Cell Mechanosensitivity and Metastasis
2 Video Views·Sep 5, 2023
Xin Tang -- Cell Mechanosensitivity and Metastasis
Advisers: Taher Saif and Mark Kuhlenschmidt
This is the last talk of the session is by Xing Tang.
He's my graduate student, has been in my lab
for the last about four years now.
He came from aerospace engineering,
so I thought that what can he do in biology,
but he started to learn cell culture,
and then he got interested in cancer.
He started working with Professor Mark Kulinsmith,
who is in pathobiology, and then he started to look
at what happens when cancer cells go to metastatic stages.
The initial idea was to find out
whether the cell cell adhesion decreases
as metastasis happens, but then he figured out
that there is much more interesting things happening
beyond just cell cell adhesion.
So that's what he's gonna talk to you today.
Thank you.
Okay, good afternoon, everyone.
My name is Xing Tang, and today my talk
is mechanosensitivity and cancer metastasis.
First of all, I would like to thank my advisor,
Professor Tahir Saif, and Professor Mark Kulinsmith.
And this work is in collaboration
with Dr. Teresa Kulinsmith, Dr. Josh Zhou,
Mr. Philip Bill, and Professor Fei Wang.
First of all, we know 90% of the cancer deaths
are due to the cancer metastasis.
During the cancer metastasis, the malignant cancer cells
firstly make the decision to detach from the primary tumor,
and then those cells downregulate their cell cell adhesion
and upregulate some cancer-invasive proteinase
to invade into the circulation system.
Driven by the flowing blood, those malignant cancer cells
travel through the human body
and reach various distant organs.
Then those cancer cells escape from the blood vessels
to settle down in those distant organs.
Then they generate numerous secondary tumors.
It is those numerous secondary tumors kill the patient,
not the primary tumor.
If the primary tumor is detected on time,
they can be cut, and the patient will cure.
But if there are numerous tumors,
numerous secondary tumors generated,
there are just too many tumors to be cut.
The problem in the problem to cure the cancer
is mainly because we don't know which signal
causes the malignant cancer cells to make this decision
to detach from the primary tumor at the first place.
So in our work, we discovered that by only changing
the mechanical environment alone,
the cancer cells can make the decision
to become more malignant and to prefer
some metastasis-like phenotype and behaviors.
The experimental systems are as follows.
We used the substrate with variable stiffness
to culture the cancer cells.
Those substrate stiffness are varied from 1 kPa
as softer as human brains to intermediate soft 20 kPa
as softer as the human muscles,
to very stiff substrate like 3 kPa polystyrene substrate.
This is conventionally used for the cell culture.
We choose the human colon cancer cell line,
HCT8, as our model.
The colon cancer cells, HCT8, is obtained
from the primary tumor of the human patient,
but not metastasized yet.
It is a cell line.
We use the normal cell, MA104,
and the bovine endothelial cells as the control.
To make the identical chemical environment
on the gel surface, we use the fibronectin,
laminin, or collagen as the extracellular matrix
to enhance the cell adhesion.
We only change the stiffness of the substrate,
mechanical stiffness of the substrate.
This is what we observed for the normal cells,
for the MA104 and the bovine endothelial cells.
When the normal cells are cultured on the softer substrate,
like a 20 kPa substrate, they form the clusters.
They left some blank area behind them
and form the clusters to form the cell-cell junctions.
When they are cultured on the harder substrate,
like polystyrene substrate, as you guys already know,
they form a monolayer.
And due to the contact inhibition of the normal cells,
they stop proliferate until the confluence is reached.
We did the same thing for the cancer cells.
We find on the softer substrate,
they also form the clusters.
On the harder substrate, they form the monolayer.
At the first look, they behave the same,
normal cells and the cancer cells.
However, if we keep looking longer,
something interesting happened.
Consequently, we observed on the seventh day of culture,
there are some single cells detached
from the edge of the cell colony and migrate away.
This detachment keeps going until after 10 to 14th day,
the whole colony falling apart into numerous single cells.
And those cells keep proliferating.
In contrast, on the very stiff polystyrene substrate,
the cells keep in the monolayer.
And since they are cancer cells,
they don't have contact inhibition,
they start to pile up on the monolayer.
But they never show this dissociation phenotype.
So when we observe this phenotype first time,
our first question is, is those really the cells?
Are they just dying?
