– Right, well, I’d like to thank the
organizers for inviting me, and for Steve for
inviting me to talk here. I’d like to begin my presentation, with what’s usually at the end, a little bit of acknowledgments. Although I’m giving this talk, a lot of this work was
done with Jose Alberto, who was a post doc in my lab at the time, and kind of spearheaded a lot
of our ketogenic diet studies, and also Megan and Zeyu,
who are graduate students, and Jennifer and Trina, who are scientists working with our Mouse
Metabolic Phenotyping Center, and helped us with a lot of this work. My goals for today’s talk really are to give a little bit of background on how I became interested
in ketogenic diets, and, along with that,
talk a little bit about ketone signaling, some of the work that Eric
Verdin was gonna talk about, just introduce it a little bit, because Eric and I have
kind of overlapped. I was familiar with Eric’s work, and some of Eric’s work
with ketone signaling, and that helped shape
some of our ideas on, that there might be
some cellular mechanisms beyond just utilization
of ketones as a fuel, that might affect aging. I wanna start with that, and then give a little
bit of an overview of, and summary of, our studies in animals with
ketogenic diets and aging, and then finally, end with some areas that
I think are key points, or areas that need to be addressed to get a fuller understanding of how ketogenic diets might affect aging, and potential steps to move
forward with these diets. I became interested in ketogenic diets through studying calorie restriction. Calorie restriction has been
kind of the gold standard for aging research. It’s an intervention
that, for over 100 years, has been known to increase
lifespan in laboratory rodents, and has been shown to increase lifespan in a wide range of animals from yeast, up to dogs,
and, potentially, monkeys. I became interested in
calorie restriction work over 20 years ago. At first, it was an interest
in trying to understand when animals or humans
go on weight loss diets, the energetic adaptations
and metabolic adaptations that tend to work against
maintaining weight loss. Then I was introduced
to calorie restriction, as it relates to aging, when I began post-doctoral
work with Rick Weindruch. Part of the question then was, are some of these adaptations
that occur with weight loss, that potentially occur with weight loss, also important for aging? My original interest was trying to look at whether these energetic adaptations might be influencing
the increase in lifespan with calorie restriction. Also, around 2000, Kevork Hagopian was a
scientist working in my lab, and he is a excellent biochemist, who was very interested in
intermediary metabolism. One of the questions that
he was looking at is, when animals go on calorie restriction, after they’ve lost weight
and become weight stable, are there still metabolic
changes that occur with them? He was looking at all the major pathways of intermediary metabolism, and published a series of papers. These are data from one of the papers, looking at glycolysis and
regulatory enzymes of that, like glucokinase,
phosphofructokinase, pyruvate kinase. What he was seeing was, with both relatively
short-term calorie restriction and very long-term calorie
restriction in older animals, that these animals, they didn’t return to
levels of these enzymes, or activities similar to control animals. There was some level of adaptation, or some difference that
persisted with them. This decreased capacity for
glycolysis in these animals. This was a slide that he prepared, that was kind of summarizing
just some of his work, and this idea that the metabolic pathways within these calorie-restricted animals were shifted in a way that
they were kind of poised to use fatty acids and
ketones as a fuel source, and had a decreased
capacity for using glucose. That had been one of the
questions that we had had is, are these metabolic changes important, and are they something that may be influencing
aging in these animals? In particular, when looking at ketones, we were interested in the
ketogenesis pathway in liver, and this idea that this pathway is changed with calorie restriction, and kind of chronically up-regulated, and then producing the primary ketones, acetoacetate and beta-hydroxybutyrate, which are interconvertible through beta-hydroxybutyrate
dehydrogenase. That was one of the questions: Is it the shift in metabolism, or is it possible that it’s some components
of that shift in metabolism, and particularly ketones, that might be playing some role in aging and changes we’re seeing
with calorie restriction? This is also some of Kevork’s work. This is just looking within liver. These were calorie-restricted rats that were started on the
diet at six months of age. What we had seen in those animals is that there was a chronic increase in total ketone, acetoacetate, and beta-hydroxybutyrate levels in liver within these animals, from both short term calorie restriction, all the way out to long
term calorie restriction. Along with that, we’d been doing, in addition to the biochemical work, had been doing a lot of work with indirect respiration, calorimetry, and primarily looking
at energetic adaptations to calorie restriction. At the time, we were also getting respiratory exchange ratio data, and for those of you who aren’t familiar with indirect respiration calorimetry, the respiratory exchange
ratio gives you information about substrate oxidation. Values down at 0.7 are utilization of
