922 Cell 141, June 11, 2010 ©2010 Elsevier Inc.
Bernèche, S., and Roux, B. (2005). Structure 13,
Clarke, O.B., Cputuo, A.T., Hill, A.P., Vandenberg,
J.I., Smith, B.J., and Gulbis, J.M. (2010). Cell, this
Doyle, D.A., Morais Cabral, J., Pfuetzner, R.A.,
Kuo, A., Gulbis, J.M., Cohen, S.L., Chait, B.T., and
MacKinnon, R. (1998). Science 280, 69–77.
Hoshi, T., Zagotta, W.N., and Aldrich, R.W. (1991).
Neuron 7, 547–556.
Jan, L.Y., and Jan, Y.N. (1997). J. Physiol. 505,
Kuo, A., Gulbis, J.M., Antcliff, J.F., Rahman, T.,
Lowe, E.D., Zimmer, J., Cuthbertson, J., Ashcroft,
F.M., Ezaki, T., and Doyle, D.A. (2003). Science
Lu, Z. (2004). Annu. Rev. Physiol. 66, 103–129.
Nishida, M., and MacKinnon, R. (2002). Cell 111,
Suh, B.C., and Hille, B. (2008). Annu Rev Biophys
Autophagy, or “self-eating,” is a catabolic
process that degrades and recycles cyto-
plasmic contents. Pioneering studies in
the single-celled yeast Saccharomyces
cerevisiae identified a suite of autophagy
(Atg) genes required for survival during
starvation (Mizushima, 2007). Although
many of these genes are functionally con-
served from yeast to mammals, autophagy
is probably more complex in multicellular
animals and most likely requires factors
that are absent in yeast. For example, ani-
mal tissues maintain homeostasis when
nutrients are locally restricted by trading
off metabolic and catabolic processes,
and this may be one reason that cancer
cells with altered metabolism display
elevated levels of autophagy (Mathew et
al., 2007). However, little is known about
autophagy machinery specific to animals.
Now in a tour de force study, Tian et al.
(2010) identify four previously unchar-
acterized genes specifically required for
autophagy in multicellular animals and
establish Caenorhabditis elegans as one
of the premier genetic models for uncov-
ering new autophagy genes in animals.
During autophagy, cytoplasmic con-
tents, such as proteins and organelles,
are engulfed by a double-membrane
autophagosome (Figure 1), which then
fuses with lysosomes to form autolyso-
somes. Here hydrolase enzymes degrade
the cargo, and the products are sub-
sequently released into the cytosol for
reuse (Mizushima, 2007). Besides recy-
cling cytoplasmic material during periods
of starvation or stress, autophagy (also
called macroautophagy) clears protein
aggregates, eliminates pathogens, and
influences cell death. Moreover, in many
organisms, autophagy defects are asso-
ciated with decreased life span, neuro-
degeneration, and tumor progression
(Mizushima et al., 2008).
In worms (C. elegans), flies, and
mammals, autophagy is also important
during development (Meléndez and
Neufeld, 2008). In C. elegans, germ cells
contain aggregates of protein and RNA
known as P granules, which are absent
in somatic cells. A previous study dem-
onstrated that autophagy is required for
clearing the aggregate-prone compo-
nents of P granules from somatic cells in
developing C. elegans embryos (Zhang
et al., 2009), and defects in autophagy
lead to the aberrant accumulation of
aggregates of P granule proteins in
Now Tian et al. (2010) use the persis-
tence of P granule proteins in somatic
cells to find mutant C. elegans embryos
with defects in autophagy. From the
?160 mutants identified, the authors
isolated four new genes, named epg-2,
-3, -4, and -5 (ectopic PGL granules),
which do not map to known autophagy
genes. The coiled-coil protein, epg-2,
mediates recognition of cargo (e.g.,
aggregates of P granule proteins)
for delivery to autophagosomes and
appears to be specific to nematodes.
The other three genes, epg-3, -4, and
-5, are also required for starvation-
induced autophagy. They are con-
served genetically from worms to
mammals and appear to lack homologs
In addition, the authors isolated numer-
ous new mutations in genes homolo-
gous to yeast autophagy genes, which
validate and strengthen the results of the
study. Not only do these new mutations
provide a valuable resource for probing
the structure and function of autophagy
proteins, but they also establish C. ele-
gans as a preeminent system for study-
ing the role and regulation of autophagy
in multicellular animals.
Autophagy shows Its Animal side
Christina K. McPhee1 and Eric H. Baehrecke1,*
1Department of Cancer Biology, University of Massachusetts Medical School, Worcester, MA 01605, USA
Most autophagy genes have been discovered in the single-celled yeast Saccharomyces cerevi-
siae, and little is known about autophagy genes that are specific to multicellular animals. In this
issue, Tian et al. (2010) now identify four new autophagy genes: one specific to the nematode
Caenorhabditis elegans and three conserved from worms to mammals.