So we did the live-dead assay
to test the viability of those cells.
We found those dissociated cells actually
are living even better than the cells in the colony.
Those blue, green color indicate the life
and red indicate the dead.
Not only alive, those cells, dissociated cells,
are also mortal and metastatic.
Here we show the migration trajectory
of those dissociated cells.
They migrate away from the cell island, cell colony,
with the maximum speed of one to two micron per minute.
Here I show the time-lapse video
recording the dissociation process of a small cell cluster.
So you can see, within seven to eight hours,
the cells are dissociated completely from the cell cluster.
This picture shows a zoomed-out view
of the dissociated cells on the whole gel.
The diameter of this gel is about one centimeter.
So here you see, this is after 14 days.
I'm sorry, the top of the caption is cut off,
but this is after 14 days culture
of HCT8 cells on the 20 kPa gel.
So remember, we start up with the epithelial phenotype cells
and now we end up with all the rounded-shaped single cells.
From now on, we call those epithelial phenotype cell E cells
and the rounded dissociated cells, R cells,
to make it simpler.
Then we harvest those R cells as well as E cells
to perform the invasion assay.
What we use is the basement invasion assay.
We culture those E cells and R cells
on the top of a 100 microsecond basement membrane.
We culture the cells with the media without any serum
and put the media with serum beneath the basement membrane.
So due to the chemotracting,
those cells will penetrate the basement membrane.
After two days, we find at the bottom
of the basement membrane,
the R cells penetrate 15 times more than E cells.
These surprising results told us
this cell dissociation process
is not a simple cell dispensation.
It's actually a cell-state transition.
After triggered by the appropriate
mechanical microenvironment,
the cells are actually transferred
to a more malignant cell state.
Then, the next question we ask is,
why those dissociated cells don't form the cell cluster
again on the soft substrate?
Another parallel question is,
if these cells can be driven to dissociate
due to the appropriate mechanical signal,
can they be reverted back
if we change the mechanical signal
back to the unfavorable signal,
such as the polystyrene substrate?
These two questions can be answered by one experiment,
which is to harvest the dissociated cells
from the polyacrylamide gel
and replete them back to the fresh polyacrylamide gel
as well as the fresh polystyrene substrate,
soft and hard substrate.
What do we observe this?
Very surprisingly,
after those dissociated cells are repleted back
to the fresh, soft, and hard substrate,
they behave identically.
If I don't tell you what substrate they are on,
you will not tell whether they are on the soft
or hard substrate.
Totally opposite to the original cancer cells' behavior,
on the soft substrate, they form the colony.
On the hard substrate, they form the monolayer.
Those dissociated cells totally remain dissociated,
remain around the phenotype,
regardless of the substrate stiffness.
This experiment told us two things.
One is the cell state transition.
This cell state transition is stable and permanent.
It can be reverted.
It can be triggered by the mechanical signal,
but it cannot be reverted by the mechanical signal alone.
Second conclusion we find from this experiment is
that these cells actually lost
the substrate stiffness sensitivity,
which means they are more automotive.
And that's an important hallmark
of the cancer metastasis.
As I said at the beginning,
during the metastasis,
especially at the onset of the metastasis,
cancer cells need to downregulate their cell-cell junction
to downregulate the cell-cell adhesion molecules
to fill the cell-cell junctions.
So to test whether the dissociated cells
also behave in the same way,
we stand the E-cadherin on both E-cells and R-cells.
E-cadherin is a well-known cell-cell adhesion molecule.
Here, the green indicates the E-cadherin expression levels.
On the top two pictures showing the E-cadherin
expression level in the cell colony,
and these second two pictures showing
the E-cadherin expression in the dissociated cells.
By estimating the fluorescent intensity,
they are downregulated five to six times.
We also verified this E-cadherin expression level
using the Western protein.
They are really downregulated.
Also, as the dissociated cells are replated back
to the polystyrene substrate,
even the mechanical microenvironment is changed,
they don't upregulate the E-cadherin.
They don't restore that surface cell-cell adhesion molecules.
So here, this experiment verified the downregulation
of the cell-cell molecules.
It only tells us the presence
of the cell-cell adhesion molecule,
but does not tell us the function of the cell-cell adhesion.
So that's not sufficient.
So to test the function of the cell-cell adhesion,
whether that is really filled,
we used the code-counter assay
in Professor Colin Smith's lab.
