies (Yun et al., 2008; Tain et al., 2009). Likewise,
htrA2/omi knockout/mutant mice displayed no apoptosisrelated phenotypes (Jones et al., 2003). It is also interesting
to note that the IBM of HtrA2/Omi is not conserved in all mammalian homologs (Li et al., 2002). Therefore, disruption of the
IAP-caspase interaction may not be the main cell death function
of HtrA2/Omi in at least some mammals.
Very little is known about how HtrA2/Omi protease activity
can trigger caspase-independent cell death. Overexpression
of an extramitochondrial form of HtrA2/Omi in mammalian
cells induced caspase-independent cell death with atypical
morphology, largely reminiscent of the morphological features
of GCD, including some apoptosis-like morphological changes
(i.e., cell rounding and shrinkage) and the lack of other apoptotic
events, implying that HtrA2/Omi may either directly or indirectly
trigger the cleavage of at least some caspase substrates (Suzuki
et al., 2001). Along this line, several HtrA2/Omi cytosolic
substrates were identified in a proteomic study, most of which
are also known targets of caspases (Vande Walle et al., 2007).
In conclusion, whereas it is still unclear why relatively high
numbers of spermatogonia undergo cell death during normal
testis development, cell death of premeiotic male germ cells
has also been reported in many mammalian species (Allan
et al., 1987). Two phases of developmental germ cell death
have been reported in rodents: an early coordinated wave of
cell death, which is estimated to eliminate approximately 80%
of the newly generated spermatocytes during the first round of
spermatogenesis and constant spontaneous cell death in the
adult testis (Allan et al., 1992; Rodriguez et al., 1997). In striking
similarity to Drosophila, however, this spontaneous cell death is
mainly restricted to spermatogonia and is estimated to eliminate
about 30% of these newly generated germ cells. Moreover,
whereas the early wave of cell death was blocked in mice overexpressing the apoptosis inhibitory proteins, Bcl2 and BclxL,
or in mice deficient for the proapoptotic protein, Bax, the spontaneous cell death in the adult testis was not affected in these
animals, suggesting a possible divergence from the conventional
apoptotic program (Knudson et al., 1995; Rodriguez et al., 1997).
Indeed, these spontaneously dying spermatogonia exhibited
morphological features of both apoptosis and necrosis (Allan
et al., 1992). Although the advantages in using alternative cell
death pathways are generally unknown, it is interesting to note
that the conventional apoptotic machinery, including caspases,
is activated later in spermatogenesis during the vital process of
spermatid individualization (Arama et al., 2003; Kaplan et al.,
2010). Because many anatomical and molecular features of
gametogenesis have been conserved in evolution, it is attractive
to propose that GCD may constitute a more ancient form of cell
death, whereas caspases might have originally evolved to
promote non-cell-death processes (Aram and Arama, 2012).
Future studies of GCD in Drosophila shall help to determine the
cause of this spontaneous cell death and the reason why these
cells are executed by an alternative cell death pathway.
EXPERIMENTAL PROCEDURES
Fly Strains
All fly strains and the generation of all the constructs and transgenes
mentioned in this study are described in detail in the Supplemental Experimental Procedures.
The small deficiency lines, omiDf1 and omiDf2, were generated by applying
the FRT-FLP recombination procedure (Parks et al., 2004), using the following
pairs of FRT-bearing lines: P{XP}d04326 and PBac{WH}HtrA2[f03785] (for
omiDf1) and P{XP}d06999 and PBac{WH}f00974 (for omiDf2).
The omiIFD and omiV110E are EMS-induced mutant alleles, which we originally identified in the Zuker collection lines using the Fly-TILL service (Cooper
et al., 2008; FHCRC, Seattle, Washington).
The Deficiency Screen
We used the ‘‘Bloomington deficiency kit,’’ a collection of several hundred
chromosomal deletion lines, which collectively cover 98% of the genome
(Cook et al., 2012). We initially screened through the third chromosome subset,
encompassing about one-third of the total collection, and identified three
deficiency lines with nonoverlapping chromosomal deletions that significantly
suppressed GCD as heterozygotes. We then screened deficiency lines with
smaller deletions in the respective suppressive regions, which limited one
of these regions to a relatively small chromosomal segment, bands 88B288C9, containing only 35 genes, including htrA2/omi.
TUNEL, LysoTracker, and DHE Labeling
TUNEL labeling of the testis was carried out using the ApopTag kit (Millipore,
Billerica, MA, USA) as described in Arama and Steller (2006).
For TUNEL and cleaved caspase-3 double staining, the rabbit anti-cleaved
caspase-3 antibody (1:100; Asp175; Cell Signaling Technology, Danvers, MA,
USA) was added to the anti-digoxigenin solution. The secondary antibody was
added after the washes of the following day.
Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc. 39
Developmental Cell
Omi and Lysosomes Mediate Alternative Cell Death
For LysoTracker and Hoechst double staining, testes were dissected in
PBS, incubated in staining solution (100 mM LysoTracker Red DND-99
[Molecular Probes, Carlsbad, CA, USA] and 16.2 mM Hoechst 33342
[Molecular Probes] in PBS) for 10 min, rinsed in PBS, mounted in a drop of
PBS, and immediately photographed. For DHE and Hoechst double staining,
we performed a similar procedure using instead a staining solution with 20 mM
DHE (Molecular Probes) and incubating for 30 min.
