Featured Science Watch® Newsletter
literally means "self-eating," which seems like a
particularly gruesome brand of cannibalism. As it turns
out, however, we all do it. Just wait long enough
between meals and we start to consume our own fat
tissue, and if we wait even longer, we start to live
off our muscle as well. Researchers have known for half
a century that cells do it, too. They surround
defective proteins with double-membraned vesicles and
then drag them off to be broken down and the parts
reused, but for many, this phenomenon was thought to be
a mechanism for clearing away garbage and little more.
All this changed in the last decade, as defects in autophagy were linked to
a host of diseases from cancer and neurodegeneration to microbial infection
and even aging. Among the leaders of this autophagy revolution is the
University of Michigan cell biologist Daniel Klionsky, who, according to a
recent extraction from the
Hot Papers Database, has coauthored five highly cited reports on
the topic in the last two years. One of these, a Nature review
from February 2008, "Autophagy fights disease through cellular
Cuervo) has been cited nearly 150 times in the year and half since
then—more than 100 times in the last six months alone.
Klionsky’s 2000 review in Science, "Autophagy as a regulated
pathway of cellular degradation," co-authored with his post-doctoral
Emr (link: article# 2) of Caltech (now at Cornell), has been cited over
600 times (see adjoining table, paper #1). It’s one of over 20
Klionsky papers published in the last decade that have garnered more than
100 citations each.
Klionsky, 50, received his bachelor’s degree, magna cum laude, from
UCLA in 1980, and went on to Stanford where he obtained his Ph.D. in
biological sciences in 1986. He spent the next four years as a
post-doctoral fellow in Emr’s lab at Caltech until moving, in1990, to
the University of California, Davis, where he eventually became a full
professor of microbiology. In 2000, he left California for the University
of Michigan, where he now holds joint appointments in the Life Sciences
Institute, the Department of Molecular, Cellular and Developmental Biology,
and the Medical School.
Klionsky spoke to Science
Watch from his office in Ann Arbor.
What prompted your research on autophagy,
considering that it wasn’t exactly mainstream science when you
began in the early 1990s?
When I was starting my career at UC Davis, I had a new post-doc who
came from a laboratory in Spain that had cloned a particular gene
encoding a vacuolar enzyme; this was before the yeast genome was
sequenced. I had been interested in this gene when I was a post-doc
studying vacuolar protein targeting but had never pursued it. For a new
faculty member, it was an easy project to begin with, but it turned out
that the gene product didn’t go to the vacuole by the normal
What’s the normal pathway?
At that time, we thought that all proteins going to the vacuole first
enter the endoplasmic reticulum and then go from there to other
locations inside vesicles that bud off from those
organelles—first to the Golgi complex, then usually to the
endosome, and then to the vacuole. This is part of the secretory
pathway, which is also used to get proteins out of the cell. The
protein we were looking at is called aminopeptidase I, and it turned
out that it didn’t go through the secretory pathway. So we
figured we were looking at a new pathway to the vacuole that no one had
ever shown before. We got mutants, because we were working in
yeast—I’m still working in yeast—and it’s easy
to do genetics. And one of the things we did is compare our mutants
with other mutants that are already published in the literature to see
if they’re novel. What I found was that these mutants overlapped
almost 100% with mutants from two other labs—one in Germany and
one in Japan—that were looking at autophagy.
Is 100% overlap good or bad?
Cited Papers by
Daniel J. Klionsky and
(Ranked by total citations)
D.J. Klionsky, S.D. Emr, "Autophagy as
a regulated pathway of cellular
290(5497): 1717-21, 2000.
B. Levine, D.J. Klionsky, "Development
by self-digestion: Molecular mechanisms
and biological functions of autophagy,"
Develop. Cell, 6(4): 463-77,
T. Shintani, D.J. Klionsky, "Autophagy
in health and disease: A double-edged
sword," Science, 306(5698):
D.J. Klionsky, et al., "A
unified nomenclature for yeast
autophagy-related genes," Develop.
Cell, 5(4): 539-45, 2003.
D.J. Klionsky, Y. Ohsumi, "Vacuolar
import of proteins and organelles from
the cytoplasm," Ann. Rev. Cell
Develop. Bio., 15: 1-32, 1999.
I was shocked and, I have to say, not very happy about it, for two
reasons: one, it obviously meant that I was not alone in looking at
this set of mutants; and, two, at that time autophagy was pretty much
considered a garbage pathway. The thought was that aberrantly folded or
nonfunctional proteins were taken to the vacuole or lysosome to be
degraded—sort of garbage-recycling pathways. Who cares about
that? Garbage pickup is important, but it is not what I planned to
study—my interest was in protein targeting. So at that point
I’m studying autophagy and thinking, my goodness, this is
terrible. This was the mid-1990s. None of the genes had been cloned at
the time. Autophagy itself had been studied in mammalian cells since
the 1950s, but not a single gene had been identified, because it is
relatively difficult to do genetics in that system. As far as the
molecular aspects then, this whole field was going nowhere, which was
not exciting to someone who does a lot of molecular genetics.
