Extracellular Matrix Remodeling during Cancer Progression and Metastasis (Manu Platt)
1 Video View·Sep 5, 2023
ManuPlatt 08 07 14 youtube
I came last year, I had a great time here,
and great talking to the students afterwards.
So yeah, so I hope you'll find that topic interesting.
So I'm talking to you today about extracellular matrix
degradation during cancer, and it's actually
going to segue into some personalized medicine
technologies that we're working on to look at disease severity
and to predict patient prognosis.
But yes, I do bring you greetings
from Atlanta, Georgia, the Wallace Coulter Department
of Biomedical Engineering.
And we're a joint department between Georgia Tech
and Emory University School of Medicine.
So if you're interested about that as well,
I would love to answer questions about that.
So my undergrad training is in biology,
and then I got my PhD in biomedical engineering.
So I approached it from the more biological side,
so that's why I was kind of getting
a flavor for how much to discuss.
But I think it'll be accessible to everyone.
So this is me.
I run the Platt Lab for Repair Regeneration and Remodeling
since January 2009.
We really, we have a huge cardiovascular disease focus
where we study sickle cell disease.
And my children with sickle cell disease get strokes.
And again, that's going to be a lot of matrix degradation.
We also study HIV and how people with HIV
have increased risk of heart attacks or strokes.
And I actually have some collaborations
here with people in Illinois on some of these projects.
But what I'll be talking to you mostly about today
is our early cancer detection technology,
and also our system biology and predictive medicine strategies.
And plattlab.com, check us out.
OK.
So when I say we study remodeling,
so what is tissue remodeling?
So I look at tissue remodeling as how do you
go from this healthy tissue.
So this is a coronary artery, a human coronary artery.
And you can imagine that the blood
comes through this cross-section to go
throughout the body, oxygenate all of your tissues.
However, when atherosclerotic plaques form,
you can see this huge plaque has now formed.
And this healthy vessel is now completely remodeled.
And you can see now here where the blood should
be passing through.
It's no longer able to pass.
If this is a coronary artery like this,
if you get a heart attack, if this
is a carotid artery or cerebral artery, you get a stroke.
But for this remodeling to happen,
there's a large amount of protein degradation
that helps to start off some of these processes.
New proteins get synthesized and deposited.
And then, of course, there are cells
that infiltrate into this area, changing
the structure of this tissue.
Not only the cellular properties,
the mechanical properties, the chemical properties,
they all change.
And that's where BMEs add a lot of value to the field.
Most people know about osteoporosis.
You have this normal bone matrix.
It's dense.
It's got a lot of type 1 collagen, a lot of mineral
on it.
But again, in certain women postmenopausal,
that will start to break down.
And you can see this loss of the protein matrix.
And then that leads to the loss of the mineral matrix.
And you have this now brittle bone
that will be osteoporosis.
And so again, this tissue goes from this healthy
to this diseased tissue.
And again, we'll talk a lot today about breast cancer.
It's just always not good to show a breast cancer
picture in the morning after you've had breakfast.
So I just showed the pink ribbon to be polite.
But so I told you, my undergrad degree was in biology.
So when I came over to bioengineering,
what they kept saying to me the first year of grad school,
structure relates to function.
Structure relates to function.
I guess that's what engineers believe.
And it's actually true.
So I like to model this for people.
So if you think about this castle,
back from medieval times, right?
So in this castle, it wasn't built to look like a castle.
All of the features of it are there for a particular reason.
So why does this thing have these narrow windows?
Archery?
OK, we'll give you that.
Archery?
What else?
Why don't they have big windows, so you can really
fit out the window?
What is that?
Privacy?
OK, maybe privacy, too.
That's good.
Yeah.
It's for compliance.
Right.
It's for compliance.
Right.
So they could do that.
Well, also, remember, they didn't
have central air conditioning back in the day, either, right?
So they want to regulate temperature.
You can't have these wide, open spaces that you can't control.
So again, that structure relates to the function.
And then, of course, these turrets at the top,
what are these here for?
Cannons.
Cannons, right.
Archers, and the other forms of defense and protection, right?
So they didn't say, let's put some turrets,
because that looks really cool.
No, they're there for a reason.
And you can contrast that with this condominium building
down in Miami, which is built to be luxurious.
Lots of windows.
Are Miami in the house?
OK.
Lots of windows, built for luxury and decadence, right?
So again, the structure relates to the function.
And it's actually, of course, very true for the body.
So let's see if the electrical engineers in the room.
So what is this structure right here?
What is this right here?
Anybody know?
Your intestines, right?
We're a smart group.
And so you can see the small intestines
are completely folded and convoluted inside of our body.
And they're not folded, because they should just be folded.
It looks pretty if they're folded.
You know, we take one person, and they
stretch all the way across the room.
But they're folded up so they fit.
Why do they need to be so long?
Absorption, right?
So they want to take in the nutrients
from the food you digested.
And if you have a greater surface area,
then you can transport more of those nutrients.
So again, the structure relates to the function.
And then on the wall of that small intestine
are these finger-like projections called what?
That's cilia.
What was it?
Villi, right?
So these are your microvilli on the surface.
And again, they increase the surface area
for even greater absorption of the nutrients
that you've been digesting.
So your structure relates to your function.
And I love the ear.
I think of this ear as just this incredible architecturally
designed organ that just takes small sound
vibrations to this membrane right here
and transmits those into biochemical signals.
All right, just beautifully designed structure.
So the structure relates to the function.
However, if you mess up the structure,
you mess up the function.
So again, your nose is actually built
to breathe in oxygen and fresh air
so that the oxygen can go to your brain
and your brain can work.
But if you close off that passage,
you get a loss of oxygen deprivation,
and then you end up with brain damage.
So these are the things that we end up studying.
How, if the structure of your tissues change,
does it lead to a loss of function?
So what I'm going to tell you about today
is I'm going to talk a lot about extracellular matrix.
So just so the group is all on the same page,
we'll just describe that a little bit first
and some of the key molecules people care about.
And then I'm going to talk a lot about proteases.
So proteases are enzymes, which means they're proteins.
But they're proteins that degrade other proteins.
So specific enzymes that cleave other proteins.
And so they, of course, can degrade this ECM,
which I'm going to use this a lot,
instead of extracellular matrix.
And I'm going to focus on this, my favorite brand of proteases
called cathepsins.
And then, of course, we'll put that in the context of cancer
and metastasis.
And then, as I mentioned, I'll also
transition from that into some personalized medicine
techniques that we're developing and how
we're looking at cancer prognosis.
And some of this is unpublished, so there you go.
But it's first to think about the cell
and its microenvironment.
So it's funny, when I talk to my engineering undergrads,
they think if you take a pill, you get better
because you've taken a pill.
But no, you get better because that pill
has a drug that will interact with some receptor
or some cellular process.
So everything happens for me down at the level of a cell.
That's where everything happens.
But the cell, what I appreciate about BME
is that the cell actually lives in the same physical environment
as we do.
It responds to gravity.
It responds to chemistry.
It responds to gradient.
All of those things happen for the cell.
So you can think about the physical cues.
They're actually bound to these matrix proteins, which
I'll talk about in a second, which also will stimulate
the cells with different signaling things that all
go to the nucleus.
Many people know about the soluble signals
that cells receive that are just in the blood.
They can take them up.
That will stimulate some sort of signaling pathway.