Images were taken on a confocal microscope (LSM510 Meta Inverted
Axio Observer; Carl Zeiss) and captured using the LSM510 operating
software.
Ultrastructural Studies
Testes were prepared for TEM analysis as described in Arama et al. (2006).
Ultra-thin sections were prepared with ultramicrotome Leica UCT (Leica),
analyzed under 120kV transmission electron microscope Tecnai 12, and
digitized with EAGLE CCD camera using TIA software (FEI, Eindhoven). The
electron microscopy studies were conducted at the Irving and Cherna
Moskowitz Center for Nano and Bio-Nano Imaging at the WIS.
Fertility Test
Young adult males were placed individually with three wild-type virgin females
in separate vials at 25 C, and vials were scored for offspring pupa after
12 days.
Quantification of GCD
The number of fixed TUNEL or LysoTracker-positive spermatogonial
cysts in each testis was manually counted under the confocal microscope. The mean score of the wild-type control group was set as 100%,
and the values of the other genotype groups were expressed as percentage of the wild-type group. GCD levels are represented as the mean ±
SEM of pooled results obtained from the indicated number of samples.
Statistical analysis was performed by Fisher’s protected least significant
difference posttest for multiple comparisons using the StatView Program
(Abacus Concepts, Berkeley, CA, USA). Significance level was considered
as p < 0.05.
SUPPLEMENTAL INFORMATION
REFERENCES
Allan, D.J., Harmon, B.V., and Kerr, J.F.R. (1987). Cell death in spermatogenesis. In Perspectives on Mammalian Cell Death, C.S. Potten, ed. (London:
Oxford University Press), pp. 229–258.
Allan, D.J., Harmon, B.V., and Roberts, S.A. (1992). Spermatogonial apoptosis
has three morphologically recognizable phases and shows no circadian
rhythm during normal spermatogenesis in the rat. Cell Prolif. 25, 241–250.
Aram, L., and Arama, E. (2012). Sporoptosis: sowing the seeds of nuclear
destruction. Dev. Cell 23, 5–6.
Arama, E., and Steller, H. (2006). Detection of apoptosis by terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling and acridine orange in
Drosophila embryos and adult male gonads. Nat. Protoc. 1, 1725–1731.
Arama, E., Agapite, J., and Steller, H. (2003). Caspase activity and a specific
cytochrome C are required for sperm differentiation in Drosophila. Dev. Cell
4, 687–697.
Arama, E., Bader, M., Srivastava, M., Bergmann, A., and Steller, H. (2006). The
two Drosophila cytochrome C proteins can function in both respiration and
caspase activation. EMBO J. 25, 232–243.
Ashkenazi, A., and Dixit, V.M. (1998). Death receptors: signaling and modulation. Science 281, 1305–1308.
Bader, M., and Steller, H. (2009). Regulation of cell death by the ubiquitin-proteasome system. Curr. Opin. Cell Biol. 21, 878–884.
Berry, D.L., and Baehrecke, E.H. (2007). Growth arrest and autophagy are
required for salivary gland cell degradation in Drosophila. Cell 131, 1137–1148.
Blum, E.S., Abraham, M.C., Yoshimura, S., Lu, Y., and Shaham, S. (2012).
Control of nonapoptotic developmental cell death in Caenorhabditis elegans
by a polyglutamine-repeat protein. Science 335, 970–973.
Bredesen, D.E. (2008). Programmed cell death mechanisms in neurological
disease. Curr. Mol. Med. 8, 173–186.
Budihardjo, I., Oliver, H., Lutter, M., Luo, X., and Wang, X. (1999). Biochemical
pathways of caspase activation during apoptosis. Annu. Rev. Cell Dev. Biol.
15, 269–290.
Challa, M., Malladi, S., Pellock, B.J., Dresnek, D., Varadarajan, S., Yin, Y.W.,
White, K., and Bratton, S.B. (2007). Drosophila Omi, a mitochondrial-localized
IAP antagonist and proapoptotic serine protease. EMBO J. 26, 3144–3156.
Supplemental Information includes five figures and Supplemental Experimental Procedures and can be found with this article online at http://dx.doi.
org/10.1016/j.devcel.2013.02.002.
Chew, S.K., Akdemir, F., Chen, P., Lu, W.J., Mills, K., Daish, T., Kumar, S.,
Rodriguez, A., and Abrams, J.M. (2004). The apical caspase dronc governs
programmed and unprogrammed cell death in Drosophila. Dev. Cell 7,
897–907.
ACKNOWLEDGMENTS
Cook, R.K., Christensen, S.J., Deal, J.A., Coburn, R.A., Deal, M.E., Gresens,
J.M., Kaufman, T.C., and Cook, K.R. (2012). The generation of chromosomal
deletions to provide extensive coverage and subdivision of the Drosophila
melanogaster genome. Genome Biol. 13, R21.
We are grateful to A. Bergmann, J. Chung, M.B. Feany, L. Gliboa, E. Kurant,
K. McCall, P. Meier, T.P. Neufeld, I.P. Nezis, L. Pallank, H.D. Ryoo, B.Z. Shilo,
H. Steller, H. Stenmark, N. Tapon, and the Bloomington Stock Center for
providing additional stocks and reagents. We thank the Arama laboratory
members for encouragement and advice. We are indebted to S. Zaidman
and V. Shinder for constant help and guidance with the EM studies. We
warmly thank S. Brown for editing the manuscript. This research was supported in part by the Israel Science Foundation (grant no. 308/09), Ministry
of Agriculture of the State of Israel, Minerva Foundation with funding from
the German Federal Ministry of Education and Research, Fritz Thyssen Stiftung, Israel Cancer Research Fund, Israel Cancer Association, and the Moross
Institute for Cancer Research. E.A. is also supported by research grants from
Rudolfine Steindling, the Jeanne and Joseph Nissim Center for Life Sciences
Research, and the Y. Leon Benoziyo Institute for Molecular Medicine. E.A. is
the Incumbent of the Corinne S. Koshland Career Development Chair. K.Y-S.
is supported by a postdoctoral fellowship from the Israel Cancer Research
Fund.
