- Welcome everyone to Wednesday Nite @ the Lab.

I'm Tom Zinnen.

I work here at the University of Wisconsin-Madison Biotechnology Center.

I also work for the Division of Extension, Wisconsin 4-H. And on behalf of those folks and our other co-organizers, PBS Wisconsin, the Wisconsin Alumni Association, and the UW-Madison Science Alliance, thanks again for coming to Wednesday Nite @ the Lab.

Tonight, it's my pleasure to introduce to you Scott Coyle.

He's a relatively new professor from right across the street in the Department of Biochemistry.

And he's gonna have what I think is an extraordinary talk because he has extraordinary video of cells.

Cells that move and do things and molecules that move and do things.

And it's very different than the microscopic visions I had in college, or at least the ones through the microscope.

Scott was born in Livermore, California and then went to Granada High School there in Livermore.

He stayed right there in the East Bay to go to the University of California, Berkeley, where he majored in biochem as an undergrad.

Then he went across the Bay to get his PhD at the University of California at San Francisco, also in biochemistry.

And then he went down the road just a little bit to Stanford to do a postdoc.

He came to the UW-Madison in late 2019.

He single-handedly timed the COVID pandemic.

- Scott: What can I say?

- Tom: Perfectly.

I think this is an extraordinary thing.

Veit Bergendahl gave the first talk for Wednesday Nite @ the Lab on February 22 of 2006 on human embryonic stem cells.

For me, it's great to have this extraordinary talk also on cells to wrap up my time as the emcee here.

Would you please join me in welcoming Scott Coyle to Wednesday Nite @ the Lab?

[audience applauding] - Scott: Thank you.

Make sure my pointer's working.

Can you all hear me?

PBS, you can hear me?

Good, all right.

Well, it's my great pleasure to be here and talk to you about kind of my thoughts about biology and cells.

And really, I am a biochemist and a synthetic biologist interested in understanding and engineering cells.

And that's kind of a daunting task because everything I'm showing you on this screen here is a cell, right?

And they look very different.

They're doing very many different things.

And what I'm hoping to tell you about in today's talk is ways to think about the answer to a simple question, which is, "What are cells?"

They're very complicated, obviously, but I'm gonna share with you some ways I like to think about cells that make them easier to understand.

And through that lens, we will see also that that approach to thinking about the cells may give us an answer to another question, which is, could cells be more than they already are?

Can we engineer them and change them to do new things for us in the world?

And so with that in mind, I'm gonna try to tell you about three different sort of arcs in this talk as we think about these fascinating little creatures that we call cells.

I'm gonna give you kind of a view of the kind of classic biology of cells.

How do cells work and why are these kind of the basic unit or the kind of the atom of all life on earth?

And I'm going to try to make it very simple because every cell is very different, and the biology of cells is very complicated.

But there's some useful abstractions in there that will help us kind of generalize our thinking about what a cell is and what the key things that have gone into making this an important unit for biology.

And from that abstraction, I'm gonna challenge us as an audience to think a little bit about cells from different perspectives.

So instead of thinking about the cell as a biological entity in and of itself, we're gonna start to reimagine cells as different kinds of things.

A manufacturing system, for example, or a chemical computer, or even some sort of self-driving machine or microscopic robot.

And then in the final stage, I'm gonna tell you a little bit about how those new perspectives of kind of projecting cell biology onto these other concepts can help us think about ways to engineer biology and living systems through this lens of other kinds of technologies, really with an aim in the long term of kind of paving a way forward to sort of a new, more biologically-powered future, okay?

That's a little bit of a synthetic biology futurism there, but I hope you'll join me on this fun excursion, okay?

So let's start by just thinking a little bit about cells, how do they work, and why have they become kind of the unit of all life on earth?

And the first thing we need to recognize is that cells are really the minimal system that's capable of executing all the processes of a life form, okay?

So what I mean is if we have a cell here, and real cells look a lot more interesting than this little schematic there [laughs], but these cells will have a few things that are really important to how they're able to function, okay?