lipids as a fuel source, one, carbohydrates above one are lipogenesis from carbohydrates. Although we had seen this data, it kind of changed our focus. This paper that had come out
from Marc Hellerstein’s group, that was looking at
calorie-restricted animals and saying, these
animals typically consume very high-carbohydrate meals, 60% of calories from carbohydrate, and yet what they do, the typical way of feeding these animals, they’re given one meal,
they eat it rapidly, they convert much of that carbohydrate through lipogenesis into lipid, and rely heavily on lipid as a fuel. That was the purpose of his paper. We were thinking about that. We were thinking, yes, and
they also become ptotic. Are all these changes important? This kind of finalized these questions: Is the shift in metabolism towards lipids, is increases in ketones,
are those important? And if you could achieve
this without weight loss, these shifts in metabolism, could you have an impact on aging? One of the things that helped
with this was this idea, around this time, with aging, this interest in potential mechanisms through which ketones could potentially, or shifts in metabolism
could impact aging. One of the things was, during this time we had learned a lot more about energy-sensing pathways. These are thought to play a
very important role in aging, and the functions of calorie restriction, and potentially ketogenic diets. In particular, with aging, there has been a lot of
interest in mTOR pathway, and so rapamycin, an inhibitor of that, is one of the interventions. The first intervention, from NIA’s Aging Interventions
and Testing Program, has been shown to change aging. There’s also interest in metformin, and effects on AMP kinase. Part of the idea was that ketogenic diets produce many of these same changes that you see with calorie restriction, and so they might be working through some of these energy-sensing pathways, and inducing changes in
metabolism that can affect aging. The other thing which made us kind of zero
on a little bit on ketones, was this idea that there was
all this emerging information that ketones were actually
signaling molecules, that they were doing something more than just serving as a fuel
source that was being shuttled from the liver to the rest
of the body for use as fuel. Now we know that there’s
a number of functions that ketones can do, and that these give potential mechanisms through which they could be
changing gene transcription, and changing function of proteins, and potentially affecting aging. One of the things that had been noticed is that there’s two cell surface receptors that ketones are ligands
for these receptors: the first HCAR2 receptor,
or the niacin receptor. There’s been a lot of interest in the effects that
ketones may have on this, and effects on decreasing
lipolysis and adipocytes, effects on inflammation, DAT integrity. That’s one option, is that ketones could be
having effects through that. The other is through the
free fatty acid receptor 3. Originally, thoughts were that it was primarily an antagonist, or inhibitor for this receptor, and that that might be
part of the adaptation to starvation or calorie restriction, things that help decrease
sympathetic tone through that way. There has been some recent data, though, that suggests that this
isn’t always the case, and that it may actually
stimulate this receptor, at least under certain circumstances, so there’s a lot of interest in this receptor-mediated
possible functions of ketones that are going on now. One of the other things that we saw, and this is data from Eric Verdin’s paper, is that Eric was doing
some work with ketones, and looking at histone acetylation. He published this really important
paper in Science in 2013. Part of it was this idea that
beta-hydroxybutyrate itself was affecting histone acetylation, and histones are proteins that then influence gene transcription. This was providing a way that histones, that beta-hydroxybutyrate
and ketone bodies could be producing changes
in gene transcription, and also providing a potential mechanism that they could influence
physiological function and aging. One of the interesting
things with this paper was that he was noticing
that they were seeing changes under physiological conditions. For example, under fasting,
he was looking at acetylation at two different lysines,
nine and 14, and histones, and seeing increase of fasting, and also an increase in
calorie restriction in rodents. This fit with this idea that there could be something going on when you have this shift in metabolism that’s beyond just a change
in energy utilization, but could be affecting pathways that can affect the function of the cell. This has been the general idea with beta-hydroxybutyrate at this time, that it’s a histone deacetylase inhibitor, and so beta-hydroxybutyrate can go in, and in particular inhibit
class one histone deacetylases, and in this way change
acetylation of these histones, which in turn affect gene transcription. In particular, HDAC1 and 3,
there’s convincing evidence for changes for beta-hydroxybutyrate,
inhibiting those. There is also some evidence of effects on, possibly, class IIa, at least on HDAC. Some of the important genes that we’re just beginning to understand, some of the genes that might
be changed by this process, but there is a few: FOXO3a, brain drive, neurotropic factor, FTF-21, PGC1-alpha. These are all genes that