And then, of course, there's these cell-cell interactions
that also lead to some signaling that all
let the cell know who I'm touching, what environment am
I in, and what kind of surface am I sitting on.
And the cell processes all of that information.
So as it changes, then the cell changes its response to it.
But what I like to remind people is
that when you get these proteins that
are outside of the cell, this extracellular matrix,
as I'll talk about, where do you think that comes from?
Who makes that?
The cells make it, right?
So the cells make it and put it outside.
And then they respond to the fact they're sitting on it.
And they might change and even make more or make less.
So there's a constant dynamic feedback
with their environment.
So here's a nice picture of this extracellular matrix
that the cells are all sitting in.
Any cell that you guys have been growing in dishes,
you let them sit there long enough,
they are laying down and depositing
more matrix over time.
Most cells will make their own matrix over time.
But they also will remodel and degrade that matrix.
So this is a nice micrograph of the basement membrane,
or it's commonly called the basal lamina.
So all cells sit on the basal lamina.
But very important for this cancer discussion,
epithelial cells, I think 98% of cancers
are from epithelial cells.
And epithelial cells have a polarity.
And they all sit on a basement membrane here.
So you can see this little scratched away piece
right here that the cell sits on.
And so it actually is comprised of several different matrix
molecules.
There can be growth factors bound
within this matrix, in this lamina.
And depending on the type of tissue that you're in,
the basal lamina will have different compositions.
But collagen IV is a major component.
Laminin is a major component, the basal lamina.
But again, proteins that the cells have deposited
that they sit on.
Now, when we talk about in terms of cancer,
so here's an open normal duct, which
you can see the lumen inside.
You've got your apical surface and your basal lateral here.
When they talk about a tumor being benign,
it doesn't mean it's not cancerous
or whatever that means.
Benign simply means the cells have not
broken through the basement membrane.
It hasn't started its initial invasion.
That's all benign means.
So it doesn't mean that you're safe
and that nothing will ever happen.
Because there could be a chance that over time,
they start to break in.
That's what they mean when they say benign or malignant.
And then your malignant one, which we all
associate with being more dangerous,
is the one that has now started to invade through that basement
membrane and now it's breaking into the rest of the tissue.
And that's where we know it starts
to become a major concern.
But again, you can imagine to do that,
it has to start to degrade some of these protein
structures in the basal lamina.
And that's where my proteases come into play.
So it's an interesting thing.
Not interesting, actually, it's a bit of a tragic thing.
The five-year survival rate for surviving cancer
drops dramatically as the cancer invades.
And I think we all are aware of this in our mind.
So if they catch it early on, then you
have a great 61% chance of making it
past this five-year survival rate.
Once it gets into this regional invasion,
you can see it drops off precipitously.
And then, of course, we know when it has metastasized.
Now the five-year survival rate really drops.
And this is different for different types of cancers,
but the general trend is mostly true.
So what are these steps in metastasis?
We've talked about that early benign, or just even not
benign, normal epithelium.
And then as it starts to grow up into this space,
it becomes benign.
And then it penetrates through that basal lamina.
And now we have a malignant tumor.
And then, of course, as the cells
that you guys have had cancer all week, I'm sure.
So then as the cells proliferate,
they recruit vasculature that then comes into the tumor.
And then the cell, again, has to break through the protein that
actually outlines the artery, or the vessel
that they're going into.
And it'll locally proteolyze that to break into the vessel,
get flushed downstream.
And then it will stick somewhere.
And when it adheres to that wall,
again, it's now sticking to a new type of matrix.
So it's sticking to something which will tell that cell, hey,
we're stuck.
Let's do something to get out of here.
Signals get turned on.
It starts to local proteolyze.
So it can now break back into the wall
and settle in some new tissue.
So there's proteases involved in the early stages
and also in the settling stages as well.
And then they also help the tumor grow in space and volume
because we need to clear out some of this crap around us
so we can all live.
But of course, one of the worst things that can happen
is metastatic bone disease.
And so breast cancer and prostate cancer,
they account for about 80% of all metastatic bone disease.
So this is tough.
OK, so in the bone, you know about bone marrow?
Anybody know about bone marrow?
So in the bone marrow, it's highly vascularized, right?
And that's where all of your blood cells
are made is in your bone marrow.
And so when these cancer cells get in there,
actually, there's no good way to cure bone cancer right now.
So this is when the prognosis is actually the poorest.
And what you can develop, you can
see this picture of a person with metastatic breast
cancer in the pelvis.
You can see it's riddled with these holes where
there are these osteolytic lesions, meaning
the cancer cells start to eat away the bone.
It's also deeply painful because bone pain is a very deep pain.
And then it's hard to get all of that cancer out of there.
So again, for patients with the advanced breast and prostate
cancer, 65% to 75% develop these bone nets.
And 40% of patients with lung cancer
will develop bone nets.
So I'm going to talk about this in a bit, too,
because these enzymes that I love
are actually expressed by osteoclasts that actually
remodel the bone, and they start to partner up
with the cancer cells to really start to tear things apart.
So let's talk about my baby.
And I don't have all the slides in there,
because some of them are just, I can use a lot of paper,
but the important ones are in there.
So I love these things called the cysteine cathepsins.
So cathepsins are, well, I'll tell you more about them later,
but they are actually involved in several different types of cells
that make up a tumor.
So a tumor is made of more than just those cancer epithelial cells.
They recruit vasculature, as we know.
There's also an inflammatory response,
as the body maybe tries to fight the tumor.
And so tumor-associated macrophages come in,
which we'll talk about quite a bit.
And the epithelial cells make up the blood vessels
and several other cell types.
But they all have been shown to produce these enzymes
in this tumor environment.
And really, the key trick is that the tumor cells themselves
initially start to upregulate these proteases
so that they can start to escape.
And normally, they're turned very low in most epithelial tissue.
So it's a nice binary signal, which I'll show you.
So a little bit more about them.
So they were originally called the lysosomal cysteine proteases,
because they were identified in lysosomes.
What are your lysosomes?
Anybody know what they are?
Yeah.
They're like your trash compactors in the cell.
They degrade all the extra protein that it wants to turn over.
So these are really powerful proteases,
because they eat all the proteins that your cells want to destroy.
But they've been shown to be secreted by osteoclasts.
As I mentioned, these are the bone-resorbing cells.
Your bones are also for calcium homeostasis,
as well as for structure.
And you're constantly remodeling your bones.
So they break down the bone to release calcium,
and the osteoblasts build it up.
And then macrophages and other cell types.
But here's what I love about these things.
They are the most powerful human collagenases and elastases.
So they degrade collagen better than any other protein
we have in the body.
And they also degrade elastin, which
is a structural protein that we also have,
better than any other human protease.
But here's the issue.
This is why I think they're less studied
in some of these contexts, is they function best
at an acidic environment.
So I started studying these things in graduate school.
And it used to be the thought that, well,
when they get outside of the body, they don't work,
because it's neutral outside of the body.
They don't work.
But we now know, and I think it's now much more broadly
accepted, that outside the body, especially in these disease
environments, the local conditions
can become slightly hypoxic.
I mean, with the hypoxia, they can become acidic.
And all these other things are now known.
Plus, cathepsins are now known to maintain
some level of activity at a neutral pH.
Not forever, but for some short time.
But they're so powerful, in that short time,
they're degrading a lot of matrix.
And they do prefer a reducing environment.