The cell is gonna have a genome.

And what that means is that there is chemical material inside the cell that provides the information for this thing to be able to make an exact replica of itself.

And so I'm sure many of you have seen the beautiful structure of the B-form double-helix DNA structure.

And it's a really remarkable chemical entity for encoding information because it's actually this beautiful helical structure here.

And the complementarity between the two strands, the pairing of As with Ts and the pairing of Cs with Gs through Watson-Crick-Franklin based pairing interactions that make this molecule this perfect copyable entity that can be replicated and passed down from one to the next.

But this material, the genome, is only part of the story, right?

And this genome encodes the things that enables a cell to build materials, okay?

And cells are amazing builders.

They can create nanoscale machines called proteins that perform specific biological functions.

And so for many of you, proteins might be like the soy powder that you would get at GNC, like my students over there, bulking up and things like that.

But really, proteins are kind of, each one of them is a very special entity that can perform chemical reactions.

They can create structure.

They can do all sorts of amazing things.

And our genome contains the instructions for the cell to build all kinds of different proteins with different functions that let them execute the tasks that they need to perform.

And finally, a key thing about cells is that they can react to what's happening around them.

So they're not like a static thing that just sits there doing nothing.

They can change what they make and do in response to signals that they experience from the environment.

And that's because they have all sorts of things on their surface.

They're like this little red thing, we zoom in.

There's all these things on the surface.

We call them receptors.

They're special proteins in the membrane of these cells that combine to things in the environment.

And when they do, they transmit information to the inside of the cell, where networks of additional protein molecules work together to interpret what those signals are, process the data, and then communicate to our genomes what types of new proteins they should make or what kinds of changes to the biology should happen.

And so really, we can sort of abstract the cell to kind of just a few key components that are really critical to driving this function.

The genome, which is providing these chemical blueprints with which we can make the molecular parts, and then kind of the proteins that are the actual molecular parts that the cells are gonna use to do most of their work.

And if anyone out there works on lipids, I apologize.

I left out some critical molecules that exist.

But this is kind of like the key set for me in terms of what's going on.

And then it is through this interaction between our genome producing proteins, and then proteins doing important work and doing important tasks that feed back into our genome.

This sort of leads to this integrated loop in which we end up with a cell.

And so we can then start to think of the cell as this network of dynamically interacting molecules that are working together to produce this functional, self-replicating system that we call a cell.

And the things that cells, the kinds of machinery that comes out, the kinds of cells that we find in nature, and the diversity of forms that we can see is just truly astonishing.

So I'm showing you here two videos of two cells, both of which, I think, you could spend your whole life trying to understand how they work, and think about their remarkable capabilities.

Over here on the left, we have a video of a T cell interrogating a target cell in its environment.

So T cells are immune cells, they're in your body, and one of their functions is to investigate or talk basically to other cells in your body and say, "What's going on in you?

"Do you have a viral infection?

"Do you have some sort of, is there something inside you that shouldn't be there?"

And what you're seeing this cell doing is it's actually searching around on the surface of this target cell, and it's looking for information about what's going on in the inside to make an evaluation as to whether it should let this blue cell live on or if it should destroy it because it thinks it has a virus or something else like that inside it.

And so here's a cell investigating another cell.

On the right-hand side, we have an extraordinary single-celled animal.

So this is basically a single-celled organism that's almost indistinguishable from the kind of complex animal structures that we form.

And we found this cell, my students and I, we just went out to the lake, fished it out.

You find these things.

If anyone wants to know where you can find them, I can tell you where we get them.

But this cell is almost able to do the same thing, but with a completely different kind of structural apparatus.

You see, it's built this unusual hunting trunk, and it swings it around, and it looks for food in its environment.

And when it recognizes something that it's evolved to eat, it's able to detect it, paralyze it, and eat it.

And so both of these cells are basically doing totally different things, but they're sort of built into the same framework of they all have genomes, they're all making proteins, and yet we're getting so many different kinds of results at the other end, okay?