have been talked about for potential roles that
they may play in aging. It’s very nice, then, to
see that, under this case, that beta-hydroxybutyrate may be modulating the
expression of these genes. One of the final things
with signaling functions, or functions of beta-hydroxybutyrate, is how it’s affecting
post-translational modifications of proteins. One of the ideas that, as beta-hydroxybutyrate is metabolized, that it forms acetyl-CoA, and this acetyl-CoA can be a substrate for protein acetylation. We now know that virtually
all of the proteins in the mitochondrial intermediary metabolism are acetylated. Many of the important
proteins with metabolism undergo acetylation. We don’t entirely know what
that does to most proteins, but there’s thoughts
that it may be regulatory in the way that
phosphorylation-dephosphorylation can be. The other thing is that when beta-hydroxybutyrate is metabolized, it consumes succinate, and so there’s been some speculation that it may decrease
protein succinylation, which is another regulatory step. Finally, some of the more recent data has also suggested that
beta-hydroxybutyrate itself can attach to proteins, and in that way can
potentially change the function of some of these proteins. There’s all these
exciting areas now on ways that ketone bodies and
beta-hydroxybutyrate can affect metabolism, beyond just serving as a fuel itself. That led us to the key question of, does a ketogenic diet
have an effect on aging? At the time when we were considering this, there was a fair amount of work that was suggesting that high-fat diets have a negative effect on aging, primarily because of weight gain. There wasn’t a lot of studies, or we couldn’t really see studies with really long-term ketogenic diets, or even studies that
were using high-fat diets that were fed in ways to restrict intake or prevent obesity. We thought these type of
studies needed to be completed, and that it was time to take a
look at these types of diets, and see if shifts in
metabolism through diet could have an impact on aging. When we were considering these studies, we had a few goals for the diets. When sitting down with Jose Alberto, we said, one, we wanna start
the diets in middle age. We wanna use middle-aged animals, a time when many humans
would make a diet change. The other thing is, we didn’t want the diets
to cause weight gain. There was already plenty of data out there about effects of different diet, and especially high-fat
diets with obesity. We didn’t want that component. We also didn’t want the
animals to lose weight. We weren’t trying to do another way of studying severe calorie restriction. We also wanted to devise
a couple of diets. One that would induce an
increase in fatty acid oxidation, but not cause continuous ketosis, because we were still very interested in, is it important just to shift
towards fatty acid oxidation, and the ketones aren’t
necessarily important? The other diet that we wanted
to do was develop a diet that produced animals that
were in sustained ketosis. That was our goal, was
basically to take a diet, take middle-aged animal,
feed them in a way that we basically maintain
that middle-aged body weight through the rest of their life. When considering diets, there’s many different ways
that we could have gone. I think this is an important area, when we’re looking at these, whether we’re talking about
human studies or animal studies, is that just saying something
is a low-carbohydrate diet, or even, sometimes, a ketogenic diet, these can be a range of different diets. They could have different
types of effects. We had to carefully consider what type of diet we were gonna use. For our control diet, we decided to use just a
standard, semi-purified diet that we had experience, and others had experience
with life span studies. We had a control where we knew that we could
get reasonably long life spans with animals. Our control diet was just AIN-93G diet. In this case, a little over 60% of
calories from carbohydrate, roughly 20% of calories from both protein and fat. We used the same ingredients, and then just manipulated
the diets for low-carb. We had 10% of calories
coming from carbohydrate, and this was enough that
these animals weren’t ketotic. In our ketogenic diet, 90% of calories came from fats, and 10% from protein. The animals were fed 11.2 kcals per day. This was an amount that we had to adjust, just to make sure that the animals maintained
middle-aged body weight. For our study, we had a couple of goals. One was the question on, what effect do these
diets have on life span? We had a group who, that
was their whole goal, was to measure age of death, and at death, to be able
to do a histopathology, and look at cause of death. The other component were
our cross-sectional cohorts, which I think, in aging research, this is an area that’s
become critically important. In the past, one of
the gold standards was, could you find an intervention
that increased life span? Now, there’s a lot more
interest in saying, not can we increase life span, but along with that,
can we increase measures that show that health is being improved? That you’re creating an
older animal that’s healthy, not just that you’re extending
life span, and along with it, potentially increasing time at which an animals is impaired, or experiencing age-related disease. We had two cross-sectional groups. One where we were looking at just short-term acute time
on the diet in middle age, and then the others, where we
