Integrating Mechanical Cues and Biomolecular Patterns for Tendon-Bone Junction Repair
Sep 5, 2023
"Integrating Mechanical Cues and Biomolecular Patterns in a Collagen-Glycosaminoglycan Scaffold for Tendon-Bone Junction Repair"
Laura Mozdzen, Chemical and Biomolecular Engineering
(Advisors: Brendan Harley and Amy Wagoner Johnson)
CNST Nanotechnology Workshop 2013
My name is Laura Mosden, and I'm here to discuss my research on how to create a collagen GAG
platform to regenerate the tendon bone junction.
So the tendon bone junction is a highly specialized tissue that, which connects the elastic, highly
aligned tendon to a more mechanically stiff and amorphous bone.
What's so special about the insertion region is that it can actually dissipate the stress
concentrations that are caused by the mechanical mismatch between the tendon and the bone.
And it does this through a complex series of gradients in both growth factors and extracellular
matrix proteins, along with a gradient in the mineral content, which increases the elastic
modulus as you move from the tendon to the bone.
So after injury, all these complex gradients don't actually repair themselves.
They are just replaced by scar tissue.
And even after surgeons go in and repair that site, you still can get a refailure rate as
high as 94%.
So the goal of our project is to see if we can do something better than just mechanically
fixing the tendon to the bone, and see if we can get some sort of better functional
response through creating a biomaterial that can regenerate the tendon bone junction.
So as I said, the overall goal of our project is to create one single construct, a collagen
GAG platform, with which we can seed adipose-derived stem cells from the patient's own body.
And then these stem cells will differentiate into a tendon region, a bone region, and an
insertion region, which can functionally repair the tendon bone junction.
So one of the things that I'll talk to you about today is this idea of patterning biomolecular
cues across this scaffold in order to cause this differentiation of the stem cells.
Another thing that we want to focus on is creating a mechanically robust insertion site.
Because right now, when you essentially staple or attach the tendon to the bone, you get
all these stress concentrations that cause refailure right there at that insertion region.
So we need to figure out how to dissipate those stress concentrations.
Our choice of material is a collagen GAG scaffold.
So we take a suspension of collagen and GAG, or we can also mineralize this suspension
for bone repair.
So you take this suspension, and you can freeze it down, and then sublimate away the ice crystals
to form a dry, porous sponge-like material.
By controlling the freezing properties, you can also control the pores that the ice crystals
leave behind.
So you can control size as well as alignment.
And here's a picture of that dry, porous sponge.
We can also, instead of freeze-drying our suspension, we can just let it evaporate to
form a two-dimensional membrane, which we can use to more easily screen cellular conditions
before we move into our three-dimensional platform.
So we've also shown that we can layer our non-mineral slurry on top of our mineralized
slurry, kind of like you would find at a bar in a black and tan.
So for this, we can create a multi-compartmental scaffold that can regenerate both the tendon
and the bone sections of a tissue, along with having a continuous interface—sorry, I think
my laser pointer's dying—along with a continuous interface where you can see our strands of
collagen actually are continuous.
So previous work in our lab has shown, or has looked to nature for inspiration.
And there's always this trade-off between, for a mechanically strong structure, but also
to have something that's porous for cell viability and nutrient transport.
So a paradigm that we find in nature is this core-shell composite, which consists of a
porous inside with a more mechanically strong outside.
So we can wrap our collagen gag membranes around our more porous scaffolds to create
the same kind of core-shell composite.
So here's an SEM image of it, which it looks pretty similar to the porcupine quill.
Oh, thank you.
And then—I like this one better.
Alright, so you can see that our core-shell composite looks pretty similar to the porcupine
quill.
And we've also shown that the tensile modulus of it can increase up to 36-fold while maintaining
cellular viability.
So our next question is, can we take natural inspiration not only for the whole construct,
but can we look at something that pertains to a mechanically robust interface?
And the answer is yes, we can.
There are structures in nature, such as in a turtle, in an armadillo, or in an arapaima,
which is an armored fish.
And all of these animals need their armor for protection against predators, but they
also need to be able to respirate, they need to be able to grow, eat, move, all of these
things.
So the way—or so what you find in nature is this interdigitation, which provides strength
in tension, in compression, and in sheer stress.
We also—or rather, Guy Jennings' group has also shown that there is some interdigitation
in the native rotator cuff as well.
So this is really exciting for our work.
So one way to measure the degree of interdigitation is through the suture complexity index, which
is essentially the total degree of interdigitation divided by the shortest distance across the
interface.
And you can also correlate this to the angle of the tooth.
So Mary Boyce's group at MIT has shown that you can actually model this kind of interdigitation,
and that in tension, there's actually an optimal angle where the tensile forces are evenly
distributed across this angle to create sheer forces.
So this is something that we would like to look at in a biomaterial and see if we can
create something that is stronger than simply a flat interface.
So in order to do this, we take a divider and—or sorry, a mold with a removable divider,
and we fill one side with a non-mineralized slurry and another side with mineralized slurry
for bone repair.
Then you can remove the divider and freeze-dry it.
And as you can see here, here's the mineralized slurry and here's the non-mineralized.
And you can see that there's a degree of interdigitation that looks like a series of teeth across that
interface.
So we can not only vary how many teeth we have, but you can also vary the diffusion
time to create more complex degrees of interdigitation.
So this is a micro-CT image of our interface.