So what did you do?
Well, all three labs—the German lab, the Japanese lab, and
mine—started to clone the genes, and it was pretty amazing how
rapidly it developed. Literally in just over ten years the field
exploded, even in yeast, with the identification of 32 genes in this
pathway and the discovery, in effect, of a pathway that was almost
completely unknown, at least in terms of the molecular aspects. Because
of the timing of when we did this work—before the entire genome
of yeast was sequenced—this may be one of the last examples where
several large-scale screens revealed an entirely unknown molecular
pathway. In 1997, the first gene was published for autophagy in yeast.
That was by our collaborator/competitor, Yoshinori Ohsumi. Two years
later Beth Levine published the first paper showing a connection
between autophagy and disease—in this case, cancer.
How is cancer connected to autophagy, and how
did this one discovery affect the evolution of the field?
Mice that have a mutation in one of the autophagy genes show a much
higher rate of spontaneous tumor formation. And that, of course,
attracted considerable attention. People in the field were thrilled to
see this sort of connection between autophagy and human disease. Since
then, we’ve continued to see this amazing networking of
connections. Autophagy is involved in the elimination of certain
invasive bacteria and viruses from host cells; it’s involved in
protection against certain types of neurodegeneration; it plays a major
role in cellular remodeling in response to various environmental
changes or stresses; it is induced by starvation, by low oxygen, and by
reactive oxygen species. Because of the unique membrane dynamics of
this process, the cell has the capacity, through autophagy, to actually
sequester essentially any-sized cargo, which means that this is the
only process in which you can enwrap an entire organelle in a membrane
and deliver it to the lysosome or vacuole for degradation.
Do you understand the mechanism linking
autophagy to cancer, or is it still a black box?
It’s not so clear what’s going on there. One possibility is
based on work from Eileen White’s lab. Autophagy is basically a
protective mechanism for cells, one that’s induced if cells are
stressed. If autophagy is defective and then the cell is stressed,
it’s possible that there’s now selective pressure for
hypermutation. So, many of the cells may die, but those cell mutants
that survive may undergo a change that allows them to survive the
stress, but they may also undergo a change that causes them to become
How did the research evolve after
Levine’s discovery and the genetic work?
Once the genes started to come out from the yeast work, people who were
working in these various other fields started to ask if it’s
possible that this pathway is playing a role in what they were
studying. For example, what happens if we either induce this process or
block it genetically or with a chemical inhibitor? So there was this
explosion because of all the other fields now looking at autophagy
connections, especially with regard to human disease.
You’ve noted in your papers that
autophagy is turned on during starvation. What’s the
The idea there is simple. It’s similar to what happens during
starvation on an organismal level. If you fast for long enough, your
body first starts to break down your fat reserves and then your protein
reserves. If you’re not taking in any nutrients, you have to keep
the essential processes going. The same thing happens on a cellular
level. If the cell is not getting enough food for a certain threshold
of time, it needs to cannibalize parts of itself; it takes up cytoplasm
non-selectively and delivers it to the vacuole or lysosome, where it
can be broken down and the products released back into the cytosol to
be reused for new synthesis. The cell can make essential proteins by
reusing parts of itself and thereby survive the starvation condition.
During starvation most new synthesis is shut down, but the cell will
make things that are essential, and autophagy provides it with the
resources to do that.
So the cell is effectively eating what it can
afford to eat of itself until more nutrients come along from
Exactly. In yeast, if we starve wild-type cells for a month or so, they
can do pretty well. If they’re autophagy-defective, they’ll
die within one or two days. That’s a huge difference in life
Can you describe the link between autophagy and
neurodegeneration, and perhaps what the mechanisms are?
Many people are now working with autophagy in neurodegeneration, either
looking for additional ways to regulate the process or to understand
how it’s functioning, for example, in neurons. It appears that
autophagy may be protecting neuronal cells from the buildup of certain
proteins that become toxic. It’s become clear that in some cases
autophagy is definitely protecting us from neurodegeneration.
It’s not crazy to imagine that in some not-too-distant future
we’ll be able to turn on autophagy in a neuron-specific manner.
This might be beneficial, particularly in people who are genetically
predisposed to a neurodegenerative disease. It might, at least, delay
the onset of symptoms.
What do you think are the big questions in
autophagy research that still need to be answered?