were looking at older animals, at 26 months of age. We had a number of different
goals within these studies, and one of the first was, did we achieve what we were looking to do? One of them: Why were we
getting these dramatic shifts in substrate oxidation. The other question that we were
having with these diets is, a lot of people assumed,
with these high fats, that we were gonna get negative
impacts on body composition, and so we wanted to take a look
and see what really happens, in these older animals,
to body composition. The first, using indirect
respiration calorimetry to look at substrate oxidation. Again, a value of 0.7 is
almost-complete reliance on lipid as a fuel, or lipid and ketones, values of one are carbohydrates. What we saw was, with our ketogenic diet, that the animals basically just maintained metabolizing ketones and fatty acids. It was almost a flat line. Our low carb was intermediate to that. When you fed them, they had some spike, which would show an
increase in utilization of the carbohydrates
and some of the protein, and then they rapidly then
went back down to fat. Our control did, as you’d
expect, they ate big meals, they mostly relied on carbohydrates, they had some level of lipogenesis that was occurring after that. After they had consumed all of their food, they went to using fatty acids as a fuel. We also achieved our goal of
looking at continuous ketosis. The ketogenic diet, even after a meal they maintained ketosis, and were significantly higher
than the other two groups. As far as body composition,
what we saw was, with fat mass, that
there was no differences between the ketogenic diet
group and the control group. Our low carbohydrate group was significantly heavier
than the other two groups. We don’t entirely know why. We have some guesses as to
why that may have occurred, but because of that, they had significantly-increased fat mass. When we look at lean mass,
the ketogenic diet animals were able to maintain their
lean mass throughout the aging. What we saw is, that the
low-carbohydrate group had a higher lean body mass,
they were also bigger animals. The control group, their lean mass went up at very late life, and so there was a significant difference between the ketogenic
group and the other groups. One thing that we wanted
to take into consideration is that some of these
lean body mass changes in these animals, because this is including
all of lean body mass, including internal organs, that some of these
changes aren’t desirable. Some of the increases in lean body mass that may occur under these cases may be internal organ changes that are reflecting
some level of pathology. When looking, particularly,
at skeletal muscle, and this is skeletal muscle
adjusted for body weight, what we saw in the 26 month,
or the old ketogenic animals, is that they’d maintained lean body mass, and compared to the other groups they had comparable or greater amounts of the major skeletal
muscle within the high LIM. The next big question
that we were faced with had to do with life span. That was what most people were asking: What effect were these diets gonna have? We had colleagues who were
making all kinds of suggestions. Many of them thought that this was gonna be a
relatively short experiment. We were quite interested in
seeing how this would play out. These are the life span curves. There was a significant difference, where there was an increase
in life span in the ketogenic, compared to the control,
using logrank tests. There was about a 13.6%
increase in median life span in the ketogenic diet group. The low carbohydrate group
was a little bit interesting, ’cause it was kind of intermittent to the other two diet groups. It straddled between the two, and wasn’t significant different from either the control or the ketogenic. One of the things was that,
when these animals died, was do a necropsy and histology on them. One of the most striking
things about these were changes in tumors. There were some other pathology
changes that were observed, but one of the things with mice, and particularly C57 black six, which we used for these studies, is that the primary cause
of death in these animals are due to tumors, and in
particular histiocytic sarcomas. About 50% to 60% of these animals typically will die of cancer, and particularly histiocytic sarcomas. One of the things that
was noticeable on necropsy with the ketogenic diet was how few of them had tumors, or had visible tumors
that could be detected. The first slide is just
showing total tumors. This includes both sarcomas, but it could include, also, adenomas. We also had some mast cell tumors, and a few other different kinds. This is also including tumors that weren’t likely cause of death, and may have not
contributed major pathology, but just looking at total, and in this case some animals
could have had more than one. The ketogenic diet was
lower than the others, and for the major type of cancer
within this type of mouse, we had one out of 10 ketogenic animals that showed that tumor. It was striking, because the other groups
all kind of hit the pattern that you would expect
for this type of mouse. That was one of the pathology changes that was very clear with
animals on this diet. The next things that we
really were interested in, we thought it was very interesting that there were changes in life span, but the bigger question now
that was coming out were, were these animals healthy? Were we getting increases in life span? What we’d wanted to do was try to take a number