And they're highly regulated.
Again, because in the normal tissue,
a lot of normal tissues, unless it's
a cell that specifically is using them for one function,
they're not turned up.
They're just in lysosomes working,
not being overproduced.
And when they get outside of the body,
there's all these regulations.
The neutral pH will minimize their activity.
They're also synthesized as an inactive precursor that's
got to be activated to start working.
So there's all these ways that the body regulates it.
Plus, it releases lots of its inhibitor
constantly in all of your bodily fluids.
So when I say that it has to be activated,
there's this proform here, this procathepsin.
This is how it's synthesized or translated,
where the active site is blocked.
So to turn it on, you have to cleave this propeptide off.
This can happen under really low pH.
Or it can happen where other proteases can come along
and hydrolyze a piece.
Then you get this smaller cathepsin now,
because you've lost some of that protein component.
But now it's active.
When it's active, it can bind to its protein substrates
and degrade them.
But it's simple.
So the proform cannot bind the substrate.
Or if it becomes denatured, then it cannot bind substrate.
So these are ways we think about this.
So if we want to detect its activity,
we need it in this mature form.
So when I was going through grad school,
again, these things were just starting
to make its rise as being important in diseases.
And back in the day, biologists, you
had to show an assay three different ways.
So we had to do the mRNA.
We had to do the protein.
Then we had these synthetic substrates you had to use.
The issue is they're quite promiscuous.
They all like to degrade the same substrates,
but with different affinities and different timescales.
And so there's all this cross-reaction.
You have to use sRNA to knock yours down, to look for the lot.
So when I started my own lab, I wanted
to make it very easy for my grad students.
And they don't fully appreciate how easy I've made it for them,
because they take it for granted.
They really don't.
It's really shocking.
But that's fine.
And so I tweaked this assay, this assay called Zymography.
And so I'll explain it all to you now.
And it's been our bread and butter since we started it.
How many of you guys have heard of SDS-PAGE?
Western blotting?
OK.
All right, so it's based on the SDS.
Well, it is SDS-PAGE, actually.
So SDS-PAGE is SDS polyacrylamide gel electrophoresis.
So we take this polyacrylamide gel,
which is inert to proteins.
But what we'll do instead is we actually
will also impregnate it with gelatin.
Gelatin is just denatured collagen.
So it's a simplified form of collagen.
So it's all throughout this gel.
Then we'll take our cells and use isolated enzyme.
And then we'll put them into this loading buffer,
which has SDS in it.
SDS gives them a negative charge,
so that if we could apply an electric field,
then they'll move through the electric field.
It also will denature the proteins partially
and then give them that charge.
Then we can load them into our gel,
apply an electric field or voltage,
and the proteins will begin to separate by the molecular size.
Because the smaller proteins can navigate
this tortuous matrix of this polyacrylamide gel
a lot faster.
So the smaller proteins move faster to the bottom.
Larger ones stay more to the top.
But you can get this separation.
Then in a Western blot, you would normally
transfer this to a membrane so you
could probe it with antibodies.
And we're not going to do that.
We're going to transfer it to this renaturing buffer,
where we can wash out this negatively charged
SDS, which is partially enfolded
with the protein because it's interfered
with ionic interactions, et cetera.
We'll wash that out, and the proteins
will actually refold into their native conformation.
Now, not all proteins refold, but the ones that do,
we can refold.
And then we put them into an assay buffer,
which has everything that's perfect for them to work.
So it's acidic environment, it's got slightly reducing space,
perfect for them to work.
And we can incubate that overnight.
And the cathepsins will now degrade the gelatin
at the site to which they were separated
during the electrophoresis.
Come back the next morning, and we
can stain the entire gel with Coomassie blue.
Coomassie blue, as we know, binds to protein.
But what's different is there's so much protein in the gel,
the whole gel turns blue.
But you get these cleared white bands
where the enzymes have actually degraded the gelatin.
So it's like the negative of a Western.
And then you can stain it.
And because we've separated the proteins by their size,
they actually migrate to different distances
in this gel.
And we can actually identify them.
So we've been able to multiplex this for cathepsins K, L, S,
and V. You can see they all come to a different size,
such that even if we load them all together,
we can now tell which one we have.
So I love this thing, because not only do we get,
we can quantify this with densitometry,
but we also have visual confirmation
of which enzyme it is, instead of trusting
some fluorescent signature and just believing
that it's what we want.
And you can add inhibitors to the incubation space
and block them and do all these wonderful things.
So this is fantastic.
And when we first did it, first I
just wanted to get it working for K.
And one thing we found, so this is a Western blot
for cathepsin K. And we can buy the proform.
And if you remember, they get activated from the proform
to the mature form by getting that piece cleaved off.
And you can see there's a difference in size.
And so a Western blot will tell you the protein amounts.
But the complicating thing about Westerns,
if you're doing them in other kind of ways, or ELISAs
and the rest, is you're going to detect
the proform and the mature form.
Remember, only the mature form is the active form.
But in our zymography, we only detect the mature form,
because these enzymes must be active to cleave
this substrate.
So I'd say that to you, but in the proform, I have a band.
And all the smart people in the audience are like,
but you have activity there.
What happened?
This puzzled us for a while when we first were doing this,
because it shouldn't be active.
But then when we looked at the molecular weight,
we were like, oh wait, it's at the lower size.
Now separately, I had an undergrad
who was doing computational modeling for me
of the kinetics of this assay, for different things.
And he was a person who had a great with models,
and he was learning the biology.
And he kept coming to me, he's like, Dr. Platt,
I know you did the experiment, but I think you did it wrong.
And I said, well, what do you mean?
Because it doesn't fit my model.
So your experiment must be off.
So I don't know what you're doing.
I was like, Zach, that's not how it works.
The experiment is true.
Your model must not fit this.
So he said, well, the only way I can get the model to fit
is if I let it try to fit the initial enzyme that you use.
And it will only fit if it says that about 20%
of the inactive form is already active.
So I don't know what you're doing.
And I was like, Zach, and I'm kind of animated, right?
So Zach, and I'm screaming in the office,
this is what we've been seeing here, and we can't explain it.
Because we weren't seeing it in the Western,
but we're getting it here.
Oh my god, Zach, this is brilliant.
And so I love it when models tell you something
that you didn't know, because you can actually
trust the model even more.
So this also told us something else.
Well, we're seeing it in zymography,
but not in the Western blot.
It must be more sensitive than the Western blot.
So we went back and did a limiting dose.
And we actually can see where the Western blot,
with the best Cathepsin K antibody,
stops, around 1 to 5 nanograms.
But at zymography, we can go all the way down to 0.1 nanograms.
And so that's actually, if you do the molar ratio,
it's actually 3.45 femtomoles of Cathepsin K
that we can quantify with our zymography.
So this was huge.
So not only can we get just the mature form,
we can also very sensitively detect it.
And guess how much antibodies cost?
A lot of money.
You don't use antibodies, really cheap assay.
And so then again, that's when later we figured out
we could actually multiplex it by changing
some of the conditions.
And so we published that as well.
So this is our zymography.
So what you'll see is you'll see these cleared white bands,
meaning that's more protease activity in our sample.
So people have been talking about Cathepsins
being turned up in breast cancer and prostate cancer.
But the way that they were quantifying it,
they were doing Western blots.
So they were getting the proform mostly,
because the mature form doesn't stay around so long.
So a lot of proform.