Received: August 26, 2012
Revised: January 3, 2013
Accepted: February 4, 2013
Published: March 21, 2013
40 Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc.
Cooper, J.L., Till, B.J., and Henikoff, S. (2008). Fly-TILL: reverse genetics using
a living point mutation resource. Fly (Austin) 2, 300–302.
Degterev, A., Boyce, M., and Yuan, J. (2003). A decade of caspases.
Oncogene 22, 8543–8567.
Degterev, A., Hitomi, J., Germscheid, M., Ch’en, I.L., Korkina, O., Teng, X.,
Abbott, D., Cuny, G.D., Yuan, C., Wagner, G., et al. (2008). Identification of
RIP1 kinase as a specific cellular target of necrostatins. Nat. Chem. Biol. 4,
313–321.
DeLuca, S.Z., and O’Farrell, P.H. (2012). Barriers to male transmission of mitochondrial DNA in sperm development. Dev. Cell 22, 660–668.
Denton, D., Shravage, B., Simin, R., Mills, K., Berry, D.L., Baehrecke, E.H., and
Kumar, S. (2009). Autophagy, not apoptosis, is essential for midgut cell death
in Drosophila. Curr. Biol. 19, 1741–1746.
Eastwood, M.D., Cheung, S.W., Lee, K.Y., Moffat, J., and Meneghini, M.D.
(2012). Developmentally programmed nuclear destruction during yeast
gametogenesis. Dev. Cell 23, 35–44.
Edinger, A.L., and Thompson, C.B. (2004). Death by design: apoptosis,
necrosis and autophagy. Curr. Opin. Cell Biol. 16, 663–669.
Developmental Cell
Omi and Lysosomes Mediate Alternative Cell Death
Fan, Y., and Bergmann, A. (2010). The cleaved-Caspase-3 antibody is a marker
of Caspase-9-like DRONC activity in Drosophila. Cell Death Differ. 17,
534–539.
Feinstein-Rotkopf, Y., and Arama, E. (2009). Can’t live without them, can live
with them: roles of caspases during vital cellular processes. Apoptosis 14,
980–995.
Florentin, A., and Arama, E. (2012). Caspase levels and execution efficiencies
determine the apoptotic potential of the cell. J. Cell Biol. 196, 513–527.
Franc, N.C., Dimarcq, J.L., Lagueux, M., Hoffmann, J., and Ezekowitz, R.A.
(1996). Croquemort, a novel Drosophila hemocyte/macrophage receptor
that recognizes apoptotic cells. Immunity 4, 431–443.
Fuchs, Y., and Steller, H. (2011). Programmed cell death in animal development and disease. Cell 147, 742–758.
Galindo, K.A., Lu, W.J., Park, J.H., and Abrams, J.M. (2009). The Bax/Bak ortholog in Drosophila, Debcl, exerts limited control over programmed cell
death. Development 136, 275–283.
Galluzzi, L., Vanden Berghe, T., Vanlangenakker, N., Buettner, S., Eisenberg,
T., Vandenabeele, P., Madeo, F., and Kroemer, G. (2011). Programmed
necrosis from molecules to health and disease. Int. Rev. Cell Mol. Biol. 289,
1–35.
Golstein, P., and Kroemer, G. (2007). Cell death by necrosis: towards a molecular definition. Trends Biochem. Sci. 32, 37–43.
Gyrd-Hansen, M., Farkas, T., Fehrenbacher, N., Bastholm, L., Høyer-Hansen,
M., Elling, F., Wallach, D., Flavell, R., Kroemer, G., Nylandsted, J., and Jäättelä,
M. (2006). Apoptosome-independent activation of the lysosomal cell death
pathway by caspase-9. Mol. Cell. Biol. 26, 7880–7891.
He, S., Wang, L., Miao, L., Wang, T., Du, F., Zhao, L., and Wang, X. (2009).
Receptor interacting protein kinase-3 determines cellular necrotic response
to TNF-alpha. Cell 137, 1100–1111.
Igaki, T., Suzuki, Y., Tokushige, N., Aonuma, H., Takahashi, R., and Miura, M.
(2007). Evolution of mitochondrial cell death pathway: Proapoptotic role of
HtrA2/Omi in Drosophila. Biochem. Biophys. Res. Commun. 356, 993–997.
Jones, J.M., Datta, P., Srinivasula, S.M., Ji, W., Gupta, S., Zhang, Z., Davies,
E., Hajnóczky, G., Saunders, T.L., Van Keuren, M.L., et al. (2003). Loss of Omi
mitochondrial protease activity causes the neuromuscular disorder of mnd2
mutant mice. Nature 425, 721–727.
Kaplan, Y., Gibbs-Bar, L., Kalifa, Y., Feinstein-Rotkopf, Y., and Arama, E.