of different measures of health in these animals, and try to get kind of a broad assessment of how these animals were doing, and at the same time, though, not trying to overdo what we
were exposing these animals to. One of the things that
we wanted to look at was effects on cognition. The test that we used for
that was novel object test. What the novel object
test does for mice is, it’s basically, it recognizes normal behavior in a mouse, that if you put a mouse
into a new environment and then give it a couple of objects, it’ll explore those objects. The test is that, if you take that mouse
out of that environment, put it back in, and
change one of the objects, is the animal gonna be able to remember that the object is changed. If it does, the natural
tendency for the mouse is to spend a little bit more
time exploring the new object. That’s what this test was focused on. This is a test that has clearly been shown that memory decreases in mice with age, and that’s what we saw in this case, that with the low-carbohydrate diet there was a significant difference. There was a strong trend towards a decrease in
performance in this test, also in the control group. No age-related change in
the ketogenic diet group, and the ketogenic diet group
did significantly better than both the control and
the low carb in this test, at 26 months of age. In these older animals,
at least with this test, evidence of increased memory
performance in these animals. These are areas where we’re
expanding these tests now, and looking at other different
regions within the brain, and using more tests to look
at effects on late life, on memory. The other is, we had a series of tests that we wanted to look at motor function and muscle strength. One of the typical muscle strength studies that’s done with mice is grip strength. It’s basically just taking a force meter that has a bar attached to it. The mouse grabs a hold of the bar, and then the mouse is pulled
until it releases the bar, so an evidence of ability to strength to be able to hold onto the bar. In this case, there was one change that
occurred in middle age, with low carb, compared
to the control diet. The control diet, there was, as expected, an age-related decrease
in muscle strength. The ketogenic diet group had significantly better performance in this measure of muscle strength, compared to both the low
carb and the control group. Another test that is done
is hanging wire test, which is a measure of both
muscle strength and endurance. Essentially, what it
is, is a wire is placed, the mouse is put onto the wire, and its ability of mouse
to support its weight for a period of time, in this
case a three-minute test. As you noticed, in middle-aged
animals they all did great. They basically all completed
the full amount of test. This is one of the tests where we saw the most
striking changes with age. With advanced age, in both the control and
the low-carbohydrate diet, there was a dramatic decrease
in performance in this test. Whereas, with the ketogenic diet, we didn’t have a significant
decrease in the performance from middle age, and they did significantly better than both the control and
the low-carbohydrate group. Another test that we looked
at is called the rearing test. This is one where animals are
placed in a clear cylinder, in an environment, and it’s based on normal mouse
behavior to want to explore, and along with that, to kind
of move along the cylinder, and be able to rear up on their hind limbs as part of exploration. What we saw, again, with this, this measures both a
willingness to explore, but also the coordination to be able to rear up on their hind limbs. In this case, there was also a trend towards an age-related
decrease in the control group, and in this case, the ketogenic group did
better than the controls at 26 months of age. Then, one of the final
physical performance tests that we did, or that I’m
gonna show you today, is something called the
locatronic ladder test. What this is, is basically
a coordination test. It takes a mouse that’s
in a brightly-lit area. There’s a ladder that separates that from a darker area where
the animal wants to move. It basically measures
the ability of the animal to run across that ladder and get to this darker
area where it wants to go. The faster that an animal
can complete this task, the better. In this case, all animals performed
similarly well in middle age. With advanced age, there was increase in this performance with the control group, and that the ketogenic diet did significantly better than the control, and that they were able
to complete this task in a faster amount of time,
shorter amount of time. In addition to those measures
of physical performance, there were a number of other tests that we wanted to complete, just to give an overall idea
about health in these animals. One of them is, particularly
with aging, is inflammation. There’s not doubt, with
advanced age in animals, you see increase in
markers of inflammation. There also tends to be an increase in variability
in inflammation, which reflects changes in
pathology within the animals. What we wanted to do was select a series of
markers of inflammation. The three that we selected were ones for which there had
been quite a bit of evidence that these change in mice in response to different interventions. We tested those. In middle age, we didn’t
really have evidence of increased inflammation
