The future of transparent tissue
30:03 Share

This is Stanford Engineering's The Future of Everything, and I'm your host, Russ Altman. I thought it would be nice to revisit the original intent of this show. In 2017, we wanted to create a forum to dive into and discuss the research my colleagues across the campus are doing in science, technology, medicine, and other areas.

My goal was to show you that these are people working hard to improve the world. The university has a long history of doing work to impact the world, and it's a joy to share with you the motives and the work of these colleagues as they try to create a better future for everybody.

I hope you'll remember that when you think about universities and their role in society. I also hope you'll walk away from every episode with a deeper understanding of the work that's in progress here and that you'll share with your friends and family as well. What we want to hope to achieve is that we don't have to cut open the tissue, we don't have to insert an optical fiber, we don't have to insert a microendoscope, but we'll be able to make the tissue transparent by itself.

This is Stanford Engineering's The Future of Everything, and I'm your host, Russ Altman. If you're enjoying the show or if it's helped you in any way, please consider rating and reviewing it. Please consider giving us a 5.0. Your input is extremely valuable, and it helps others discover what the show is all about. Today, Guo-Sung Hong from Stanford University will tell us that he has figured out how to make living tissue see-through. Yes, transparent. It's the future of transparent tissue.

Before we get started, another reminder to rate and review, particularly if you learned something new or find this podcast to be stimulating.

Well, you know, you look at a living organism and it is not see-through. It is not transparent. Sometimes when we were kids, there were those models made out of plastic where you could see through the transparent skin and then see the guts, the heart, the lungs, the intestines. Well, guess what? Scientists have figured out how to make tissue, living tissue, and also tissue.

dead tissue, how to make that see-through. It's related to a complex physical phenomenon where by manipulating the physical properties of both lipid and the water that it interacts with, lipids are fats, water is water, they interact and that often causes the cloudiness that leads to not being see-through.

Well, Guo-Sung Hong from Stanford University is an expert in physics, material science, and biology, and he and his colleagues have figured out how to add something to the water so that it's similar to the lipid and it all looks see-through. It first worked on a chicken breast, and then they moved to mice, and it's an amazing story.

Guo-Sung, you're a specialist in material science, and yet you work on biology and the brain, which might not be the first thing that people think of when they think about materials. So can you explain how do you bring these things together and what's your interest? So my interest actually started from when I was a graduate student at Stanford. I actually was working on using...

some very exotic nanomaterials that are called carbon nanotubes that have very interesting emission properties that allow us to see deep into the brain. So that's all my interest in brain science studies. And I moved to Harvard to continue my postdoctoral research.

And then during that time, actually, I had the privilege of working on using flexible materials for brain-machine interfaces. That's actually further strengthened my knowledge and interest in using material science for brain science. And then coming back to Stanford,

Now my lab is mainly focused on developing new tools for minimally invasive imaging and new modulation, but also for neuroscience and biology in general. Great. Now, one of the things that has gotten a lot of people very excited is this work. I don't even know how to characterize it, but making some tissues be see-through, like transparent. And so this is amazing.

And I know it's very complicated, but I wanted to try to understand how it works. So can you take us through what is the technical reason for even doing this for the science? And then what do we need to know about the physics to understand how this works? Yeah, sure, Tony. So...

The motivation behind this work is actually for me, for us, is actually the challenge. This is actually a quite universal challenge that is present for all light-based methods used in biomedicine. So we know that the light actually does not penetrate very deep into the tissue. So this can be attested by just looking at your hands. You can see the surface features, but you cannot really see through. That's because our body actually made of, you know, basically it's a

a giant bag of water, 70% water, but it also has a lot of teeny tiny lipids structures such as the membranes and also all the organelles made up of lipids and also proteins. And all these components are actually having a refractive index much higher than water. So then when we have components

