Tip:
Highlight text to annotate it
X
While we ultimately want to build whole parts of the heart,
The more immediate goal that's more--
is reachable not in five to ten years,
but maybe in the more near future,
is to be able to regenerate new muscle
in an existing heart.
In existing parts of the heart.
And this could be tremendously advantageous
in numerous forms of both pediatric and adult
forms of heart failure--
one of which, as shown here,
is caused by heart attacks, largely in adults,
where a coronary vessel is occluded.
The blood flow to that part of the heart is lost,
and those muscle cells die.
And the result is the scar that you see on your right, here,
that's indicated in blue
where the wall of the ventricle becomes very thin
and is--this is really fibrotic tissue.
The heart has little or no capacity to regenerate itself,
and so one is left with this scar permanently,
and over time the heart gets-- deteriorates
and gets worse and worse.
And this is what five million people in America suffer from.
Ultimately, the only course right now is a heart transplant
for any of these types of heart failures at the end stage.
But only about 2,000 heart transplants
are done in the United States every year
because of a clear shortage of donors.
Now, what we know is that the heart
is actually made up not just of muscle.
In fact, only less than half of the cells in the heart
are muscle.
The other half of the cells in the heart
are what we call "fibroblasts,"
which are really just support cells for the architecture of the heart
and secrete important signals to the neighboring muscle.
But it's also those fibroblasts
that get activated after injury and migrate to the site
of damage and are the ones that lay down this scar tissue.
And so we asked a few years ago-- began to ask is,
"If the heart has this great reservoir of cells
"already there in the organ,
"then might we be able to not reprogram fibroblasts
to stem cells"--
as Shinya Yamanaka did--
"but rather, directly convert those fibroblast cells
"that are already in the organ
into newly born cardiomyocytes?"
If we could do that, then we could harness this
vast potential of cells already in the organ
and turn them into new muscle cells for regenerative medicine
without having to put new cells into the organ--
and I'm sure you've heard and seen through many--so many grants--
all the hurdles that we are trying to overcome
with cell-based therapies of introducing cells.
I think we can overcome those with time,
but there are clearly many hurdles we face there.
And this might be a little bit easier.
And so, the approach we took was to go back
to how nature normally builds a heart--
which we had invested the last 15 years
in trying to figure out those networks--
and we asked, "If we essentially leveraged that information
"and threw the kitchen sink, if you will,
"of those networks at a fibroblast cell,
"could we get a cell to become more like a cardiomyocyte
in a dish?"
And, in fact, we were able to.
And it turns out that a combination
of just three key master regulatory genes,
some of which I told you about earlier,
are enough to turn a...
cardiac fibroblast into a muscle-like cell.
And those were Gata4,
which I mentioned causes human disease,
Tbx5, which causes a similar disease,
and a third factor, Mef2c.
Now, in a dish, these three factors could push cells
to a cardiomyocyte-like state,
but most cells were partially reprogrammed,
and a few went all the way to become beating heart cells,
even on plastic.
But the question we really wanted to know is not,
"Could we do this on plastic?"
but, "Could we do this in a living organism,
"in an organ--
specifically in the heart in an animal?"
And so we did that experiment more recently.
We tied off a coronary artery in a mouse,
created--mimicked-- to mimic a heart attack.
We injected the three factors
with the virus--the sort of a gene-therapy approach,
waited four weeks and then took those hearts out
and dispersed the cells in the--
from the heart into a dish
so we could examine each individual cell
and see how well we
were able to reprogram the cells into new muscle cells.
And about 50% of the cells looked something like this.
We labeled--genetically traced those cells that would--
started off as a fibroblast and become a muscle cell
with a red marker that you see here.
And about 50% of the cells looked like this,
where they looked essentially just like
a real cardiac muscle cell, which is shown here.
So that looked pretty good.
And if we stained those with markers of the--
of the what we call "sarcomeres,"
or the beating units in the muscle,
it looked something like this.