Cell 141, June 11, 2010 ©2010 Elsevier Inc. 923
Tian and colleagues found that
EPG-3 is similar in sequence and func-
tion to the human vacuolar membrane
protein 1, VMP1. Expression of human
VMP1 remarkably rescues the P granule
degradation defect in worm embryos
with mutations in epg-3, and VMP1 is
required for autophagy during starva-
tion. In yeast, autophagosomes form
at a specific cellular location called the
pre-autophagosomal structure. Mam-
malian cells do not contain a clearly
defined pre-autophagosomal structure.
Instead, autophagosomes form from
called isolation membranes, which
nucleate at multiple sites in the cytosol,
including endoplasmic reticulum (ER)-
derived structures termed omegasomes
(Figure 1) (Axe et al., 2008). Defects in
epg-3 or VMP1 lead to the accumulation
of omegasomes in C. elegans embryos
or rat kidney cells, respectively. Thus,
although the exact roles of EPG-3
and VMP1 remain unknown, both pro-
teins must function at an early step in
autophagy. Perhaps EPG-3 and VMP1
facilitate the elongation of the isolation
membrane or the closure of the double-
membrane vesicle during autophago-
some assembly (Figure 1).
Isolation membranes and omegas-
omes also accumulate in embryos with
mutations in epg-4. Consequently,
EPG-4 probably also functions in an
early step of autophagosome forma-
tion. EPG-4 localizes to the ER, sug-
gesting that it helps to convert ER
membranes to autophagic membranes.
In contrast, reducing the expression of
EI24, the mammalian homolog of epg-4,
does not affect omegasome formation
in mammalian cells but instead results
in the accumulation of autolysosomes
that fail to degrade their contents. This
suggests that EI24 functions later in
the autophagy pathway than epg-4.
The apparent phenotypic differences
between defects in epg-4 and EI24 may
be due to inefficient silencing of EI24,
or these homologs may have diverged
functionally during evolution.
Mutations in epg-5 lead to the persis-
tent colocalization of P granule aggre-
gates with protein markers known to
associate with autophagosomes (Figure
1). Thus, although protein aggregates
are near the autophagic machinery in
these mutant embryos, the aggregates
are probably not properly degraded.
Mutations in other autophagy genes (i.e.,
atg-3, atg-13, or atg-5) suppressed the
epg-5 phenotype, and these epistasis
analyses suggest that epg-5 acts down-
stream of genes that regulate autopha-
As with EI24, silencing the epg-5
mammalian homolog mEPG-5 led to
the persistence of autolysosomes that
fail to degrade their contents. Trans-
mission electron microscopy images
revealed significant differences in the
ultrastructures of the autolysomes pres-
ent in embryos with reduced levels of
EI24 and mEPG-5. Thus, future studies
are needed to determine if these genes
function in different steps in the degra-
dation of autophagosome cargo.
In this landmark study, Tian et al. define
discrete steps in the autophagy pathway
that are specific to multicellular animals.
Defects in autophagy are associated
with numerous pathological conditions,
including aging, neurodegeneration, and
cancer (Mizushima et al., 2008). There-
fore, it is interesting that the mammalian
homologs of EPG-3, -4, and -5 are all
associated with either human diseases
or models of human disease. VMP1 is
highly expressed in the pancreas of rats
with acute pancreatitis (Dusetti et al.,
2002), and it will be interesting to deter-
mine if VMP1 specifically functions in
autophagy in the pancreas. EI24 expres-
sion is activated by tumor suppressor
p53 and by etoposide, a chemotherapy
drug that activates p53 (Gu et al., 2000).
It could be that EI24 functions in a p53-
independent or -dependent process.
Notably, mEPG-5 is altered in human
figure 1. Genes Required for Autophagy in
Autophagy degrades and recycles cytoplasmic
contents. For example, during the development
of Caenorhabditis elegans embryos, autophagy
clears from somatic cells aggregates of proteins
and RNA molecules known as P granules. Upon
autophagy induction, isolation membranes nucle-
ate at structures derived from the endoplasmic
reticulum (ER) called omegasomes. As an isola-
tion membrane expands, it engulfs P granules and
then closes up to produce an autophagosome.
Lysosomes fuse with the outer membrane of an
autophagosome to form an autolysosome, where
hydrolase enzymes degrade the inner membrane
and cargo. Degraded cargo is then released into
the cytosol for reuse. Using a genetic screen, Tian
et al. (2010) now uncover four new genes in C. el-
egans (epg-2, -3, -4, and -5), which are required
specifically for autophagy in animals. Whereas
epg-2 is required for recognizing cargo, muta-
tions in epg-3 and epg-4 lead to the accumula-
tion of isolation membranes and omegasomes.
Embryos with mutations in epg-5 accumulate
autolysosomes that fail to degrade cargo. Strik-
ingly, mammalian homologs of epg-3,-4, and -5
are also required for autophagy in cell cultures.
VMP1 functions at an early step of autophago-
some formation, whereas mEPG-5 and EI24 act at
924 Cell 141, June 11, 2010 ©2010 Elsevier Inc.
breast tumors (Sjöblom et al., 2006).