In this experiment, what we did is,
we firstly suspended the cells,
the R-cells, E-cells, and the normal cells as control.
When cells are suspended,
we put them into the identical series of vise,
and let them settle down in the 37-degree C water chamber
for a varied time, from up to two hours.
As the incubating time increases,
cells tend to aggregate.
In the code-counter machine, there is a small hole,
which is small enough,
just small enough to let only one cell to get through.
So at a different time point,
we pick up the cell samples
to let the cells penetrate the holes.
More cells penetrate through the holes
means more single cells remain in the tube, in the vise,
which means cells are less tending
to form the cell-cell adhesion.
So our data suggested, throughout the two hours incubation,
the R-cells remain constant.
You may notice the single-cell amount of R-cells,
mostly constant.
For the E-cells, they reduce,
and especially for the control, they reduce dramatically,
meaning the cell adhesion activity is going on.
Then I also developed a traction force microscopy
to study the interaction between the cells and the substrate.
The traction microscopy is used to measure
how much traction force cells can exert
on the substrate underneath them.
What I find is, for the E-cells colony,
they apply maximum 300 pascal traction force
on the soft substrate.
In contrast, after the cells dissociate from the colony,
they apply very weak traction force on the substrate.
The maximum is 100 to 150 pascal,
two to three times level down.
This experiment tells us the cells not only reduce
the interaction between their neighbor,
they also reduce the interaction to the substrate.
They become more, more isolated.
Basically, after the transition,
they stop listening to the community.
They behave by themselves,
which is a very important hallmark
for the cancer metastasis.
So recently, we have also performed
the deep RNA sequencing analysis
to analyze what gene might be activated
in the R-cells after the dissociation.
What we find is, in the R-cells,
there are about 90 genes differentially expressed
in comparison to the E-cells.
Among those 90 genes, many of them are related
to the apoptosis inhibition
and related to the cell migration promotion.
In particular, there are two genes
related to the cancer stem cells.
So all these results suggest, during this transition,
the R-cells not only show the phenotype change,
but also show the genotype change.
And noticeably, all this transition
is due to the mechanical microenvironmental stimulation.
So, as a summary, what we discovered so far is,
if the mechanical microenvironment is appropriate,
the HCT8 cancer cells can express
a serious cancer metastasis hallmark,
including the reduced cell-cell adhesion,
increased cell motility activity,
the upregulated invading capability,
as verified by the invasion assay,
the upregulated apoptosis inhibition gene,
as well as the cancer stem cell genes.
Also, we find that this transition is permanent,
and it can be triggered
by the appropriate mechanical signal,
but it cannot be reverted by the mechanical signal alone.
So all what I'm saying here is,
during the cancer metastasis,
the onset could be triggered by chemical signals,
but the mechanical signals may not be neglected,
and they may play as important a role
as the biochemical signals.
Thank you very much,
and I would like to take any questions you have.
Thank you.
Yeah.
So, on the soft cells,
did you image the E-cadherin at the junctions,
and what do you see in terms of over,
is that basically on the top?
Yeah, actually the-
It wasn't clear.
Yeah, this picture showing the E-cadherin
standing on the soft gel, 20 kPa.
This picture showing the dissociated cells
on the 20 kPa gels.
So both of them are the cells on the soft gel,
but at a different cell state.
These two pictures showing the magnified view
of those black box.
This is on the soft gel.
Yeah.
Yeah, so what I'm wondering is whether or not
you're seeing a decrease in the amount
of E-cadherin at the junctions,
or is the actin cytoskeleton,
what does the actin cytoskeleton look like
adjacent to those E-cadherin junctions on the soft gel?
Yeah, that's a very good question.
I didn't show the actin structure here,
but we did the actin cytoskeleton standing.
What we find is on the soft gel,
the actin stress fibers in the cell island is diffused.
Cells mostly show the cortical actin
around the cell periphery.
The only actin fiber showing up
is at the bottom of the cell colony,
but at the top of the cell colony,
they don't have the clear filaments,
which indicates the cell on the soft gel
actually are experiencing less intracellular force.
If you are talking about the co-standing
of the E-cadherin and actin,
we have not done that.
I did the, yeah.
You answered my question.
Okay, okay.
Yeah.
Xin, what a beautiful talk,
and such interesting results.