(2010). Gradients of a ubiquitin E3 ligase inhibitor and a caspase inhibitor
determine differentiation or death in spermatids. Dev. Cell 19, 160–173.
Khan, F.S., Fujioka, M., Datta, P., Fernandes-Alnemri, T., Jaynes, J.B., and
Alnemri, E.S. (2008). The interaction of DIAP1 with dOmi/HtrA2 regulates cell
death in Drosophila. Cell Death Differ. 15, 1073–1083.
Kirkegaard, T., and Jäättelä, M. (2009). Lysosomal involvement in cell death
and cancer. Biochim. Biophys. Acta 1793, 746–754.
Knudson, C.M., Tung, K.S., Tourtellotte, W.G., Brown, G.A., and Korsmeyer,
S.J. (1995). Bax-deficient mice with lymphoid hyperplasia and male germ
cell death. Science 270, 96–99.
Kroemer, G., and Jäättelä, M. (2005). Lysosomes and autophagy in cell death
control. Nat. Rev. Cancer 5, 886–897.
Li, W., Srinivasula, S.M., Chai, J., Li, P., Wu, J.W., Zhang, Z., Alnemri, E.S., and
Shi, Y. (2002). Structural insights into the pro-apoptotic function of mitochondrial serine protease HtrA2/Omi. Nat. Struct. Biol. 9, 436–441.
Lindsley, D., and Tokuyasu, K.T. (1980). Spermatogenesis. In Genetics and
Biology of Drosophila, M. Ashburner and T.R. Wright, eds. (New York:
Academic Press), pp. 225–294.
Lüthi, A.U., and Martin, S.J. (2007). The CASBAH: a searchable database of
caspase substrates. Cell Death Differ. 14, 641–650.
Metzstein, M.M., Stanfield, G.M., and Horvitz, H.R. (1998). Genetics of programmed cell death in C. elegans: past, present and future. Trends Genet.
14, 410–416.
Nezis, I.P., Shravage, B.V., Sagona, A.P., Lamark, T., Bjørkøy, G., Johansen,
T., Rusten, T.E., Brech, A., Baehrecke, E.H., and Stenmark, H. (2010).
Autophagic degradation of dBruce controls DNA fragmentation in nurse cells
during late Drosophila melanogaster oogenesis. J. Cell Biol. 190, 523–531.
O’Riordan, M.X., Bauler, L.D., Scott, F.L., and Duckett, C.S. (2008). Inhibitor of
apoptosis proteins in eukaryotic evolution and development: a model of
thematic conservation. Dev. Cell 15, 497–508.
Parks, A.L., Cook, K.R., Belvin, M., Dompe, N.A., Fawcett, R., Huppert, K.,
Tan, L.R., Winter, C.G., Bogart, K.P., Deal, J.E., et al. (2004). Systematic generation of high-resolution deletion coverage of the Drosophila melanogaster
genome. Nat. Genet. 36, 288–292.
Plun-Favreau, H., Klupsch, K., Moisoi, N., Gandhi, S., Kjaer, S., Frith, D.,
Harvey, K., Deas, E., Harvey, R.J., McDonald, N., et al. (2007). The mitochondrial protease HtrA2 is regulated by Parkinson’s disease-associated kinase
PINK1. Nat. Cell Biol. 9, 1243–1252.
Riedl, S.J., and Shi, Y. (2004). Molecular mechanisms of caspase regulation
during apoptosis. Nat. Rev. Mol. Cell Biol. 5, 897–907.
Rodriguez, J., and Lazebnik, Y. (1999). Caspase-9 and APAF-1 form an active
holoenzyme. Genes Dev. 13, 3179–3184.
Rodriguez, I., Ody, C., Araki, K., Garcia, I., and Vassalli, P. (1997). An early and
massive wave of germinal cell apoptosis is required for the development of
functional spermatogenesis. EMBO J. 16, 2262–2270.
Rubinstein, A.D., Eisenstein, M., Ber, Y., Bialik, S., and Kimchi, A. (2011). The
autophagy protein Atg12 associates with antiapoptotic Bcl-2 family members
to promote mitochondrial apoptosis. Mol. Cell 44, 698–709.
Ryoo, H.D., and Baehrecke, E.H. (2010). Distinct death mechanisms in
Drosophila development. Curr. Opin. Cell Biol. 22, 889–895.
Salvesen, G.S., and Dixit, V.M. (1999). Caspase activation: the induced-proximity model. Proc. Natl. Acad. Sci. USA 96, 10964–10967.
Sevrioukov, E.A., Burr, J., Huang, E.W., Assi, H.H., Monserrate, J.P., Purves,
D.C., Wu, J.N., Song, E.J., and Brachmann, C.B. (2007). Drosophila Bcl-2
proteins participate in stress-induced apoptosis, but are not required for
normal development. Genesis 45, 184–193.
Srivastava, M., Scherr, H., Lackey, M., Xu, D., Chen, Z., Lu, J., and Bergmann,
A. (2007). ARK, the Apaf-1 related killer in Drosophila, requires diverse
domains for its apoptotic activity. Cell Death Differ. 14, 92–102.
Sun, L., Wang, H., Wang, Z., He, S., Chen, S., Liao, D., Wang, L., Yan, J., Liu,
W., Lei, X., and Wang, X. (2012). Mixed lineage kinase domain-like protein
mediates necrosis signaling downstream of RIP3 kinase. Cell 148, 213–227.
Suzuki, Y., Imai, Y., Nakayama, H., Takahashi, K., Takio, K., and Takahashi, R.