components of the tissue with different refractive indices mixed at the microscopic scale, it necessarily makes the tissue actually scatter light as the light goes through. So then the scattering really limits a lot of the applications in biomedicine. For example, we use fluorescent imaging to understand, to visualize the structures and understand, to study the activity and the function

but fluorescent imaging typically works the best for an X-variable piece of tissue, which you can get perfect resolution. But when we're talking about EV,

in vivo animals is actually quite challenging because how are you going to get a light in and out to, let's say, millimeters of tissue, let alone like the entire body of the human. Right. So what you're saying is that it's great if you just want to study the surface, but anything deeper, you would either have to cut them open, which is what we do, what surgeons do, biologists, or come up with a very amazing technology. Exactly.

- Exactly, so what we wanted to hope to achieve is that we don't have to cut open the tissue, we don't have to insert an ophthalmophibra, we don't have to insert a microendoscope, but we'll be able to make the tissue transparent by itself. And actually, apparently we're not the first to tackle this challenge.

there has been amazing work done in the past decade where people looked at various transparency technologies. These are called tissue clearing. So, you know, a lot of really amazing work done in this field. So, conventionally, for tissue clearing, we have to either remove

these are the high-index components from this water bag of the body. Or we have to replace water with high-index components, such as mixing the tissue with organic solvents, such that the water is replaced to match the index up to the lipid and proteins. Apparently, as you have seen, either approach will necessarily have to remove some vital components that are vital to maintaining life. Right, right. So just to repeat back,

The water if you remove the water and replace it with an organic solvent that tissue is not going to be functioning normally in the future And then so so you so you okay gotcha keep going. Yeah, so then what we want to understand is that can we find a way to not having to Replace either water or lipids or something else but maintaining well, I've got actually achieving the two transplants. Okay, so

So that requires us to look at why tissue, why water and lipids have different refractive indices. It turns out that water and lipids, despite being transparent and colorless in the visible spectrum, they are actually very strongly absorbing light in the deep UV. And this actually absorption... Ultraviolet. Ultraviolet, exactly. And this deep UV absorption, ultraviolet absorption, is the reason why they have different refractive indices in the visible. This is kind of like a mind-boggling...

because we typically think that different wavelengths or different colors are basically independent from each other. When we're talking about red light, we don't mix it up with green or blue. But in fact...

All wavelengths are actually causally connected via some kind of deep underlying physical and mathematical principle, which is known as the Cremas-Koenig relations, which basically tells us the absorption at one wavelength is going to determine the refractive index of a different wavelength and vice versa. So that basically tells us...

water and lipids have different refractive indices because they derive their different refractive indices from the different absorption in the outer body. And along a similar line, we could come up with an engineering principle by making water more absorptive in the UV or in a shorter region of the visible spectrum such that water will be as refractive

Earth's lipids at a longer wavelength, such as the web spectrum, are divisible. So this is now getting into physics, and just to keep it accessible, what you're saying is that what we would like is to match the performance, the refractive index, the color, so to speak, of water and lipids in the visual range. But in order to do that, we have to manipulate how they respond to UV light. Exactly.

That's right. Which is amazing, and we're just going to say it's a miracle of physics that this is a relationship that might not have been taken advantage of previously. Exactly. That's right. Okay. So now then, how do we do this? So yeah, this really opens up a lot of opportunities for doing this. So one of them is if we can identify really absorbing, strongly absorbing, such as dye molecules,

with absorption in the UV, absorption in the shorter range of the visible spectrum, such as in the blue or violet region, then it will have the effect of raising the refractory index of

of water at a longer range of the visible spectrum to match the refractive index of lipids without changing the chemical properties of water. And this is a dye that you would dissolve into the water. Is that right? So that it changes the overall properties of the solution that you're looking at. Exactly. That's right. Wow.

Okay, so does it work? Yes, it did work. Actually, you know, I have a team of very talented postdocs and graduate students. They actually did a screening of many, many dyes. And it turns out one of the dyes, which is actually used in the food industry, this is typhusine, or people also call this yellow number five. This is actually used in snacks such as the Doritos you hear. Oh, one of my favorites. So to the naked eye, it looks orange?