"There was this explosion
because of all the other fields now
looking at autophagy connections,
especially with regard to human
disease," says Daniel Klionsky of
the University of Michigan, Ann
You can ask that question at two levels. One is the basic-science
level. There we have questions such as what is the membrane source for
forming the sequestering vesicle, called an autophagosome? No one knows
for sure where that membrane comes from and how it actually forms. This
is a major question in the field. Another one is how this process is
regulated. You don’t want to have too little autophagy, but you
can imagine that too much is also a problem. If you eat too much of
yourself, you’ll probably die. So it has to be regulated pretty
tightly, and it’s not known in detail how that happens. We now
know 32 proteins that are involved primarily or exclusively in
autophagy, but for most of them the functions have not been
At the other, clinical, level there are huge questions to be answered.
A lot of people are now trying to see if they can manipulate autophagy
as an anti-cancer treatment. But here’s the problem: it cuts both
ways. In other words, autophagy is generally a cytoprotective
mechanism—meaning that it’s cell protective. But
that’s true not just for healthy cells that need it to survive
starvation, but also for cancer cells. So it appears that some type of
cancer cells induce autophagy to help them survive. In these cases, we
may want to shut off autophagy to promote the death of the cancer cells
during treatment. In other cases, we might want to induce autophagy to
push cancer cells over the edge and kill them during treatment.
The problem is that cancer is so heterogeneous—even within an
individual, the cancer cells are not uniform—so we don’t
know when one method might work and when the other might. At the moment
a lot of that work is being done empirically. If we treat with a given
anticancer drug and then with another drug that induces autophagy, is
the combined effect beneficial or not?
What questions are you focusing on in your own
One of the main things we’re focusing on is this question of
regulation. This has been largely a black box. People have known for
many years, for example, that TOR, a kinase, is a negative regulator of
autophagy. All we know, however, is that it’s involved in
regulation. Other than that, there are huge unknowns. We’re doing
a lot of work on that. We’d like to be able to reconstitute the
autophagy process—to really determine what happens step by step.
There’s no system right now to do the whole thing in a test tube,
but we think we can at least reconstitute the individual steps to
determine exactly what proteins are needed at each step and what those
proteins are doing. Right now there are just many unanswered questions
in terms of the mechanisms of this process.
We’ve also recently started on a new area, studying a subset of
autophagy called mitophagy: selective mitochondrial
degradation. Autophagy can be nonselective, which is what we’ve
been talking about so far, but it can also be selective. If you have
damaged mitochondria, for instance, those can be selectively targeted
by an autophagic-like mechanism. This is what we call mitophagy, and
it’s very interesting because a lot of diseases are associated
with mitochondrial dysfunction, and it’s starting to become clear
that this mitophagy is involved with the disease process.
Parkinson disease is an example; a defect in
mitophagy is apparently part of the problem. So we’ve carried
out screens for mutants defective in mitophagy. A lot of those
overlap with genes already discovered for general autophagy, but
some appear to be specific to mitophagy.
After starting your career in a field that no
one seemed particularly interested in at the time, how does it
feel to be at the forefront of a field that’s now extremely
competitive and evolving at a furious rate?
Well, at one level, of course, it’s great. One of the nice things
in a practical sense is that although the field has exploded, very few
people have come into the yeast field of autophagy. So there’s a
tremendous increase in almost every other organism you care to mention,
and I can look at that work and appreciate it and benefit from it, but
I don’t have to worry about direct competition with my own
One of my goals now is to build the community of autophagy researchers.
I’m editor-in-chief of the journal
Autophagy, and we’ve been continually trying to
establish guidelines for the field and the research. In 2003 we came
out with a unified nomenclature for autophagy in yeast in a paper
Developmental Cell. We’re now trying
to standardize that in other organisms and plan to publish that
paper in Autophagy. We published a paper last year with
over 200 authors on guidelines for monitoring autophagy. What do you
need in a paper to say, yes, this is autophagy occurring?
What’s adequate? What would or should convince a reviewer?
Because, not surprisingly, in such a new field, with new researchers
coming in all the time, they’re not sure about these crucial
details. So we set down these guidelines. We also established a
reagent forum at the journal site, to give people a place to list
their experiences with different reagents so that other researchers
can save time and money by using reagents that others have already
tested. We’re now about to launch an online protocol database.
So I guess I’m a nut for organization.
According to our Special Topics analysis of
Autophagy research over the past decade, the work
of Daniel Klionsky ranks in the
top 20 authors at #3 by total cites, #2 by papers,
and #9 by cites per paper, as well as several papers in
2-year top 20 lists for the topic..
ScienceWatch.com talks with
Autophagy's Editor-in-Chief, Dr. Daniel
Klionsky, about the journal's history and citation