Or they were looking at the mRNA.
And all those things don't really get to the activity.
So we said, well, let's test our new zymography protocol out
with actual human breast tissue samples.
So we actually purchased some breast tissue samples,
patient matched, so the same woman,
normal tumor, normal tumor, normal tumor.
And this is what I was saying,
it's off in the normal tissue, right?
And on in the tumor.
And Cathepsin K is the one at the top,
which actually had the greatest change,
with a 50-fold increase in these breast cancer tumors.
And we did it for a number of patients,
and again, we applied a dose curve to it
so we can quantify absolute value.
And again, what I love about Cathepsin K,
it's like you could draw a line here
and call it off and on, right?
And so it's a great little test that we could do,
and we compared it, sorry,
and in doing the statistics on it,
it had 100% sensitivity and 100% specificity.
Cathepsin L was good, but it wasn't at 100%,
because some patients had higher amounts.
And the other protease that everyone loves
are these matrix metalloproteinases.
Anybody heard of these, MMPs?
Everybody loves those.
Those actually have a lot more homeostatic function,
and so they only change by like two to three-fold,
and your sensitivity for those was much lower,
as far as predicting the tumor sample.
So this is exciting.
And so we published this,
and we actually expanded this to do breast, lung,
and cervical cancer sample,
and we did it over several different stages of the tumor.
And we actually see some interesting profiles
of how the cathepsin activity per tumor stage
actually does change.
So now we're showing for the first time,
yes, you can detect this mature enzyme in these tumors,
and it's off on.
I mean, binary, right?
Like what in biology is off on?
This is, I don't know, I'm more excited than you get that.
But the other trick is pharmaceutical companies
for a long time have been trying to develop
cathepsin inhibitors to use to target cancer metastasis,
bone metastasis, osteoporosis, arthritis,
but they've been failing clinical trials.
So they developed, so the organic chemists
really, really spot on binding to that active socket,
active pocket, and they get through phase one,
no problem, they get through phase two even,
and they don't get through phase two.
They move into phase two,
and that's where they fail due to side effects.
They're efficacious, they stop the disease,
but there's side effects, like rashes
and all kinds of other things that are,
had caused the trial to end.
And there's been more than 30.
Do you know how much money that is?
It's like more than 30 of them that have failed.
Now just maybe last year,
this one here has finally made it to phase three,
Merck, this is a Donica tip, which is,
I mean, they're gonna take over the market
when it gets through.
And so we were wondering, why are they failing, right?
Because these are great design active site inhibitors,
but I mentioned that they're quite promiscuous,
and so my belief is that there's a dosing issue,
because before, there's no good way
to quantify just the mature enzyme.
People are doing ELISAs, right?
And ELISAs, people know what ELISAs are?
Yeah, ELISAs are like western blots in a dish, right?
So it's antibody-based, but you still,
whatever you bind to, you don't see it.
You just get a color or a fluorescence
when it's all said and done.
You dose on that, you're gonna have way too much drug
in the system, because the proform is stable
at neutral pH, it's around.
You dose on that, and then you guys know
how kinetics work, you have too much,
it's concentration and affinity.
Even if it's lower affinity, high concentration,
it'll still bind.
These are things that we believe are happening.
So I got some, I now have some catheptin inhibitors
from GlaxoSmithKline that failed clinical trials,
and we're testing those out.
But why this is important is we have a great way
to quantify just the mature active site
that will bind to that inhibitor,
so we can maybe get some better dosing.
But the other thing we thought about is,
also, it's a great way to do diagnosis.
So I used to call it, I said to my lab group,
let's, we wanna make this like a breast cancer test
on a stick, right?
And, well, I said it differently,
but okay, I just said it now in public.
But then I talked to my breast cancer collaborator,
and I was like, hey, listen, we've got this thing
it's off on, and he loved it, he was like,
man, this is great, it's fantastic.
He said, and I was like, but I'm sorry
it takes overnight assay to test anything overnight,
and that it's probably too long.
He was like, are you kidding me?
You know how long it takes me to tell one of my patients
that he has breast cancer?
I was like, I've got like four hours or a day,
like I, you know.
He's like, no, it's about 10 days
when it has to go to the pathologist.
I was like, what?
He said, so if I could tell my patient the next day?
I mean, can you imagine these women
going through for 10 days?
I was like, no, that sounds amazing.
So what he does, he said, what normally happens
is the patient will detect the lump,
or maybe they come up mammogram positive,
go to a clinician, he'll do a biopsy,
send it to the pathologist who does all of these assays,
growth description, microscopic, blah, blah, blah,
then gets it back to the clinician.
And that is enough, a 10-day process, right?
With antibody, staining, labor, equipment, and upkeep.
So he said, so I'm really excited about your thing,
and I'd love to try to get this into the clinic.
He said, but I don't think they can do that
electrophoresis down in my lab,
or in the lab at the clinic, which I think they can.
But anyway, so what happens is the engineer,
we get the design principles,
and so now we've been going back to make it faster
and to minimize it so that he wants to be able
to take the biopsy in his office, the needle biopsy,
chuck it into the machine, process, or cook,
and then give him his answer.
And so these are minimization ways that we're working on,
which also led me to the micro and nano people
here at Illinois as well.
Sorry, my student made this animation.
So, so we want to improve diagnosis
and hopefully get earlier diagnosis,
but this article came out in the New England Journal
of Medicine at the end of 2012,
that said about after three decades of screening
for mammography and breast cancer,
maybe early screening isn't really doing anything
to stop deaths due to breast cancer, right?
Which is kind of a big bummer.
So there is substantial over-diagnosis
accounting for nearly a third
of all newly diagnosed breast cancers,
and that screening is having, at best,
only a small effect on the rate of death from breast cancer.
It's kind of a bummer.
And so with the assumption
of this constant underlying disease burden,
only eight of the 122 additional early-stage cancers
were expected to progress to advanced disease.
So what's happening?
Women are getting early diagnosed,
and a lot of women are undergoing
prophylactic mastectomies and saying,
I don't even want to take it at risk,
or my aunt had it, my mother had it.
Like, you know, do the double mastectomy.
And some women with the BRCA2, with the gene test,
they're getting the gene test, I'm at risk.
Like, we're done with it.
And that's actually, so some of these women,
what this meta-study is showing you,
that some of those women were not probably advanced
to the advanced forms of the disease,
and that they could have been saved
some of the hardship and the stress
of the double mastectomy,
the reconstruction surgery, and the rest.
But how do you know which women are gonna be at risk, right?
And then the New York Times said,
cancer survivor or victim of diagnosis.
Of course, they're gonna make it a bit more dramatic
in how to do it.
What's the best way to improve cancer survival?
Tell everyone they have cancer.
But I mean, who are we to decide how to make that choice?
So you look at these two famous actresses,
so Angelina Jolie, she just had the gene,
the BRCA1 gene, and she underwent a double mastectomy
when she found out she had the gene.
Now, her mother did pass away from breast cancer,
or ovarian cancer, I think, and those are kind of linked.
But just with the gene,
which is the possibility of the protein being on mastectomy.
And Wanda Sykes, she actually found a lump
through mammography,
and she underwent a double mastectomy as well.
Would they have advanced to advanced stages?
You know, who's to know?
So we started to ask ourselves that question.
Why does one person's cancer advance faster than another's?
So we've done several subtypes of cancer.