(2001). A serine protease, HtrA2, is released from the mitochondria and
interacts with XIAP, inducing cell death. Mol. Cell 8, 613–621.
Tain, L.S., Chowdhury, R.B., Tao, R.N., Plun-Favreau, H., Moisoi, N., Martins,
L.M., Downward, J., Whitworth, A.J., and Tapon, N. (2009). Drosophila HtrA2
is dispensable for apoptosis but acts downstream of PINK1 independently
from Parkin. Cell Death Differ. 16, 1118–1125.
Tittel, J.N., and Steller, H. (2000). A comparison of programmed cell death
between species. Genome Biol. 1, reviews0003–reviews0003.6.
Turk, B., Stoka, V., Rozman-Pungercar, J., Cirman, T., Droga-Mazovec, G.,
, K., and Turk, V. (2002). Apoptotic pathways: involvement of lysosomal
Oresic
proteases. Biol. Chem. 383, 1035–1044.
Vande Walle, L., Lamkanfi, M., and Vandenabeele, P. (2008). The mitochondrial serine protease HtrA2/Omi: an overview. Cell Death Differ. 15, 453–460.
Vande Walle, L., Van Damme, P., Lamkanfi, M., Saelens, X., Vandekerckhove,
J., Gevaert, K., and Vandenabeele, P. (2007). Proteome-wide identification of
HtrA2/Omi Substrates. J. Proteome Res. 6, 1006–1015.
Varfolomeev, E.E., Schuchmann, M., Luria, V., Chiannilkulchai, N., Beckmann,
J.S., Mett, I.L., Rebrikov, D., Brodianski, V.M., Kemper, O.C., Kollet, O., et al.
(1998). Targeted disruption of the mouse Caspase 8 gene ablates cell death
induction by the TNF receptors, Fas/Apo1, and DR3 and is lethal prenatally.
Immunity 9, 267–276.
Vercammen, D., Beyaert, R., Denecker, G., Goossens, V., Van Loo, G.,
Declercq, W., Grooten, J., Fiers, W., and Vandenabeele, P. (1998). Inhibition
of caspases increases the sensitivity of L929 cells to necrosis mediated by
tumor necrosis factor. J. Exp. Med. 187, 1477–1485.
Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc. 41
Developmental Cell
Omi and Lysosomes Mediate Alternative Cell Death
White, E. (2006). Mechanisms of apoptosis regulation by viral oncogenes in
infection and tumorigenesis. Cell Death Differ. 13, 1371–1377.
Youle, R.J., and Strasser, A. (2008). The BCL-2 protein family: opposing
activities that mediate cell death. Nat. Rev. Mol. Cell Biol. 9, 47–59.
Widlak, P., and Garrard, W.T. (2005). Discovery, regulation, and action of
the major apoptotic nucleases DFF40/CAD and endonuclease G. J. Cell.
Biochem. 94, 1078–1087.
Yuan, J., and Kroemer, G. (2010). Alternative cell death mechanisms in
development and beyond. Genes Dev. 24, 2592–2602.
Wyllie, A.H., Kerr, J.F., and Currie, A.R. (1980). Cell death: the significance of
apoptosis. Int. Rev. Cytol. 68, 251–306.
Yun, J., Cao, J.H., Dodson, M.W., Clark, I.E., Kapahi, P., Chowdhury, R.B., and
Guo, M. (2008). Loss-of-function analysis suggests that Omi/HtrA2 is not an
essential component of the PINK1/PARKIN pathway in vivo. J. Neurosci. 28,
14500–14510.
Xu, D., Li, Y., Arcaro, M., Lackey, M., and Bergmann, A. (2005). The CARDcarrying caspase Dronc is essential for most, but not all, developmental cell
death in Drosophila. Development 132, 2125–2134.
Zou, H., Henzel, W.J., Liu, X., Lutschg, A., and Wang, X. (1997). Apaf-1,
a human protein homologous to C. elegans CED-4, participates in cytochrome
c-dependent activation of caspase-3. Cell 90, 405–413.
42 Developmental Cell 25, 29–42, April 15, 2013 ª2013 Elsevier Inc.
Developmental Cell, Volume 25
Supplemental Information
Alternative Germ Cell Death Pathway
in Drosophila Involves HtrA2/Omi,
Lysosomes, and a Caspase-9 Counterpart
Keren Yacobi-Sharon, Yuval Namdar, and Eli Arama
Inventory of Supplementary Materials:
Figure S1, related to Figure 1. bag of marbles (bam) and nanos (nos) expression domains
in the Drosophila adult testis
Figure S2, related to Figure 3. TEM analysis of apoptotic spermatogonial cysts
Figure S3, related to Figure 4. Global cellular acidification, and not macroautophagy, is
associated with GCD
Figure S4, related to Figure 5. Epistatic relationships between the mitochondrial and
lysosomal pathways during GCD
Figure S5, related to Figure 7. The mitochondrial-associated components, Pink1, the
Bcl-2 family proteins, Debcl and Buffy, and EndoG are all involved in GCD
Supplemental Experimental Procedures
Supplemental References
1
SUPPLEMENTAL INFORMATION
Figure S1. bag of marbles (bam) and nanos (nos) expression domains in the
Drosophila adult testis, related to Figure 1
(A) bam-Gal4 and (B) nos-Gal4 driver lines were each crossed to a fly line carrying a
UAS-GFP transgene to reveal their respective patterns of expression in the adult testis.