The green cell here is a real, if you will--
or a native cardiac muscle cell.
And the yellow cell is a newly born cardiac muscle cell
that used to be a fibroblast cell.
And you can see that these
also have these beautiful striations
indicative of a cardiac muscle cell.
But the real proof in the pudding in this situation
is, "Can these cells beat?"
because for heart cells, we have a very rigorous test
of whether they're fully reprogrammed or not.
And so, it turns out that when we reprogram cells
in the "in vivo," native environment
with all the right signals around them
and the stretching--the mechanical forces they might experience--
over 50% of the cells
that are reprogramming
go all the way to beating.
And that's what you see here if I play this movie for you.
Here this cell that's florescent green
is genetically labeled to indicate
that it used to be a fibroblast, or a non-muscle cell.
And this is a real muscle cell
that was already in the heart
that we've taken out.
And so now you can see when we electrically stimulate,
both of these cells are able to contract--
and as I mentioned, about 50% of the florescent green cells
could be fully reprogrammed like this.
And so we were very encouraged by this
and proceeded to see if they could make a difference in vivo.
But before that, we had to confirm one thing
that's very important.
It's not enough to make a cell--
a muscle cell that beats in the heart
if it doesn't electrically couple with its neighbors.
It would beat by itself, would not generate force,
but moreover be a potential source for rhythm disturbances.
And so this experiment that I'll show you here was critical.
And that is to determine if these cells can actually couple
with one another-- both the newly born ones,
but also the previously existing one.
So here what you see is a cluster of cells
that we've isolated now from a reprogrammed heart in vivo.
And now it's not a single cell, but there are three cells here.
One, two, and three.
Only the red cell is newly born, and these two are old cells.
And when I play this movie down here,
we were looking at the wave of calcium
that goes through these cells.
And what you'll see is that when we stimulate,
all three cells have the wave in synchrony,
which means that they are electrically coupled
with one another,
and therefore, they beat at the same time.
And so with this knowledge in hand,
then we proceeded to do the-- the trial.
Okay, but I forgot I was gonna show you this movie.
So just to illustrate for you
what we've done here in a diagrammatic form,
many of you, I suppose, have seen this Waddington diagram
of this hill of a ball rolling down,
and a pluripotent cell becoming a specific cell type.
And you're probably with Shinya Yamanaka's work
where he took these balls, essentially--if you will,
and rolled them up the hill back to a pluripotent state
where they could go back into any one of these states.
What we've done here is something different,
where we've essentially taken these cells
and jumped them over the hill, if you will,
from one cell type in the adult
directly into another without having to go
back up the hill and then come back down.
So this is something we refer to as "direct reprogramming"
because it's going directly from one adult somatic cell type
to another.
Okay, so here's the experiment, then--
the sort of pre-clinical experiment, if you will.
We took a large cohort of mice,
did these infarctions,
and then injected the reprogramming factors,
and then a serial time-points did echocardiography on the mice
to see even with this crude injection in to the muscle,
could we make a difference in their function,
and at the end did MRIs
to look to see--
because that's really the best test to look at cardiac function.
And so I'm showing you here a representative MRI
on your left,
and then the quantification
across tens of mice in each category.
And the dashed line here
indicates the cardiac function by MRI of a normal mouse
that hasn't had an infarct.
This blue bar represents control mice
that had an infarct,
but had a virus injected that was just a control virus
encoding a control gene.
And you can see that it's decreased.
And then the green bar
indicates the experimental mice
that had the three reprogramming factors.
It's clearly not normal,
but it's much better than without treatment.
And the MRIs-- representative MRIs
are on the left--
and we'll go through right now.
If we section these mice after sacrificing them,
the controls look something like this,
where you see now in a cross-section
where the left ventricle is here,
that this is the thin left ventricle wall
with the scar tissue.
And the treated mice looked something like this,
where there's still a thinning of the left ventricle,
there's scar,
but you can begin to see muscle interspersed now
through the scarred area.