We know about the HER2, the ER positive,
the triple negatives.
We know about those, and you know,
we can treat the HER2 and the ER positives.
The triple negative is the tougher one.
But why does this happen?
What's different about these patients?
What's different about their environments?
What's different about their tumors?
And so it's easy with the clonal population, right?
And so, you know, but people are not oompa loompas.
So if you have the same person,
you can get the same kind of cancer, everything can happen.
But that's not how people are.
So patient-to-patient variability and diversity
complicates everything, right?
And as a clinician, that's what they have to consider.
You are my patient, now what am I gonna do for you?
So that's what we started to think about
these tumor-associated macrophages.
So as I mentioned earlier,
these other cells come into that tumor
and kind of help it grow and survive and thrive.
And one of the major cells types are these macrophages,
which are part of your immune system.
And in some later stage cancers,
macrophages can be as much as 50% of the tumor volume, okay?
And why I'm really interested is macrophages, of course,
pump out my babies, the cathepsins,
and they pump out these cathepsins
and actually help the tumor advance and grow.
And so there's a nice study,
this is actually a pancreatic tumor in an animal,
in a rat model, that actually showed that
as the tumor progresses,
you can see that the macrophages actually align
at the edge, the invasive front of the tumor,
and they're pumping out cathepsins at that invasive edge,
showing that these things are really playing a part
in helping the tumor advance.
And where do macrophages come from?
Macrophages actually come from circulating
white blood cells called monocytes.
So monocytes are circulating in our blood all the time,
ready to fight infection,
and if you have a disease or something happens,
it homes to that signal, enters the tissue,
becomes a macrophage, and starts to fight.
But when they get there,
they produce a lot of these proteases.
Macrophages also differentiate into osteoclasts,
which are these bone-resorbing cells I was telling you about.
These are fusions of monocytes
that fuse and form these osteoclasts.
So we started to think,
well, monocytes are easy cells that we can access,
and then monocytes become macrophages in the tumor.
So is there something about people's monocytes
and macrophages that are different that might,
how am I giving away the ghost?
Maybe we went, hold on, let me hold on that thought.
So we said, well, okay, we have this assay,
we can quantify these proteases.
The thing about these proteases,
they seem to be good biomarkers for the tumors,
but they are functional.
It is known that they help the tumor advance,
and they help degrade the matrix.
So they're more than just a side, right?
And they're things that people
want to target pharmaceutically.
Then we also, we're engineers,
so we can do some math to kind of look
at the matrix degradation over time.
We're also developing assays to look at this location
of these things in the tumor,
so we can look at distance,
and can we put together some sort of risk profile
of the advancement of the tumor?
We first started by gathering all of these.
We were bleeding lots of people.
These are actually mostly Georgia Tech students,
a few others.
We collected their circulating white bloods,
took their blood.
We isolated the monocytes from it,
and in a dish, we would turn them into macrophages, okay?
And then we looked at their cathepsin profile
of those macrophages,
and it was actually quite interesting.
We turned them into macrophages and osteoclasts also,
because we cared about both,
because osteoclasts help the bone metastasis as well.
And what was interesting is you could see this spread
over different people.
So this is from actually what's inside of the cell.
This is from what the cells have actually secreted
into the medium,
and you can see there's different values.
Like you see these people have very different amounts,
and if we quantify this for over seven patients,
you can see this wide range of variability
in the cathepsin activity
from different people's macrophages.
What was interesting is my student could go back
and get the same person again,
and the same person would like always make
high amounts of this,
and some people always make lower,
so it wasn't even just a factor of
were they sick that day?
Was it morning or noon?
And this was actually quite surprising,
and of course there was more spread
in what they were actually secreting out.
What's also interesting to note,
between the two different cell types
that the monocytes had become,
you do see a pattern,
and it's a different pattern,
but you can see they're all within this range.
So I do some systemology work also,
and we develop some statistical ways
that we can actually predict how much they make.
But again, our question is,
why does one person's cancer advance faster than the others?
And our hypothesis being that the person
who's making more of these enzymes,
will probably,
if the macrophages are making more of these enzymes,
make it help the tumor cell advance faster.
And here's the issue.
You can't say,
oh, we've got your blood today,
you have cancer,
don't get in treatment,
let's see what happens in six years
so we can prove my model correct, right?
Some people say that's unethical.
The ones that don't say that,
don't work with them,
so they go to jail.
And so, my students found this cartoon,
which I like,
but come back in six years and let's compare notes, right?
So this is the challenging part,
because we want to test our predictions,
but how do you test them?
So that's why we've been developing
these in vitro disease models.
So if you go to my website,
it says tissue engineering,
I don't really do tissue engineering.
We do tissue engineering disease models, okay?
Because we don't care about implanting it,
but they make great physiological systems.
And so again, our question was,
how can these different people's macrophages
that make different proteases,
help the tumor advance?
And so what we did is we collected donors
white blood cells from several different people,
turned them into macrophages.
So that's where your variability is.
Then we would stain them with a blue dye.
But then we take,
we had a breast cancer cell line,
MCF7, which is the ER receptor positive.
So this is gonna be like our oomphaloomphals,
where the cancer is the same
for all of these things
that we're gonna culture them with.
And these actually were genetically modified
that they express green fluorescent protein.
So they're green, our monocytes are blue.
And so we could co-culture those together
and then do a collagen invasion assay
to ask the question,
how do the macrophages help the cancer cells
invade into this collagen gel?
And so we co-culture them.
You can see this down here.
And then she'd let them sit for 24 hours
and we'd watch the cancer cells
actually start to migrate up
or the blue cells migrate up.
And what's interesting is,
and she developed this invasion index down here
to quantify the data.
So she was counting these cells by hand.
And it was interesting when she first started doing it
because without the macrophages,
the MCF7s really actually don't invade
this collagen gel so well.
But when you co-culture them with the macrophages,
which is the blue cells,
you can see them start to invade up
into the collagen gel.
And so we're doing this.
So we were doing this for different patients,
again, with the hypothesis
that if a person's cells make a lot of these proteases,
then they will help the cancer cells invade more.
And of course, what do we get,
which is what we thought we'd get,
is donor-to-donor variability
in these macrophage-mediated cancer cells.
So it is being quite different for the people.
So pop quiz.
So who would you think probably had the most perceptions
if you were in our school,
thought you believed everything I said?
Which of these patients?
What did you hope it to be
if you went up and do an ED round of publication?
Four.
Right.
Of course, science doesn't work out that way.
And so when we quantified the enzymes,
it was not four, it was three.
So three actually made the most amount of these proteases.
Right?
Now, don't kick me out yet.
We're gonna have the answer.
But three, this actually really surprised us
when she did this.
But then we went and looked,
because remember,
there's this whole proteolytic network that's happening.
These are only like two of them.
This is just from the secreted conditioned media.
But then we went back and looked,
because as I mentioned earlier,
our most nucleated cells make and produce
the endogenous inhibitor of these enzymes,
because the body does not want them to work
when they're outside of the cell
until it wants them to work.
So cystatin C is produced and secreted,
and it will bind to these enzymes
and block them from degrading matrix.
So Kanye went back and looked
and quantified the cystatin C
and check this out.
The one with the highest amount of protease activity
also had the highest amount of cystatin C
and the lowest invasion.
Right?
And so what she realized
is there's actually this ratio
of the protease to the inhibitor
that was actually the driving factor
and the greatest influence
on how much the invasion was.