Note that nos expression is confined to spermatogonial cysts at the 2-8-cell stages,
whereas the bam is expressed in 8-16-cell cysts. Testes were counterstained with DAPI
(blue). Scale bar 20 m.
2
3
Figure S2. TEM analysis of apoptotic spermatogonial cysts, related to Figure 3
(A-D’’) Electron micrographs of cross sections through the apical tip of adult testes from
(A and A’) wild-type and (B-D’’) flies expressing the diap1 RNAi transgene in
spermatogonial cells by the bam-Gal4 driver line. Areas confined by yellowish squares
and marked by yellow letters were magnified and presented in separated micrographs
with corresponding letters.
(A and A’) A normal wild-type spermatogonium containing mitochondria of similar
width but with different lengths (yellow asterisks mark some of the elongated
mitochondria). Note the fine and almost invisible cristae.
(B, B’ and D’) Apoptotic spermatogonia display regular mitochondrial morphology
(yellow asterisks mark some of the mitochondria).
(B) An apoptotic spermatogonia exhibiting highly crenellated nuclei with some
fragmentation (magenta asterisks).
(C and C’) Apoptotically dying spermatogonial cyst with significantly shrunken cells (a
magenta arrow; note the dark color of these cells), displaying condensed and fragmented
chromatin (green arrowheads).
(D) The detachment of the apoptotic spermatogonia from their surrounding cells is highly
pronounced (blue asterisks mark empty spaces).
(D’) Membrane blebbing is often detected in circumference of the dying cells (light blue
arrowheads).
(D’’) Apoptotically dying spermatogonia are eventually disintegrated to apoptotic bodies
(red arrows; the blue arrow points at a newly generated apoptotic body confined within a
membrane.
4
(E) Spontaneously dying spermatogonial cysts are engulfed as single entities by
phagocytes. Electron micrograph of cross section through the apical tip of wild-type adult
testis. At least two dying spermatogonial cysts at advanced GCD stages (red and yellow
asterisks) are engulfed by the same macrophage.
(F and G) TUNEL labeling (red) of testes expressing GFP (green) under the control of
the phagocyte-specific driver, croquemort (crq)-Gal4. The nuclei are in blue (DAPI).
Professional phagocytes are visualized, which engulfed several dying TUNEL-positive
spermatogonia found at advanced disintegration states (white arrows).
Scale bars in A, 2 m; B and C’, 1 m; C, D and E, 5 m; A’, B’, D’ and D’’, 0.5 m; F
and G, 20 m.
5
6
Figure S3. Global cellular acidification, and not macroautophagy, is associated with
GCD, related to Figure 4
(A) Quantification of GCD levels in a mutant of atg8 (atg8KG07569) and a null allelic
mutant combination of atg7 (atg7d14/d77), and in a separate experiment, flies expressing
GFP (serving as control) or a dominant negative form of Atg1 (Atg1KQ) in spermatogonial
cells using the nos-Gal4 driver line. All calculations were performed as in Figure 2C.
(B and C) Apical tips of testes expressing either (B) human or (C) Drosophila Atg8based autophagy reporters in spermatogonial cells using the nos-Gal4 driver line. Testes
were stained with DAPI (blue) to reveal the nuclei, (B) GFP or (C) eGFP (green), which
mark the LC3/Atg8 subcellular localization, and either (B) TUNEL or (C) mCherry (red).
The positions of dying cysts are outlines by white circles. Note the quenching of eGFP in
contrast to the persistence of the mCherry throughout the dying cyst, suggesting global
acidification of the cellular compartment. Arrowheads point at two living 8-cell
spermatogonial cysts which display some puncta. However, no puncta was detected in
dying cysts.
(D and E) Nonuniform/incomplete DNA fragmentation during GCD in dnaseII mutants.
Apical tips of testes from dnaseII mutants stained to visualize DNA fragmentation
(TUNEL, red) and the nuclei (DAPI, blue). White arrowheads points to dying
spermatogonial cysts during early-mid GCD stages (nuclei are still readily visible) with
only partially fragmented DNA. Yellow arrowheads points to dying spermatogonial cysts
during mid-late GCD stages (note that the DAPI signal is fading out) displaying diffused
TUNEL labeling. Scale bars 20 m.
7
Figure S4. Epistatic relationships between the mitochondrial and lysosomal
pathways during GCD, related to Figure 5
(A) Quantification of GCD levels in testes from the indicated genotypes. All calculations
were performed as in Figure 2C.
(B) Quantification of LysoTracker-positive spermatogonial cysts in testes from the
indicated genotypes. All calculations were performed as in Figure 2C.
8
9
Figure S5. The mitochondrial-associated components, Pink1, the Bcl-2 family
proteins, Debcl and Buffy, and EndoG are all involved in GCD, related to Figure 7
(A-C) Quantification of GCD levels in testes from the indicated genotypes. All
calculations were performed as in Figure 2C. (A) Note that GCD levels are not
significantly reduced in the parkin hemizygous mutant (park25/Df). (B) The fact that the
debcl homozygous mutants (debclE26) display a stronger effect than the hemizygous
mutants (debclE26/Df) may suggest either some background effect or an unforeseen
dominant effect of this mutant.