Therefore, EI24 and mEPG5 may specifi-
cally regulate autophagy in cancer cells.
The identification of these new genes by
Tian et al. (2010) highlights the impor-
tance of autophagy in human diseases
and may lead to exciting new discover-
ies about the role of autophagy in cancer
and other disorders.
Our work on autophagy is supported by the NIH
Axe, E.L., Walker, S.A., Manifava, M., Chandra, P.,
Roderick, H.L., Habermann, A., Griffiths, G., and
Ktistakis, N.T. (2008). J. Cell Biol. 182, 685–701.
Dusetti, N.J., Jiang, Y., Vaccaro, M.I., Tomasini, R.,
Azizi Samir, A., Calvo, E.L., Ropolo, A., Fiedler, F.,
Mallo, G.V., Dagorn, J.C., et al. (2002). Biochem.
Biophys. Res. Commun. 290, 641–649.
Gu, Z., Flemington, C., Chittenden, T., and Zam-
betti, G.P. (2000). Mol. Cell. Biol. 20, 233–241.
Mathew, R., Karantza-Wadsworth, V., and White,
E. (2007). Nat. Rev. Cancer 7, 961–967.
Meléndez, A., and Neufeld, T.P. (2008). Develop-
ment 135, 2347–2360.
Mizushima, N. (2007). Genes Dev. 21, 2861–2873.
Mizushima, N., Levine, B., Cuervo, A.M., and
Klionsky, D.J. (2008). Nature 451, 1069–1075.
Sjöblom, T., Jones, S., Wood, L.D., Parsons, D.W.,
Lin, J., Barber, T.D., Mandelker, D., Leary, R.J.,
Ptak, J., Silliman, N., et al. (2006). Science 314,
Tian, Y., Li, Z., Hu, W., Ren, H., Tian, E., Zhao, Y.,
Lu, Q., Huang, X., Yang, P., Li, X., et al. (2010). Cell,
Zhang, Y., Yan, L., Zhou, Z., Yang, P., Tian, E.,
Zhang, K., Zhao, Y., Li, Z., Song, B., Han, J., et al.
(2009). Cell 136, 308–321.
The chromatin domains that flank DNA
double-strand breaks (DSBs) harbor
a plethora of posttranslational protein
modifications. Although these modifica-
tions decorate megabase-size regions
and are generally thought to promote
DNA repair and cell survival, the func-
tional roles of many remain to be deter-
mined, among them monoubiquitinated
histone 2A (uH2A). Stemming from previ-
ous studies that implicate uH2A in tran-
scriptional silencing (Weake and Work-
man, 2008), Greenberg and colleagues
now examine whether uH2A may also
exert similar gene silencing activities
near sites of DNA damage (Shanbhag et
To do this, the authors borrow a previ-
ously described transcriptional reporter
(Janicki et al., 2004) and re-engineer
it so that a defined DSB can be gener-
ated at a stretch of sequence adjacent
to the transcription unit. By employing
fluorescence-based designs, the sys-
tem makes it possible to simultaneously
observe, both qualitatively and quanti-
tatively, nascent transcription, protein
production, as well as DNA-damage
responses—all at the single-cell level.
Introduction of DSBs not only dis-
rupts the physical integrity of interphase
chromatin but is thought to interrupt
numerous processes that take place at
this dynamic structure. Whereas DSBs
appear to inhibit DNA replication by pre-
venting global origin firing and slowing
the progression of local replication forks,
it is not known whether and how these
DNA lesions modulate local transcrip-
tion. Now using this experimental setup,
Shanbhag et al. (2010) address this
question by measuring transcriptional
activities adjacent to the engineered
DSB site. They find that transcriptional
activities at the chromosomally inte-
grated reporter are largely repressed
when a DSB is introduced. What’s more
interesting is that this DSB-associated
gene silencing response is only effective
on regions of chromatin proximal to the
lesion and does not affect transcription
at distal sites.
The authors call this phenomenon DNA
double-strand break-induced silencing
in cis (DISC), and they uncover a strict
requirement for the ataxia telangiecta-
sia mutated (ATM) kinase in mediating
DISC. Notably, DISC coincides with two
hallmarks of transcriptional repression:
stalling of RNA polymerase II (indicated
by hypophosphorylation) and impaired
to the notion that DISC affects through
Atm creates a Veil of transcriptional silence
Michael S.Y. Huen1,2,3 and Junjie Chen4,*
1Genome Stability Research Laboratory
2Department of Anatomy
3Centre for Cancer Research
The University of Hong Kong, L1, Laboratory Block, 21 Sassoon Road, Hong Kong S.A.R.
4Department of Experimental Radiation Oncology, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Boulevard,
Houston, TX 77030, USA
The ATM kinase orchestrates diverse responses to DNA damage. By simultaneously monitoring
transcription and DNA-damage responses in single cells, Shanbhag et al. (2010) now uncover a
role of ATM in preventing transcription near DNA double-strand breaks.