And in order to measure the actual interfacial width, we set the top portion to be pure non-mineralized
scaffold, and the bottom to be mineralized scaffold, and then take a threshold of two
standard deviations to determine what the actual interfacial width is.
So for this particular case, it's approximately 500 microns.
In addition to just characterizing the interfacial width, we can also do mechanical testing on
these.
And as you can see, with increasing interdigitation, we get an increasing elastic modulus and an
increasing failure load.
Our next steps on this is to shrink it down to create a more biologically relevant size.
So one of the other things that we can look at is actually characterizing the tendon portion
and the bone portion, and eventually the insertion region in between them.
This is an alizarin red stain for mineral content, where in the tendon region, you can
see that there is very little alizarin red or mineral content.
And in the bone portion of our scaffold, there is a lot of alizarin red or mineral content.
And so our next question is, what can we see in the insertion region right in between these
and where our interdigitated scaffolds meet?
So this is an SEM of the flat interface, the one-toothed, and the two-toothed, although
this is only one of them.
And through EDS, you can actually get the composition of these, where this is the calcium
content and this is the phosphorus content.
And so you can see that that is actually localized to the bone section of our scaffold.
So far, we've been focusing on just this increase of mineral content and creating a
more mechanically robust interface.
But there's also a lot of growth factors and gradients of growth factors going on, along
with extracellular matrix proteins.
And this turns into a really complex system.
So instead of trying to look at all of this at once, we really want a simple platform,
which we can modulate the material stiffness and control the degree of immobilized biomolecules,
and then look at either signaling pathways, gene expression, or proliferation data, essentially
cellular responses of our stem cells using stiffness and biomolecular patterning independent
of one another.
So next, we want to see if we can make this tool.
So we have, so the first part is for, sorry, for benzophenone, sorry about that, for, in
order to get a correct cellular response, since our application is eventually to do
this in vivo, we can't just add our growth factor to the media.
We need to immobilize it onto our scaffold so that it stays relevant in the body.
So in order to do this, we can use benzophenone photolithography to pattern our biomolecules
onto our substrate.
So what you do is you functionalize the collagen GAG scaffold with benzophenone, and then
you essentially expose it in the presence of your biomolecule that you want to pattern
to a UV laser, which creates a bond between the benzophenone and your biomolecule.
It relaxes if there, if it doesn't create a bond, so you can actually create multiple
gradients and it, and the benzophenone photolithography is time sensitive.
So essentially this little cartoon will summarize what happens.
Here's our scaffold, you shine it with a UV laser, and you get biomolecular attachment.
So we've shown that we can pattern PDGF, which is a proliferation growth factor, at
different concentrations, or at different immobilization rates, depending on exposure
time to the UV laser.
We've also shown that through a common cross-linking method, we can control the elastic modulus
of our substrate simply by changing the ratio of EDAC to NHS to our collagen scaffold.
So these three are the one, these three ratios are the ones that we will proceed with since
they have statistically different elastic moduli from one another.
So our next question is whether these two, whether the manipulation of stiffness and
the manipulation of the factor immobilization are actually truly independent of one another.
So the UV exposure does not either degrade or further cross-link the collagen, so that
is independent of elastic modulus.
So our next question is whether the cross-linking actually affects the biomolecular immobilization.
And here are our ratios one, four, and five, and these are all statistically significantly
different than our no UV control.
So no, EDAC cross-linking does not affect BPO immobilization.
So we actually do have an independent, or an orthogonally independent tool in order
to modulate both factor presentation and the elastic modulus of our substrate.
So here are our experimental conditions for the material stiffness and our pattern biomolecule.
We chose PDGF for this one, and as you can see here, with increasing PDGF, or rather
with PDGF, either the medium or the high dose, we see increase in cell proliferation.
However, you don't see a drastic cellular response based on stiffness.
However, we did see that for osteocalcin, which is a bone morphogenic gene, with increasing
stiffness you get an increase in gene expression.
So our next goal is to, instead of taking PDGF, which as I say, since it's a proliferation
growth factor, we don't really expect to see it be stiffness dependent, we'd next like
to use BMP2, which should be stiffness dependent.
So just to give you a quick summary of my work, we've shown that through interdigitated
scaffolds we can increase the mechanical properties of our multi-compartmental scaffolds.
And I will continue to map the properties across the interface.
And then, as previously discussed, for this independent manipulation of stiffness and
biomolecular presentation, we will move on to BMP2 next, which we expect to see some
sort of stiffness sensitization to the cellular response.
And since this is an IGERT presentation, I just wanted to point out that this summer
I will be doing my international experience, and with Michelle Oyen I will map the local
properties across the interface, and with Hazel Screen I will be able to cyclically
load our scaffolds and see what kind of bioactivity our cells respond to.
I'd like to acknowledge my lab group, our collaborators, and our funding.
Thank you.
Questions?
Okay, I'll start with the first question.
Yeah.
So, I think digitation is basically just to increase the area of contact, is that right?
Yes.
That's the only reason?
No, it's a little bit more than that.
As Mary Boyce's group showed, there's a point and an optimal angle that you can find, at
least in a modeling situation, where the tensile forces actually can change.
Instead of being pure tension, they turn into shear forces so that they're evenly distributed
across that angle.
So I think it's a little bit, it's probably more of the shear forces taking over.
Okay.
For the question, let's thank Laura one more time.