Am I, this is not a question, it's a comment.
A couple slides before this,
you showed us some, I think,
eight cells up in the upper corner,
and each of the, that one, no, next one.
That one.
Okay.
Now what's really interesting about this
is five cells, sorry.
I pick up a five random.
Emerging recognition that we're getting,
and many, many requests for funding
from the National Institutes of Health
that we've been thinking about cells and tissues
as being homogeneous.
We have liver cells, we have heart cells,
we have cancer cells,
but in fact, they're very heterogeneous
within a tissue.
And what this is showing me
is that this heterogeneity is showing up
under these culture conditions,
which is very interesting.
That's probably, in this case,
part of the transition.
And it is recognized,
as far as I know from my interactions
with Dr. Bhargava,
that this is also characteristic
of tumorigenic tissue.
So the point I'm making here
is look for heterogeneity.
That's sort of what people are now,
that's the future.
Yeah.
That's a very good point.
Very good point.
Just the group and single cell analysis.
Exactly, yeah.
So the idea that.
Thank you.
When a group of cells,
so from your subgroup, right,
is there a smaller subset,
one or two or three cells
that are really the ones
that are most metastatic?
Exactly, yeah.
We consider all of this.
Yeah, there might be some bad guys
to trigger this process.
Yeah.
I mean the.
So wasn't it,
I thought that one of the,
sort of the going hypothesis
was that the cancer tumor,
the matrix or the overall tumor
might be stiff.
But the cells are softer
and hence they escape the matrix.
Right, so here you're showing
that locally the matrix could be softer
and maybe it's the soft matrix
that triggers the cells to escape.
So which one is it?
Is it the cell or the matrix?
I think, yeah,
I think there is a terminology difference.
What I'm saying is softer here.
I mean the, here.
I mean one kPa is softer.
20 kPa, I mean intermediate is softer.
But this softer is in comparison
to three gigapascal.
If you compare these two
and compare them
to the physiological stiffness,
this is quite stiff.
Based on the literature,
these cells are from the colon.
The colon stiffness is about one to two kPa.
But when the colon cancer cells
develop a tumor,
the tumor generate fibrosis.
People, not trending,
we don't know.
For the fibrosis tumor,
colon tumor,
the stiffness is about three to four kPa.
They are, but not,
that's not metastatic yet.
There are no data showing
how stiff the tumor is
before the metastasis.
So what we observed is the trend
from the one kPa to 20 kPa.
If you keep the cells in this stiffness,
but not softer,
the cells show this cell state transition.
What we are right now working on
is we want to really find out
what's really the stiffness of the tumor
before metastasis.
That has, there are no data showing that.
Okay, so your work was 21 kPa
in the polystyrene.
21 k, actually we show,
we have a series of stiffness,
21 since it's optimal.
So in one or two kPa,
it doesn't happen.
It doesn't happen.
They form a tumor, but they don't.
That's the stiffness of the normal colon.
Yeah.
So what you're doing now
is actually getting the tumors
from the patient from Provena.
And then looking into those tumors'
stiffness as well as the cells.
But are you considering looking
at the stiffness of the cells also?
I did that also, yeah.
The, after the,
if you measure the single cell stiffness,
they are really softer
than the original cancer cells.
Yeah.
By how much?
By two to three times.
So after, they dissolve.
Very nice comprehensive study
if you can correlate all of that.
So you have done all the FM study.
That's fantastic.
I saw the time is only 10 minutes,
so I didn't put that in the results.
Yeah.
Sure.
All right.
Yeah.
Can you repeat the question?
I just heard about,
you talk about the bones.
I didn't hear the complete question.
Stiffer.
Stiffer?
Yeah.
Or metastatic prostate cancer,
which metastasizes to bone.
So have you thought about the,
I guess,
the transition that would occur there?
Actually, for the colon cancer cells,
they metastasize to the liver
instead of the bone.
I think the breast cancer cells
metastasize to the bone,
preferentially.
So the-
Breast cancer will degrade bone.
Yeah, as well.
Right, right.
I think bone stiffness is too stiffer
for the colon cancer cells.
I'm wondering if you think
that the mechanism is the same, though.
Okay.
Or do you think it's different?
Yeah, I'm not very clear about that.
Yeah.
But it's a very good question.
Good.
Okay, well, thank you very much
to all the four speakers.
Thank you.