And so again,
there's other models.
This model,
I'm not showing you that we did.
We can talk about that later if you're interested,
but just want to keep it focused for time.
So then we said,
okay, well, what's causing this difference of variability?
So we got more blood,
more people, et cetera.
And we did some signaling analysis.
Did you guys know about kinases?
So kinases are these intracellular signaling proteins.
I'll just leave it there.
And a major one that we saw
that seemed to be important was J and K.
We've shown in other systems
that J and K is activated
to lead to cathepsin expression and activity.
And so what she found out
is that we could,
when we did these like 12 patients,
there were these high cathepsin producers.
We called it kind of these mid-cathepsin producers
and these low-cathepsin producers.
Okay?
So this is just how much cathepsin they made
when we turned them agitated.
And she also, of course, quantified the cystatin C now
because we knew that was playing some sort of role.
And they're not directly correlated, right?
So you can't say,
oh, with the cathepsins up and cystatin C's up.
It wasn't like that for all of these people.
And so, but then we added our J and K inhibitor.
What was interesting is
the high cathepsin-producing patients
had a significant difference in their cathepsin activity
when we inhibited this pathway.
And some of the lower ones did,
but some didn't.
And the cystatin C,
some of them did, some didn't.
But it looked like this J and K
was actually having an effect
in reducing the amount of cathepsin they were producing.
So then, of course, we care,
well, what does that do for the invasion?
Because that's what we care,
how the macrophages are helping the cancer cells invade.
And so when she went back to do that,
I'll just jump to the chase here,
we can again see that the high cathepsin producers,
not all, so there's still some questions here,
this is the new stuff,
that about three out of the five of the top producers
were significantly impacted
with a reduction in invasion when we add this inhibitor.
And you can see here the actual images.
So this is without the J and K inhibitor,
and here's with.
And you can see the green cells are not migrating up.
So when we were discussing this,
we had the case, well, right,
this is the patient variability, right?
Now, her committee, they had different questions about it.
But that's the thing,
how do you know what's going on here?
So what we looked at, we put it all together.
And it looked like, again, DMSO was just the vehicle.
Here are high-producing cathepsin patients.
But you can see that the high-producing cathepsin patients
with J and K inhibition,
they actually dropped down to the level
of the other ones, right?
And the cystatin C didn't really change as much.
I mean, these two come down mostly.
And the invasion actually also starts to get them
within the range of the other enzyme,
or the other patients.
And so this, the PI3 kinase, which is a different kinase,
actually did not have that same effect.
So it looks like by adding J and K,
we're at least bringing down the cathepsin levels
to the level of the other patient.
So is this a way that you can now normalize patients
to some baseline level
that then you can maybe start to treat
with your cathepsin inhibitors
if they're now down to this normalized level?
These are things that we're starting to consider.
J and K could be some adjuvant kind of treatment
with the protease inhibitor to block it.
And again, we do this,
system biology is basically
principal components analysis here,
where you can see these patients
with just their vehicle, no inhibition,
are kind of all over the place, right?
But when we add the J and K inhibitor,
you actually put them all to the same kind of phenotype,
which we think is interesting.
Are you getting everyone to the same baseline starting point
that now you can add additional treatments?
And again, a different kinase inhibitor
didn't actually normalize them to that same point.
So this is what we're thinking about now with this.
And again, we're still trying to get,
but again, our big point is,
so that's one way for treatment,
but we still want to get to the prognosis.
So how could we see if we could tell a person
by the activity of their tumor or their macrophage
that they're going to have a very dangerous
type of advanced metastatic tumor?
So that's where the kinetic modeling comes into play.
And so to talk a little bit more about the matrix structure,
because again, this is the nanospace,
we think about these proteins being degraded
outside of the cell.
So this is a collagen fibril.
So type one collagen is like a nutrient
and that's what all these beauty products
want you to build more collagen in your skin,
so it stays taut and fresh or whatever.
So here's a picture of a fibroblast
and there's collagen fibrils running parallel with the board
and also cross sections coming out, right?
But these are things that will be degraded over time
as this tissue becomes remodeled.
And so again, we in my group,
I think we did this paper in Journal Club last year.
And I think before this, the students were still thinking,
you cleave it once and the whole thing comes undone.
So these are these macromolecular structures
that are formed from one protein peptide
that had to be synthesized, but then they do self-assemble.
And so there's this process of degradation over time.
And again, cathepsin K,
the most powerful hemi-collagenase we have in our body, okay?
So I just want to give you this orientation
and there it says nanometer, so I fit into the nano space.
Okay, great.
But this paper is great because it actually shows
how it's degraded over time by just cathepsin K.
So the fibrils start to get loosened.
There's gags that keep them connected.
And then you can see these fibers, fibrils, bundles, etc.
And then you start to see those actually start
to get unwound and actually degraded by cathepsin K.
So we actually want to do this computational kinetic model
because we know the cells release these enzymes.
And if you think about the kinetics,
then if you know how much protein you have
or proteins you have,
which we can quantify with our zymography,
we can know the composition of the tissue
because people have done all this matrix composition
of different tissues.
Can we put that together and see if the tumors
in this type of tissue,
this is how much matrix it can tone over per unit time
where maybe the cells can escape.
And so when thinking about the kinetics,
we'll take a single cathepsin for ease,
secrete it as this pro-cathepsin that's not active.
Then of course, there's some rate
at which it's going to be activated,
which will depend on the microenvironment, etc.
Then there's some rate at which,
or it could bind to the inhibitor.
If that inhibitor's statins is present,
it'll bind to it, takes it out of the equation.
But if it doesn't, it can bind to the substrate
where it can either cleave the substrate
and release product or release it.
That's got that hill.
And then there's also some rate
at which they are degraded or denatured,
depending on the pH, depending on other factors
in the environment where they just stop working.
And so these other constants are relatively well-known
in the field.
So the protease people, which I love those people,
they're all biochemists,
and they've got these kinetics all figured out.
But these other two seem mostly to be context-dependent
because there's a lot of things that can determine
its activation or its degradation.
And so we're engineers,
so we can define these governing equations.
You can develop your system of differential equations
that describe these,
and you can solve these simultaneously in MATLAB,
and you get your nice curves
when you can fit it to experimental data.
This is just a synthetic substrate
that capacitocate likes to degrade and gives fluorescence.
So that's great.
But the biochemists are great
because they dig one enzyme at a time, right?
That's fantastic to do that one enzyme.
But the cells are releasing several at one time.
So what happens when several are released?
But what's the trick about releasing several?
They all do not have the same catalytic rate
towards the collagen one.
They have different binding affinities,
and so they have different kinetic properties.
And what this paper nicely shows,
you can see how CatK can degrade that collagen fibril.
This is actually only a 14-hour picture.
The rest of these are 20-hour pictures, either.
But you see, they still are nowhere near
as good as the cathepsin K.
So they have different rates and different responses.
And again, cathepsin K is so good
because not only does it cleave it at the ends
where everyone else does,
it also can cleave it in this intrahelical region,
which makes it so special.
And also, when it cleaves those things,
again, collagen one has the greatest
tensile strength protein in our body.
That's why it's the predominant one in bone and your skin.
But as you start to cleave it with cathepsin K,
you can see its mechanical strength is greatly reduced.
But you can see when the other cathepsin are working on it,
they can maintain it a little bit longer.