It looks orange slash red. Okay. Yeah. And then, wow. Okay, keep going. This is amazing. So Doritos is the answer to transparency. Yeah, exactly. So Doritos dye, the transparency looks red because it absorbs light in the blue region.

which is actually a very short wavelength, around 430 nanometers. And then this absorption leads to the refractive index change upon dissolution in water at a longer wavelength, which is red, which is actually beyond 600 nanometers. And then basically what we have demonstrated is that the aqueous solution of the yellow dye has the refractive index similar to that of lipids.

Okay, and that was the goal was to try to get them matched. Exactly. That's right. Okay, so how what were the first experiments? Like when you said screening, you're obviously not doing this in humans. What kind of tissue do you use? And what kind of transparency did you get?

Yeah, that's actually a good question. So we actually, first of all, we used a set of optical characterization methods, such as the UV visible absorption spectrometer, as well as the spectroscopic ellipsometer. And these are basically very fundamental optical characterization tools. And then once, you know, upon identifying some top containers, top candidates of these dye molecules, we then apply this solution of the dyes to a piece of chicken breast.

Chicken breast, there you go. Yeah. The chicken breast, you know, it actually looks pinkish and it doesn't really, it's opaque, it doesn't allow the light to pass through. But then we found that if we take the piece of chicken breast and soak that into the de-radium dioxide solution, just within nearly a few minutes, we can see the transparency significantly changes. If

It goes from complete opacity to translucency. And if you let it in for even longer, let's say half an hour to an hour, it becomes completely transparent. And then it achieves the transparency close to 80-90%. Okay, we have to stop because this is amazing. First of all, how thick is the piece of chicken breast? I'm sure it's not a whole chicken breast, or is it? Right.

Yeah, so what we have done is about 1 to 2 millimeters in thickness, which is not very thick. But actually, the NSF, National Science Foundation, they actually developed a protocol for any interested students, let's say kids, to produce at home. And then they actually did something that is as thick as 5 millimeters thick.

Exactly. Okay. And now you're getting to be like that's a thick piece of paper or a very thin piece of chicken. So you slice that and you simply soaked it and you counted on the fact that the water with the dye would diffuse by natural processes into the chicken and replace the – I'm calling it the normal water. Right. And as that happened, it became see-through. Right.

- Right, exactly. It becomes easy. - Unbelievable. Okay. So, okay, now I'm imagining you're pretty excited. How have you taken this further? Like what are the serious applications of this?

Yeah, so actually it really opens the door to many, many different applications. One of the things we're doing, we're very, very excited about is actually we're working with dermatologists at Stanford and we're actually working on the human skins. These are actually freshly dissected human skins with the goal of one day applying this to the human patients. And then the data is actually very, very promising. We're about to publish the paper in the upcoming paper. And then what we have found is that

There are many different dye molecules, well, apparently the DeVito dye is included. That has the effect of making the human skin transparent as well. And then the human skin, apparently, is actually much thicker than the mouse skin.

that we have tested in the paper so having a human skin transplant really has a lot of applications for example we'll be able to if this can be used for uh human patients then we might be able to identify some kind of a medical conditions such as let's say basal you know cell carcinoma yes skin cancer very common and maybe would look different when you could see through yeah exactly that's

That's right. So then we can actually detect these structures in the deep tissue. And this is one of the applications. And then we're also combining this technology with a variety of imaging modalities used commonly in lab animals. For example, we're doing two-photon, three-photon microscopy. We're actually applying this, combining with light sheet microscopy. And interestingly, actually, we're very delighted to see

quite a few labs very quickly picked up, you know, they adopted this technology and applied that in their own systems. For example, a group, Adam Wax Lab at Duke University, they actually applied this to OCT, the optical coherence tomography, and they realized that they could actually enhance the penetration depth by two to three times

in a live animal. - Okay, so what I'm hearing is that many of these other technologies that would have been only applicable on the surface until the light gets scattered, by effectively removing that layer, they can get better data on the deeper layers of tissue. - Exactly.