10
Supplemental Experimental Procedures
Fly Strains
yw flies were used as wild-type controls. Fly mutant alleles used in this study are
droncI24, droncI29 and droncL32 (Xu et al., 2005), arkN28 and arkP46 (Srivastava et al.,
2007), ark82 (Akdemir et al., 2006), dcp-1prev (Laundrie et al., 2003), strica4 (Baum et al.,
2007), driceΔ1 (Muro et al., 2006), cathD1 (Myllykangas et al., 2005), dor4 and car1
(Sriram et al., 2003), dnaseIIlo (Mukae et al., 2002), and omi
1
and omiR-1 (Tain et al.,
2009). The UAS-diap1-IR line was obtained from P. Meier (ICR, London), nos-Gal4 and
bam-Gal4 from L. Gilboa (WIS, Israel), UAS-myc-diap1 from H.D. Ryoo (NYU, NY),
and UAS-p35 and Df(3L)H99 from H. Steller (Rockefeller Univ., NY), atg7d14 and
atg7d77 (Juhasz et al., 2007), UAS-atg-1K38Q and atg8KG07569 (Scott et al., 2007), UASpeGFP-mCherry-DrAtg8a (Nezis et al., 2010), UAS-eGFP-huLC3 (Rusten et al., 2004),
pinkB9 (Park et al., 2006), park25 (Greene et al., 2003), crq-Gal4, UAS-GFP (Franc et al.,
1996), endoGMB07150 (Deluca and O'Farrell, 2012), debclE26 and buffyH37 (Sevrioukov et
al., 2007), and the Df(2R)ED1552, Df(2R)BSC259, Df(2R)BSC699 and Df(3L)Pc-MK
from the Bloomington Stock Center.
Expression Constructs
For generation of the omiR-2 rescue construct, a 1.97 kb genomic fragment spanning the
htrA2/omi locus was amplified by PCR from purified yw genomic DNA using the PFU
polymerase
(Stratagene),
the
GCGAATTCGTGTTGGTGGAAGGAAGA-3’,
11
forward
and
primer,
reverse
primer,
5’5’-
GCGGTACCGGGTTTGTCAGCGATTTC-3’, and cloned into the PattB vector using
the EcoRI and Acc65I restriction sites.
To generate the omiR-2(S266A) mutant form where the conserved serine residue in position
266 was replaced by an alanine residue, we performed standard site-directed mutagenesis
using
the
omiR-2
construct
as
a
template
and
the
forward
primer,
5’-
CGTTTGGCAACGCCGGCGGTCC-3’, and its complementary reverse primer.
To generate the omiR-2(-IBM) mutant form, the three putative IBM tetrapeptides in positions
74, 79 and 92 were simultaneously mutated by replacement of the first two amino acids
in each tetrapeptide, including the invariable first alanine, by two glycine residues. This
was done in two steps using the omiR-2 construct as a template. The tetrapeptide in
position
92
was
first
mutated
as
before
using
the
forward
primer,
5’-
CGCCAACGATCGCCGGTGGCAAAATGACCGGTCG-3’ and its complementary
reverse primer, and subsequently this construct was further mutated simultaneously in
positions
74
and
79
using
the
forward
primer,
5’-
GCCCTTCTCCCTGGGCGGTGGGGTCAGTGCGGGGGGCATACAGCGGGAAGAC
-3’, and its complementary reverse primer.
To generate the omiR-BAC rescue construct, a P[acman] BAC genomic clone (from a
genomic DNA library inserted in the pattB vector), spanning approximately 18.2 kb of
the genomic region encompassing the htrA2/omi locus (BAC# CH322-129B23), was
obtained from the BACPAC Resource Center [BPRC; Children's Hospital Oakland
Research Institute, Oakland, CA; (Venken et al., 2009)], and the DNA was purified from
the DH10B bacteria using the alkaline lysis miniprep method (according to
12
http://bacpac.chori.org/bacpacmini.htm). The clone was validated by restriction enzyme
digest and transformed into EPI300 bacteria (for the high copy number production step).
To generate transgenic flies, these (pattB) rescue constructs were all injected to embryos
bearing the defined attP40 landing site on chromosome 2. omiR-2, omiR-2(S266A), and omiR2(-IBM)
were injected by Genetic Services Inc. (Cambridge, MA), whereas omiR-BAC was
injected by BestGene Inc. (Chino Hills, CA).
RNA isolation and RT-PCR
Total RNA was extracted from twenty young adult male testes using the PureLink Microto-Midi Total RNA Purification System (Invitrogen) according to the manufacturer’s
recommendations. Purified RNA was reversed transcribed using Qiagen OneStep RT–
PCR kit (Qiagen). cDNA for HtrA2/Omi was amplified using forward primer: 5’GTGAGGCTATCCGATGGCAG-3’
and
a
reverse
primer:
5’-
TGAGTTCTTGATCTCCTTCTTG-3’. The primers were designed to amplify a 659 bplong cDNA fragment from wild-type (wt) and the omi transgenes (omitran), a 536 bp-long
truncated omi cDNA fragment from the omiDf1/IFD mutant (omiIFD), and a larger, 812 bplong fragment of omi genomic DNA, which includes two small introns (omigen).
13
Supplemental References
Akdemir,F., Farkas,R., Chen,P., Juhasz,G., Medved'ova,L., Sass,M., Wang,L., Wang,X.,
Chittaranjan,S., Gorski,S.M., Rodriguez,A., and Abrams,J.M. (2006). Autophagy occurs
upstream or parallel to the apoptosome during histolytic cell death. Development 133,
1457-1465.