So it's playing a part in several different parts
of what's happening with the spinal
and that tissue structure.
So we got bold.
So I said, okay, well, the biochemists are only doing one.
I don't know, maybe they're scared.
I don't know what's going on.
We're gonna do all four that we care about.
We're gonna do four,
and we're gonna do it with three substrates.
We're gonna have the inhibitor,
and we're gonna develop this model
because this is gonna be what is happening
outside of these cells in these kind of cancer environments.
With different kinetic rates, we're gonna do it.
So first, I did the experiment, and it got crazy.
Okay, so then I ate the humble pie,
and we backed up and said, well, let's just do two.
We'll do one, we'll do two.
Two is good, two is ambitious, sure.
So we said we're gonna do two,
and look at if we combine these two,
then let's look at the matrix degradation.
Because you would imagine, right,
that if you have two enzymes, substrate is not limited,
right, that you'd have the addition of one and the other,
that you would have this additive effect
of matrix being degraded over time, right?
More matrix, or more enzymes there to make more product.
People say they didn't think that
after I gave this talk before.
They're like, okay, but you would think that.
So here's cathepsin K by itself, here's cathepsin S.
This is them degrading elastin.
I'm gonna get back to collagen in one second.
So then we defined our model,
and this is actually real data,
and then we let the model predict
how much total elastin would be degraded,
which would be some additive amount.
And then we did the actual experiment,
and that's all the elastin that we got degraded
when we put the two together.
We were like, what's happening?
So me and my student were tossing ideas back and forth.
Did I mess up?
Did I do this?
What's going on?
And we actually then said,
well, let's just run a Western blot
to look at the protein accuracy.
Like, people just haven't done that.
In the field, they do these kinetic assays,
and the enzymes lose activity by about two hours.
And the field just says, well, they just lose activity.
Well, why?
Maybe they're oxidized.
Maybe they just lose activity.
That's what they do, right?
We need to be experts.
So we said, well, let's take this
and actually see the protein that's there.
And this is actually what we saw,
which is quite interesting.
Over that two-hour period,
cathepsin K was actually gone.
This is when they were incubated together.
In cathepsin S, the protein is still around.
But if you incubate cathepsin K by itself,
it stays around for about 30 minutes, right?
So it looks like putting the two together,
you actually get the disappearance
of this one really powerful enzyme.
So that's when, and then we did our zymography,
and you see the same thing, a bit more sensitive.
So you see K out to 60,
but still, it's disappearing, and the S is present.
So that's what we did with cathepsin cannibalism, right?
That was the new hypothesis we wanted to test,
where one of the cathepsins is eating the other
instead of the substrate,
thereby reducing the total amount of enzyme present
and getting less degradation.
So to test that hypothesis,
we went back and added these terms into our model
to add these degraded terms.
I'm sorry, because what I didn't mention,
because we couldn't get the model to fit.
Like, we could not get the model to fit in this structure.
There was, the kinetics just wouldn't work.
But in adding the cannibalism terms,
we were actually really able
to get a really nice fit of the model.
Now, no model's good until you test it a priori, of course.
But the other question came up.
Is there some therapeutic benefit?
These cathepsin inhibitors have not been worked.
They've been fairly clinical trials.
Well, can you use one cathepsin to degrade another,
clinically or therapeutically?
Let me not say clinically.
Because again, cathepsin S does not degrade collagen
very well, as you saw in those images.
But K degrades it like gangbusters.
But S will eat up K.
So if S eats up the K,
well, you have all your collagen protected.
So we said, let's increase this ratio,
which is gonna be a lot of money.
So I said, well, we should model it first,
so we know about where it needs to start with.
And it turned out that at a tenfold ratio of S to K,
you could get this protection of the collagen,
a loss of collagen degradation,
and you get increased cathepsin K degradation in that system.
And then we did the actual experiment,
and it was actually nice to see that cathepsin S
does actually protect the collagen
from cathepsin K degradation.
So this is just a Western blot from cathepsin K enzyme.
You can see these increasing ratios here,
where at the tenfold ratio,
you see the significant drop in cathepsin K.
We can do the zymography,
which we see both at the same time.
And again, tenfold ratio, loss of K,
high amount of S.
And then if we looked at the collagen one,
stated by SDS-PAGE, you see again, tenfold ratio,
all that collagen that we had at the beginning
is actually still there.
This was really cool, right?
And so we can quantify that and show that, yes,
one can actually eat the other
and protect the matrix protein that was present,
which then makes it back in a tumor
what's really happening in that tumor space.
And so we published this,
and the first journal didn't like
the term cathepsin cannibalism, so fine.
And the next one did.
So we got a JBC, which was awesome.
But again, it adds this new thing to protease kinetics,
which people aren't really considering,
is that we always think they just eat substrate,
but actually, one of these enzymes
can become the substrate to the other.
And again, when you take that out of the equation,
you're changing your total protein degradation.
Why is that important?
If you still have a lot of this
pharmaceutical drug as inhibitor,
you now even have less of your target,
and you might still now be over-targeting
the other ones with the nonspecific binding.
So we got back bold again,
and said, okay, we're gonna go back again
for the big map.
But now it gets a bit daunting,
because now we actually,
we already were considering substrate degradation,
but now we have to add in the cannibalism rates,
because these are not intuitive.
So we can't just say, okay, we'll bind the,
we don't, because we don't even know
which ones eat each other preferentially.
So that's why we've been doing some bioinformatics
based on their protein structure to figure that out.
But we now have to include
all of these different rate constants as well.
And again, we use the models
to actually kind of help make the experiments easier.
So again, it's the age-old question,
which we all ask since we were kids.
Who is the right one for me, right?
And so we can develop different scenarios
that we can make computational models of,
so that we can go back and test
and see what actually matches the experiment.
And again, we could say, okay, let's run this one
and see which one actually best fits the experimental data.
And this will probably be the best one
that's happening in the real system,
and then we're gonna test that one.
So we always relate pop culture to science,
I'm like, okay.
And so I think, actually, this might be me wrapping up.
And so again, the whole big picture
is can we sample the tumor,
or can we sample the cells that make into the tumor,
quantify the protease activity there,
we know the matrix properties of that tumor,
such that we can put it with our kinetic models
to predict how dangerous a person's tumor.
Oh, I was supposed to do the last one.
But right, so the big conclusions here for this whole talk
is that the ECM matrix degradation
does play important roles in tissue homeostasis
and in disease progression.
The matrix provides cues, it's not just infrastructure,
it's biochemical, biomechanical cues that it provides.
But in disease, these things get destroyed and degraded,
and the cells start to do different things.
But there's also patient-specific differences
in proteolytic activity that also could be modulating
this matrix degradation and invasion.
And by integrating these measurements
with predictive kinetic models,
we may be able to provide
some informed personalized therapies.
And that's where we're going.
So this is my group.
They're a great bunch of people.
This is us at the Sickle Cell Run,
we do this every year to support the Sickle Foundation.
And I'd love to thank my funding,
and here's my website, here's my email if you have questions,
and I'd love to take questions from you now.
Thank you.
Now we have questions.
Yeah.
I've read that for cancer,
well, first of all, it's a great talk.
Oh, thank you.
And I don't have a background in cancer biology.
Okay.
For cancer metastasis that the PCM aligns,
as opposed to being in a random arrangement,
do you think that it has,
the enzymes have a role in aligning the PCM?