So before we go on, I was thinking about the human and the dermatology and we went from chicken breast to humans. Now the chicken breast is dead and we don't really care what we do to the chicken breast. We might make it a little bit saltier and tasty but separate from that because of the Doritos. What is the impact when you're doing this? Can you do this on a live tissue and does it still work and are you doing damage to the tissue?

I'm sure you're worried about this and doing a lot of tests in that area. Exactly. Yeah. So actually, we also, that's an excellent question. We did this on live animals, so live mice, basically. So what we have shown is that we could apply this to the shaved mouse abdomen and

And then typically the shape of the mouth abdomen actually has the skin, has the connective tissue, it has fat, it has muscles. So all of this actually will make the abdominal wall not transparent. We cannot see through the abdomen to visualize the inner organs. But what we have found that is quite striking to us when we first saw this is that by applying dimolecules, dissolving water onto the surface of the skin, abdominal skin, we could actually visualize liver, intestines,

You know, even bladder and a lot of different organs inside the mouth. While the mouth is still alive, we could actually see the mouth, heart, you know, beating, we can see the lungs breathing, we can even see the gut moving. Unbelievable. We're looking at the motility of the gut through the transparentized mouth abdominal wall.

When you do it to a living mouse, of course we know that the skin gets circulation. So it's going to be clearing the water. The dyed fluid, I'm guessing, is slowly going to dissipate as it gets brought to other parts of the body. So is that an issue and do you have to kind of refresh it in some way? That's right. That's actually a very good point. Actually, if you just leave the dye in there, it probably will maintain that effect for about maximum of 30 minutes. But after that, the effect gradually gets washed out.

because of the clearing of the body, as you mentioned. So that requires us to apply this again. So what is nice about this technology is it's actually transient and reversible. Getting back to your point about the safety of this, apparently we actually support the Doritos dye. It is not the best dye that potentially we can find. We're still actually looking for other more efficient, more potent dyes. But Doritos dye actually will still have to use a...

pretty high concentration, I have to say. And that concentration, if you leave it on the skin for too long, it may cause some kind of adverse effect. And then currently what we're doing is actually we leave this transparency window for about maximum 30 minutes. And then after that, we would actually wash out dye by using saline to just extract the dye molecule. And then we saw that, we actually observed that the skin actually very nicely returned to the original opaque state.

Now, I'm going to ask one ridiculous question. I always allow myself one ridiculous question. When I eat Doritos and it's all over my tongue, am I making my tongue slightly translucent? That's actually a very good point. Actually, you know, based on the physics, yes. It will make some of the epithelial cells slightly transparent. But the thing is that people probably have never used a microscope to look at or to characterize that kind of transparency. But I believe, according to physics, that is indeed happening.

You're listening to The Future of Everything, and we'll have more with Guo Songhang next. Welcome back to The Future of Everything. I'm Russ Altman, and I'm speaking with Guo Songhang from Stanford University. In the last segment, we learned how Guo Songhang and his team figured out how to make tissue transparent. We're going to discuss now whether this occurs in nature, and then we'll move to another project where he's figured out how to introduce light into the body without any wires or electricity.

In this section, I wanted to ask more generally, is this transparency phenomenon something that occurs in nature? Yeah, that's actually an excellent question. So it turns out that many species, usually, you know, these are aquatic species, not in mammals. So these

These species are actually derived there. They also are in here in street transparency. These species such as zebrafish larvae and also glass frogs, they derive these street transparency. So zebrafish and a certain type of frog. They're larvae, but I've seen full-grown zebrafish, and they retain a certain amount of transparency. Oh, that's interesting. Okay, so I need to update my knowledge. A little bit, yeah.