Baum,J.S., Arama,E., Steller,H., and McCall,K. (2007). The Drosophila caspases Strica
and Dronc function redundantly in programmed cell death during oogenesis. Cell Death.
Differ. 14, 1508-1517.
Deluca,S.Z. and O'Farrell,P.H. (2012). Barriers to male transmission of mitochondrial
DNA in sperm development. Dev. Cell 22, 660-668.
Franc,N.C., Dimarcq,J.L., Lagueux,M., Hoffmann,J., and Ezekowitz,R.A. (1996).
Croquemort, a novel Drosophila hemocyte/macrophage receptor that recognizes
apoptotic cells. Immunity. 4, 431-443.
Greene,J.C., Whitworth,A.J., Kuo,I., Andrews,L.A., Feany,M.B., and Pallanck,L.J.
(2003). Mitochondrial pathology and apoptotic muscle degeneration in Drosophila parkin
mutants. Proc. Natl. Acad. Sci. U. S. A 100, 4078-4083.
Juhasz,G., Erdi,B., Sass,M., and Neufeld,T.P. (2007). Atg7-dependent autophagy
promotes neuronal health, stress tolerance, and longevity but is dispensable for
metamorphosis in Drosophila. Genes Dev. 21, 3061-3066.
Laundrie,B., Peterson,J.S., Baum,J.S., Chang,J.C., Fileppo,D., Thompson,S.R., and
McCall,K. (2003). Germline cell death is inhibited by P-element insertions disrupting the
dcp-1/pita nested gene pair in Drosophila. Genetics 165, 1881-1888.
Mukae,N., Yokoyama,H., Yokokura,T., Sakoyama,Y., and Nagata,S. (2002). Activation
of the innate immunity in Drosophila by endogenous chromosomal DNA that escaped
apoptotic degradation. Genes Dev. 16, 2662-2671.
Muro,I., Berry,D.L., Huh,J.R., Chen,C.H., Huang,H., Yoo,S.J., Guo,M., Baehrecke,E.H.,
and Hay,B.A. (2006). The Drosophila caspase Ice is important for many apoptotic cell
deaths and for spermatid individualization, a nonapoptotic process. Development 133,
3305-3315.
Myllykangas,L., Tyynela,J., Page-McCaw,A., Rubin,G.M., Haltia,M.J., and Feany,M.B.
(2005). Cathepsin D-deficient Drosophila recapitulate the key features of neuronal ceroid
lipofuscinoses. Neurobiol. Dis. 19, 194-199.
Nezis,I.P., Shravage,B.V., Sagona,A.P., Lamark,T., Bjorkoy,G., Johansen,T.,
Rusten,T.E., Brech,A., Baehrecke,E.H., and Stenmark,H. (2010). Autophagic degradation
of dBruce controls DNA fragmentation in nurse cells during late Drosophila
melanogaster oogenesis. J Cell Biol 190, 523-531.
14
Park,J., Lee,S.B., Lee,S., Kim,Y., Song,S., Kim,S., Bae,E., Kim,J., Shong,M., Kim,J.M.,
and Chung,J. (2006). Mitochondrial dysfunction in Drosophila PINK1 mutants is
complemented by parkin. Nature.
Rusten,T.E., Lindmo,K., Juhasz,G., Sass,M., Seglen,P.O., Brech,A., and Stenmark,H.
(2004). Programmed autophagy in the Drosophila fat body is induced by ecdysone
through regulation of the PI3K pathway. Dev. Cell. 7, 179-192.
Scott,R.C., Juhasz,G., and Neufeld,T.P. (2007). Direct induction of autophagy by Atg1
inhibits cell growth and induces apoptotic cell death. Curr. Biol. 17, 1-11.
Sevrioukov,E.A., Burr,J., Huang,E.W., Assi,H.H., Monserrate,J.P., Purves,D.C.,
Wu,J.N., Song,E.J., and Brachmann,C.B. (2007). Drosophila Bcl-2 proteins participate in
stress-induced apoptosis, but are not required for normal development. Genesis. 45, 184193.
Sriram,V., Krishnan,K.S., and Mayor,S. (2003). deep-orange and carnation define
distinct stages in late endosomal biogenesis in Drosophila melanogaster. J Cell Biol 161,
593-607.
Srivastava,M., Scherr,H., Lackey,M., Xu,D., Chen,Z., Lu,J., and Bergmann,A. (2007).
ARK, the Apaf-1 related killer in Drosophila, requires diverse domains for its apoptotic
activity. Cell Death. Differ. 14, 92-102.
Tain,L.S., Chowdhury,R.B., Tao,R.N., Plun-Favreau,H., Moisoi,N., Martins,L.M.,
Downward,J., Whitworth,A.J., and Tapon,N. (2009). Drosophila HtrA2 is dispensable for
apoptosis but acts downstream of PINK1 independently from Parkin. Cell Death. Differ.
16, 1118-1125.
Venken,K.J., Carlson,J.W., Schulze,K.L., Pan,H., He,Y., Spokony,R., Wan,K.H.,
Koriabine,M., de Jong,P.J., White,K.P., Bellen,H.J., and Hoskins,R.A. (2009). Versatile
P[acman] BAC libraries for transgenesis studies in Drosophila melanogaster. Nat.
Methods 6, 431-434.
Xu,D., Li,Y., Arcaro,M., Lackey,M., and Bergmann,A. (2005). The CARD-carrying
caspase Dronc is essential for most, but not all, developmental cell death in Drosophila.
Development 132, 2125-2134.
15