in any of the animals, and the diets behaved similarly. At 26 months of age, we saw, as been reported
in the literature, that there were age-related increases, at least in TNF-alpha and CXCL1, in the control animals. No significant age-related increases in the ketogenic diet animals. The other thing with
ketogenic diet animals that we noticed is we just didn’t see the variability that we typically see with
aging in other animals, so much tighter data. The next thing that we had
a lot of questions about was what was gonna happen to serum markers of lipid
metabolism within these animals. These are data from 26 months of age. What we saw was, really,
no significant differences. The groups all behaved pretty similarly. The only difference was,
really, with free fatty acids, and that the low-carbohydrate group had higher amounts than the other groups, but the ketogenic diet had similar values, and if anything, maybe a trend towards higher HDL cholesterol
than the other groups. The final area where we
haven’t really done that much, and at meetings, where I get a lot of questions
are about insulin signaling, and particularly looking
more within the cell at cellular measures of insulin signaling, and questions about whether
we’re seeing long-term changes in these animals. We haven’t done the cellular work, but what we did was do the
general glucose tolerance and insulin tolerance tests. In these, animals are given an IP
injection of glucose, or given glucose in insulin
for the insulin tolerance test. What we saw is, after these animals had been long term on
ketogenic diet animals, they were somewhat glucose intolerant. There was a significant difference in glucose tolerance tests
compared to the other groups. The interesting thing was, when we did an insulin tolerance test, these animals had still
maintained their ability to respond to insulin. So, our general thought was, the glucose tolerance test effect was just reflecting the fact these animals hadn’t seen
appreciable amounts of glucose for a long period of time, and were maintaining very
low levels of insulin, so they had somewhat
impaired ability to respond when given this big glucose bolus, but when given insulin, they still perfectly responded to it. This set up the last part
that I wanna talk about in the remaining minutes of this talk. These are some of the key
questions that still remain. One of them has to do with mechanism. That’s been a lot of the focus. There’s been a lot of interest in trying to narrow down mechanism, and to try to determine
if you can write a grant and figure out something
where, in five years, we can figure out the
mechanism for how this works. I’m a little bit skeptical on that, because I think this is a diet that’s having massive
changes in metabolism, and I think it’s multiple mechanisms that are contributing
in different tissues. We tried to take a look
at a few of the mechanisms that have been proposed. From an aging standpoint, one of the most important
ones is this mTOR pathway. This is an important pathway for protein synthesis and growth, and it’s an area that’s of
great interest for aging, because rapamycin, which
inhibits this pathway, has been shown to increase life span. We looked at that, and
in liver, what we saw was that just phosphorylation of mTOR, there wasn’t a significant difference. There was maybe a trend towards
ketogenic diet being higher, but if you looked at downstream
indicators of mTOR activity, so phosphorylation of 4E-BP1, rpS6, what we saw is a significant decrease with the ketogenic diet, but also a difference that occurred with the low-carbohydrate diet in liver, and also a trend towards
a decrease in rpS6. We looked at DDIT4, which is an inhibitor of the mTOR pathway. That was significantly
increased in the ketogenic diet. So, we had some evidence
that the ketogenic diet is having an effect on
mTOR signaling in liver. But this is also a complicated pathway, because we didn’t see these
same changes in muscle, and there’s still some question on which tissues are
showing changes in mTOR, and how, exactly, is
that affecting response to ketogenic diet? The next thing that we looked
at was protein acetylation, and this is one of the most
dramatic changes that we see with the ketogenic diet. These are data that
we’re showing from liver. What we see is this total protein
acetylation of lysine in those, a dramatic and significant
increase in the ketogenic diet compared to all the other diet groups. When looking at p53,
that specific protein, also a dramatic increase
with the ketogenic diet. When looking at histone, in
this case also an increase. Interesting thing here,
though, is that we also saw that the low-carbohydrate
diet had a significant change, at least in that measure of acetylation. So, this is one area
that we are focusing on, that we think is quite important, these changes in protein
acetylation with ketogenic diets, and this right now seems to be one of the
more striking changes that we see with this diet. The other thing that we
were very interested in was mitochondrial content, this idea that, especially
in very late life, whether these diets were gonna help maintain
mitochondrial mass, and potentially mitochondrial function. This ended up being a lot more
complicated than we thought. It very much depends on which
tissue you’re looking at, and it very much depends
on what’s being measured. What we saw was, in no case did we see
ketogenic diet in old animals, and these are 26-month-old animals, having a uniform increase in every marker of mitochondrial