That's a good, that's a very good question.
I was trying to fit some movies into here,
but it just got to be too long.
Now there's some nice live imaging movies,
Pete Frito, that has actually done this,
where he's actually shown that the cancer cells,
as they actually are squeezing through
and starting to metastasize out,
they will do this local proteolysis,
and by loosening it up,
it allows the matrix to be straightened out
as the cells continue to push through,
or as they grab it to move through,
it actually starts to align the fibers a bit more.
And why they do it,
at least how his work has been going
over the last couple of years,
is that because then you get these trails of cancer cells,
all following along these tracks,
to kind of get out,
and it makes it where one doesn't,
each cell doesn't have to do its own proteolysis,
they just follow the lead guy who's clearing out the way.
So that's been, and he's shown that in animals,
like with intravital microscopy and some other things.
So Frito, F-R-I-E-D-L, Pete here.
Yeah, really nice papers.
And his movies are online too, so check them out.
You have a question?
Sure.
Since there's a mechanics part to this summer institute,
and you're showing that different degradations
will affect the mechanics of it,
have you looked at that?
I mean, how the mechanics of this degradation
might also contribute to metastasis?
So we are not, I know people who want to,
we actually are not yet.
Now other people have shown, so we are not, okay?
But for different reasons,
because yeah, we have to do the mechanics stuff.
But other people have been looking at the mechanics.
And it's funny, depending on the type of tumor,
then some people say that,
because a woman will feel the stiffness of the tumor,
but it's mostly just that the cells are so together.
I don't think it's about the matrix themselves.
But other people have shown that on stiffer surfaces,
the cancer cells do migrate faster or invade faster.
So a collaborator of mine at UMass,
she makes materials, she makes biomaterials.
And so we've been putting together,
she makes these bone marrow mimetics
that can have all of the protein
that are in bone marrow, the different compositions.
And then in my lab,
I had a postdoc who was making like bony surfaces.
So then we could put the two together
and we were looking at migration on those surfaces.
And on stiffer surfaces, the cells do migrate faster.
Interestingly, she's also shown, she has a mojito,
she's searching it around,
that when she inhibits JNK in those cells,
they migrate slower, right?
So I love JNK for all these reasons.
But we haven't looked at the degradation
in response to stiffness.
So that's what I should have said first.
We haven't looked at that yet.
Yes, sir.
When you're adding the S to the T and the K,
isn't it a bit dangerous to do that?
Because if you don't add it in the right concentration,
it seems like the S actually helps out the K.
Well, so right.
Okay, so it's also, that's a really good question.
I have one slide, but I took it out again.
What we also show is it's also become,
it's going to become organ or tissue dependent
or specific how you would do that.
So for bone, right, for osteoporosis, right?
It's like, I don't want to say the number,
but it's probably 95% or greater type one collagen.
Like that's the big thing that gets messed up.
It's not all the elastin in your bone.
So in that context where there's all collagen one around
and K is degrading all that collagen one,
maybe adding a bolus of S or using a nanoparticle
to deliver it locally so you can control its delivery,
maybe that would work in a bone environment.
But in the artery wall where there's elastin and collagen
that gets destroyed during plaque formation,
it's totally not going to work
because S eats elastin just like K does, right?
So then that would not be a good use of that.
But I think there's some context dependent ways
that it can become important.
Plus, over some length of time,
that S is going to become inactive as well anyway, right?
So you can get your, you got to take out these enzymes
while they're there, and then the S won't work
on the matrix and won't continue to work.
Maybe you have a time window.
And that's why I think some of the nano strategies,
like again, time delivery, time release
could actually maybe come into play and be quite helpful.
But it's funny too, I'm sorry, I'm going to keep talking.
It's funny, S, we've now done a lot of experiments
looking at the rest of that network.
S is the big bad one who eats all of the other ones, right?
So it's interesting to see that S eats all,
and S is the one that maintains activity at neutral pH too.
So now there's the engineering application side,
but then I go back to my scientist
fundamental question side.
Is the body using these things naturally or natively
to kind of keep things in order
that we just haven't explored yet?
And so that's kind of one of the other directions
that it could go where we're open to both of those.
Yes, sir.
My name is Naeem, and I mean like when,
I mean supposedly when it happens,
it's cancer can kill people more quickly than,
you know, for cancer, I mean, once the host dies.
Right, that's, okay, I like what you're thinking.
So I tell my students, so again,
I come from the biology side,
so I talk about biology,
and I talk about cells are green systems, right?
They don't do anything to damage them,
they don't make damaging things for them,
they reuse everything, they're really smart.
So then you get this cancer,
and they're like, but cancer's out of whack.
Well, the cancer cell is doing what's best
for the cancer cell, right?
So why do they metastasize, as I understand it,
to be playing, as the tumor is growing,
like there's several mutations you have to have
for that tumor to become a successful tumor, right?
And as it's growing, so the first step, of course,
is it recruits these blood vessels
because it needs more oxygen.
So in a lot of early primary, not early,
but in primary tumors are this dead necrotic core
in the middle because the oxygen is not getting to it, right?
But as it tells, as it's starting to starve,
or some of these inner cells are starving,
it's releasing signals to bring more blood in, right?
So it'll recruit these blood vessels,
and they're really weak blood vessels,
but they bring enough blood to oxygenate the tumor,
but they're really loose and leaky,
and that's when the cancer cells can get in.
Now, what's crazy about cancer cells
is they should not be able to survive in the sheer,
like all that sheer stress in your blood.
Like cells are supposed to be one place,
and when they get other, like they shouldn't be there.
But they can make it because of the mutations
that are, like many cancers we know
have these mutations that say don't die, right?
Even though you're getting all these cues
to apoptose and kill yourself, they ignore all those things.
Then they still get to land somewhere
and enter into a harsh environment.
So like the liver, like there's lots of liver metastases.
The liver is so vascularized.
Liver is a harsh environment.
It's detoxifying all the poisons in your body.
And this cancer cell is hanging out
right next to the toxins, right?
And eating up nutrients that make it, sorry.
So it's, I think the cancer cells,
cancer cells just, mm.
So why are they doing that?
For first, it's just they have to leave
the home environment.
There's not enough space, right?
Or that they just have baby,
I don't know if it's even targeted,
that they've fallen into the massage,
or they've just broken in
because they're producing things to get more space,
and then they just start going places.
That's the short version of it.
I don't think it's a plan to go somewhere else
because they're just making it, is my thought.
So how much of your work relates
to the cancer cell leaving the initial site
to versus how much of this work
will be applicable to where it attaches?
Yeah.
So right now we focus on leaving the primary site.
And the real reason we do that
is because we cannot get our hand
onto the metastasized lesions.
Those are like, I don't know who's got those,
or they're not sharing those.
So we do play, we focus more on the primary tumor,
but in our bone mets model,
so we're making disease models of the bone metastases.
Like I told you, my collaborator at UMAC,
they were culturing the cancer cells with osteoclasts
that we've derived from monocyte
to kind of get a handle on what happens
when they get in that space.
And really we're doing that
just because there's so few treatments for bone metastases
that we wanna be able to test some of these delivery,
targeted delivery to that bone met environment
if it's even helpful when you have the cancer
working with the osteoclast in those spaces.
So that's where we,
we're focusing more on the primary tumor side now
because that's where we think
with the macrophages coming in for sure,
that's playing a part in helping you focus on that.