So, I'm actually, you know, I read some papers recently and it turns out that people only in the recent years have started realizing that some of these inherently transparent species derive their high transparency in a tissue from the high refractive index components. These are usually proteins that

in the cytosol of their body, and then which in turn comes from this intense UV absorption of these proteins. Now, once again, we're actually seeing this kind of a, you know, interconnection between the absorption of UV, which is a shorter wavelength, and also higher refractive index and longer wavelengths. So they're taking advantage of exactly the same physics that you did. The same physics.

I wanted to ask you, it just occurred to me, about the proteins in the eye. So like one of the miracles of human biology is how clear our lenses are and our eyes. Is this a similar phenomenon or is it different? I'm so glad you asked me this because we actually also recently looked into this. This is just amazing to see that we have various, we have probably, I don't know the number exactly, but thousands of different proteins in our body.

The crystalline, this is the protein that is found in cornea and the lens, which is important for the transparency in our eye and also the high refractive power for us to see things, is one of the proteins that actually is enriching the amino acids that give rise to the most intense UV absorption. Oh, wow. If you think about it, we have quite a different natural amino acids, but actually only a subset of them, very small fraction of them, have the most intense UV absorption.

And then crystalline is one of these proteins that has such a high concentration of its strongly UV absorbing molecules such that it derives very high refractive index. If you compare crystalline with other proteins in our body, crystalline is the highest, has the highest refractive index and that gives rise to transparency and refractive power.

And it preferentially uses these special amino acids. Amino acids are the building blocks of proteins. There's 20 of them, as you just said. And you said some of them are especially good at this UV manipulation. And is it enriched in those kinds of amino acids in crystalline, the protein? In crystalline, the protein.

Does that – okay, now I wasn't expecting to go in this direction, but that makes me wonder if we could engineer other proteins to be more transparent by making replacements with these kinds of amino acids in areas where it doesn't mess up the function. Exactly. So this is actually a really, really, really great suggestion. So we're actually already looking into this. So one of the things we found is that, first of all, the –

the concentration, the percentage of this UV-absorbing amino acids is one thing, and the other thing is actually interaction of the amino acids. It turns out that actually when the amino acids actually interact with each other, for example, in the polypeptide, the interaction itself could also enhance the absorption. And then now really, this really begs the question, what kind of amino acids should we put in? What kind of sequence we should design these amino acids to make the most UV-absorbing proteins such that if

If you overexpress these proteins in a cell, in a mammalian cell, we might be able to actually make a transgenic and adherence-transparent mouse and even make some of the mammalian tissue transparent. Yeah, my mind went exactly there that if you make these substitutions and they're compatible with life...

Now we have partially or fully transparent mice running around. You might not start out with mice, right? You might start with something a little bit more simple. Like there are these very small worms that we maybe could do first. But, okay, that's very, very exciting. Okay, thank you, Franz. I'm glad I asked that because that's fascinating. And it turns out that this phenomenon of the UV connection with the visible is the trick that nature has used exactly as you have. Exactly.

I wanted to move to one other thing you're working on because it just also sounds similarly magical, which is the idea of light that you introduce into the circulatory system, intravascular light sources. So just like before, why would we need these? And then tell me how we might be able to build them or how you are able to build them. Yeah, well, actually, all this work actually started from the...

the inspiration I got from reading a lot of the papers from the Diasporas lab. We got really inspired by all the work, pioneering working optogenetics from Diasporas lab. And then we realized that actually for a lot of the applications involving the delivery of light into the body, such as optogenetics,

Yes. We have to get our mind. So just for people who are not aware, optogenetics is the idea that we can control the function of genes in our bodies or gene products using light. And it was invented by, among others, our colleague Carl Deisseroth at Stanford. But this has created a revolution in the ability to turn, for example, neurons in the brain on and off using light. But you need to get the light to turn on and turn off at the right parts of the brain. Yeah.