content that we measured. What we would see is that
it’d have an increase in some, and in this case, in skeletal
muscle, and activities of two of the electron
transport chain complexes. We saw that was variable, and we also saw at times,
like in this example, that the low-carbohydrate diet sometimes had similar effects. I have a graduate student
who’s working on this, and it’s ended up being a much
more complicated experiment than I had wanted. She got a much bigger challenge than I think we both expected. The second really important question has to do with calorie restriction and intermittent fasting in ketones. One of the questions is, does ketogenesis have to be continuous? This is Eric Verdin’s paper that came out at the same time as our
paper with ketogenic diets. Eric ran into a similar
problem that we had. If we tried to feed
ketogenic diets ad libitum, the ketogenic diets that we were using, the animals overate them,
and they became obese. That didn’t meet one of the goals. So, that was part of the reason that we fed amounts to maintain
middle-age body weight. Eric saw the same thing with his animals, but his approach to get around that is, he put them on ketogenic diet for a week, put them on a control diet
for a week, and so he rotated. His goal for that was
to prevent weight gain, so his animals’ weight cycled. Despite that, coming on and off ketosis, he did see some positive changes, and particularly increases in memory. I show some of his data
from his novel object test that was in his paper. Where I think that’s important is that, with calorie restriction, and how I first got into this is, there’s usually two different approaches that people use for calorie
restriction in animals. There’s where the animals
are fed every day. Under this case, when
the animals eat a meal, they go out of ketosis, and they have a period of
several hours, eight to 10 hours, and then their ketone levels go back up. So, they’re continually
going through this cycle of increased ketosis, where they go into ketosis
for periods of 12 hours, maybe a day, or much,
and then go out of it. Or there’s the calorie-restriction method where they’re fed calorie-restricted diets three days out of the week, and they just cycle through
a few days out of the week. Part of the question is, if ketones and beta-hydroxybutyrate are playing an important role
in the metabolic adaptations and the effects of calorie
restriction on aging, it’s not doing it through
continuous ketosis under this circumstance. That’s a big question. Is it that the two things aren’t related, and we just kind of lucked
into studying ketogenic diets, and that they really aren’t playing a role in calorie restriction? Or is there some signaling role that these kind of blips
in ketosis are playing in calorie restriction? I think that’s an important question that remains to be addressed. There’s two remaining questions
that I think are important. One of the most common questions
I get with the animal diets are, when people are asking
me about making these diets, which fats should I put
into a ketogenic diet? I honestly don’t know. I mean, I’ve tested the fats
that I’ve used in my own diet, but I haven’t been able
to test a range of fats, so I don’t really know if there is an optimum fat
composition for these diets. I think that still
remains to be determined. I think, from an aging standpoint, probably the most important
question is the last one. That’s, what effect these diets
have on age-related disease. We’re never gonna be able to
complete a life span study in humans using ketogenic diets. We can be able to do those
manipulations with animal models, but we can use animal models, and potentially translate that to humans with age-related disease. That’s the big question is, which age-related diseases are gonna be responsive
to ketogenic diets? When can you start them, as far as in progression of disease? And which approach for delivering the diet is gonna be the most effective? I think those are critically
important questions in the area of aging. Then, just to wrap up, what
we’ve seen from our work is, we’re convinced now that a ketogenic diet, if it’s given in isocaloric amounts, if it’s in a situation where
animals aren’t overweight, that it can increase life span. Also, under same circumstances, this type of diet has an effect to be able to extend
physiological function, maintain healthy function in
these animals into later life. I think it’s critically important, kind of the last, part two,
is that level of intake is extremely important in animal studies, and presumably in human studies, too, in how a diet’s gonna have its effect. Those two have to be taken
into consideration together. With that, I’d like to wrap up, and use that just to
acknowledge all the groups that have played an important
role in our studies. Keith Barr’s lab, and Marita Wallace, they played important role in some of our signaling measurements, and we continue to work with Keith in trying to study the effects of ketogenic diets on signaling. Same with Floss Hodges’ group. He does similar work. Gino Cortopassi, who we’ve done some of the
work with mitochondria, and also the team at the Mouse
Metabolic Phenotyping Center, and finally, my collaborator, Jose Villalba at University of Cordoba, who’s done a lot of our work, and is somebody that we’ve talked to a lot with our diet and lipid studies. With that, I’ll go ahead and end. Thank you. (audience applauding)