And we could genetically label those cells
and ask if those red cells that you see in this scarred area,
thinned area,
were just leftover myocytes,
or newly born ones--
and we have genetically labeled these,
and I'll just tell you that these, in fact,
are all newly born and have the markers
that tell us that these used to be non-myocyte.
So they're actually newly reprogrammed cells.
So we're very encouraged by this
because we think that this represents a way to potentially
regenerate heart muscle without putting cells in--
and all the associated hurdles,
but rather harnessing the cells that are already in the organ
and tricking them to do something that they would not normally do.
Now, there's a lot of work to be done going forward
to really understand this biology.
And so, with the help of a CIRM Basic Biology grant,
we're now aggressively trying to understand the mechanism
by which this occurs--
meaning, where do these three transcription factors
that are controlling thousands of other genes--
where do they sit on the DNA,
and what are they turning on and off
that actually makes this happen?
And what I'll just tell you from our early results
is that it happens within hours.
It's phenomenal how quickly the genome
is being reprogrammed by these factors.
And within three days,
most of the transcriptional changes across the genome
have already occurred.
And then over the next several weeks,
There are fine tuning of gene changes
that ultimately result in the cardiomyocyte phenotype.
But the bulk of the activity is very, very early.
We did not anticipate or expect that to be the case.
And we have the tools now
to actually understand the epigenetics
and the transcriptional changes
that occur at hourly intervals,
and we're doing exactly that with your help.
Now even while we push to understand this
and refine the technology,
we're aggressively pushing forward with pre-clinical studies
in larger animals, specifically the pig
to begin to see if this strategy could work in a larger heart
that's more similar both in size and physiology
to the human's.
And again, this is with the help
of the CIRM Early Translational Award
that we're now able to do our studies in pigs
to test both the efficacy of this approach
as well as the safety.
And so we've already initiated our first sets of pigs
to begin to test how the delivery is occurring
and what levels we're getting.
And I'm very hopeful
that in the next year to year-and-a-half,
we'll begin to have the results
to see if this really-- this approach really can work
in a larger heart.
And we're very, very excited about this,
and are grateful, again.
This is not the kind of work we could do any other way.
It's expensive, and NIH
just doesn't support this sort of thing
that's no longer in the discovery phase,
but really is translational
to move this into a pre-clinical state.
Now I'll just close by saying that this approach
we're very excited about because we don't think it's unique
to the heart.
We--there's evidence from-- at least in a dish, now,
for many other cell types that one can directly reprogram
adult somatic cell
into not just heart cells, but brain cells
by folks at Gladstone,
as well as Marius Wernig at Stanford,
to blood cells, to liver cells,
pancreatic cells,
and it's our hope that others will now
do these types of studies in vivo.
And just last week, a group reported
that that they were able to convert, in vivo,
cells in the brain--neurons,
into spinal-motor-neuron-like cells
that are lost in ALS.
And so I think you can think of this strategy
for multiple, different regenerative approaches
because the developmental biology
of each of these cell types is now fairly well understood,
and it's that developmental biology that then allows us
to switch the-- flip the switches, if you will,
and tell a cell to become what we want it to become.
And so, I'll close there
by just saying that
I think the combination of the powers
of human genetics
along with the stem cell biology,
particularly using reprogramming technology,
puts us in a situation where we have unprecedented opportunity
to both understand human disease in ways we couldn't before
through disease modeling,
and leverage our knowledge
to begin to think of new ways
to regenerate organs going forward.
And so I just want to thank members of my laboratory,
that are shown here,
that are just a brilliant group of people.
Many CIRM scholars are in this picture
that have been supported by our training grant at Gladstone,
and all of them utilize our shared facility
that you all have generously supported at Gladstone.
And we've had great collaborators
including Shinya and his lab at Gladstone,
Benoit Bruneau, who I mentioned,
Bruce Conklin, and Shen Ding,
who moved to join us last year
from the Scripps Institute.
Thank you very much.