Exactly. Yeah. So we actually, we found that actually in order to get the light deep in the tissue, especially in the mouse brain, we have to overcome the scattering, the opacity of the inherently opaque tissue. Yes. So conventionally people are inserting optical fiber. We actually were asking the question, can we get rid of the optical fiber which sometimes is invasive in the tissue? Yes. By using a

light source that is traveling inside the body and only gets turned on with a form of energy that can non-invasively penetrate through the tissue. And apparently in order to achieve that, we cannot rely on light itself because we have just seen a home scattering like this. Yes, and we just spent 20 minutes about how hard it is to get light into deep tissue. Yes. And then one actually alternative form of energy is actually ultrasound. What is that? Ultrasound.

Ultrasound, okay. We know that ultrasound is used widely for deep tissue imaging in the hospital. It goes through centimeters of the tissue versus less than a millimeter for light. But unfortunately, ultrasound itself is not light. These are two very different energy modalities. And the question is, how can we convert ultrasound into light? We actually created a technology that allows us to quote-unquote see the sound, which is basically using a material that a

upon ultrasound stimulation can release light emission. And then one of these types of materials is called mechanoluminescent materials. Basically, you create mechanical stress in a material using ultrasound and then they become luminescent, which is light emission.

Yes. An interesting thing about this material is that they can be made so small that we can inject them into the circulatory system. And during the circulation, they only get activated by the ultrasound wherever and whenever you focus the ultrasound in a particular deep tissue location. And only at that location can you get the light emission.

So this is actually it's a very simple idea. So the this special material that you've created circulates everywhere in the body But it's dark so to speak but it can be activated when you do a focused bit of ultrasound That then makes it light up and now does it light up to the visual spectra or does it light up to or do you have control of the way? Over the way, let's we actually have demonstrated that you can give various different colors from blue all the way to red and

So this is exciting because as I explained with the optogenetics, you might want to turn certain neurons on or off. And then I'm sure people like Carl, our friend, and others would say, what is the resolution? How precise can we focus ultrasounds so that we just turn on the areas with like –

control of exactly where it is. I just don't know. I see these fuzzy pictures of my grandchildren in ultrasound. So how good is ultrasound at focusing? Excellent question. It actually depends on the frequency of ultrasound being used. Just like light, the resolution of our ultrasound is diffraction limited. So which means if you're going to use, let's say, one megahertz ultrasound, which is commonly used in our lab, it

It gets us a resolution of about 1 mm. But if we are able to go up to 5 MHz, that goes down to 200 µA.

And 200 microns resolution is actually quite similar to that achievable using any type of fiber. Yes, yes. That's 10 or 20 cells. Yeah, that's right. So that's really very precise, and they should be pretty happy with that. So what is the status of that technology? Is it still experimental? Have you injected it into mice or other living beings? And what kind of results have we seen?

We actually have published a few papers on this and the latest work that we feel very excited about is actually using this ultrasound to produce light emission at multiple locations in the same outbreak. So, imagine you think about implanting optical fiber. It's actually so invasive that you can only implant it in one, maybe maximum two or three locations. But now, this is actually with ultrasound, this becomes a quote-unquote maybe virtual light source without a physical implantation.

And then basically you can dynamically target different brain regions, producing light emission in an orchestrated way. Almost like playing the piano with 10 fingers instead of like one finger at a time. Right. Thanks to Guo-Sang Han. That was the future of transparent tissue. Thank you for tuning into this episode. You know, we have more than 250 episodes in the back catalog and you'll find an amazing array of discussions on the future of anything.

Please remember to hit follow in whatever app you're listening to. That'll guarantee that you always get alerted to our new episodes, usually coming out on Friday, and you never miss the future at all. You can connect with me on many social media platforms, including LinkedIn, Threads, Blue Sky, Mastodon, at R.B. Altman, or at Russ B. Altman on Threads. You can also follow Stanford Engineering at StanfordENG or at Stanford School of Engineering.