And this is one of the elements of a cell that's actually quite phenomenal for building even more complex organisms.

It's really this flexibility that it makes cells an exquisite building block for multicellular life forms.

Because a single genome can contain the information needed to make all the parts that you might need to make any cell in your body.

So from this one little cell, you can potentially, by regulating what portions of the genome are actively used at different places and times, you can start to create all kinds of different cells.

Cells like neurons, cells like epithelial cells, intestinal cells, muscle cells, blood cells.

They're all producible just from one set of parts inside this genome, but using them in different ways.

Not so different from how my son, who's now getting into playing with blocks and Legos.

You can make all kinds of things from the different parts that are in his collection.

It's how you put them together.

You can get different results.

And just to give you a sense for the kind of scale of how many cells might go into a multicellular organism, a very simple animal like a nematode, which my colleague, Judith Kimble, studies, they might have a thousand cells or so.

And a mouse might be made up of three billion cells, but our human cells have upwards of 40 trillion of those little things running around, working together to build us up.

And within those trillions of cells, there's hundreds, and people are discovering new ones all the time, but hundreds of different cell types that is specific machines that can be built from that single genome, muscle, neuron, et cetera.

So it's quite remarkable how much you can potentially get out of this kind of framework for building materials.

So that's kind of just like how I like to think about the cell, okay?

Try to abstract it a little bit and simplify it.

And now, I want you to think about those things that I just told you about, genomes that contain information, making things, making decisions, evaluating them.

Let's take a moment now to see if we can start to reimagine those activities of cells as other things.

That is, we can start to view the cell as something that can serve as a manufacturing device to make lots of something, or it can serve as a chemical device that can perform computations for us, not so different from the computer I'm presenting to you around here from, or even operate functionally, not so different from the kinds of robots that you might see running around campus delivering food.

I don't know if any of you have seen those, but they're out there.

So what I wanna kinda do is project onto this scheme I had for this sort of oversimplification perhaps of the cell, each of these concepts for a sort of technological paradigm that we can kind of imagine.

And so we can think about the genome and its information that it contains and the proteins that it can make, which are the sort of parts, as a step in a sort of manufacturing process, right?

Because the genome allows us to manufacture or create protein products.

And in fact, you'll find that there are many cells in your own body whose primary functions often involve the manufacturing of lots and lots of specific proteins.

So for example, myocytes that make up your muscles, a large portion of their mass is one protein called myoglobin, whose primary function is to bind oxygen so that your muscles have enough fuel to be able to power themselves, okay?

And so these cells will sit there, and they're cranking out tons and tons and making tons and tons of these myoglobin molecules so that they can make sure that they are fueled up and have enough oxygen ready to go.

And then just a little way down the, I guess I don't know where it is on my body relative, but elsewhere in your body, you'll find cells like B cells.

And these are the cells whose primary function in their existence is to manufacture antibodies.

So each B cell, for example, in your body can produce one unique antibody that will bind potentially something, right?

Maybe it will never bind anything at all.

But the purpose of this B cell is that perhaps one day, a foreign invader will come into your body, and the antibody that this B cell makes will bind specifically to that invader.

And if it does, then you have basically tagged that foreign material for destruction.

And when a B cell produces material that detects a foreign agent, it will kick into overdrive on a mass production scale of this antibody and will undergo massive changes in its metabolism and its architecture so that it can secrete as much of this protein into the environment as it possibly can to try to basically tag all of those invaders, whether they're bacteria or viruses or whatever with this material so that they'll be destroyed.

But I wanna emphasize that cells aren't really just limited to manufacturing, say, a single protein molecule.

They can actually make things much, much, much larger and much grander in scale.

And so we can also start to think about the cell as a machine that not only makes these nanoscale proteins, but also makes microscopic devices and things.

And I'll give you one example that one of my students has recently discovered.

She grows these very unusual cells in the lab called P. collini.

Their bark is much, well, I don't know.

They're cold-blooded killing machines, as you'll see in a minute.

And the reason for that is that these cells are able to manufacture a specialized structure that we call the tentacle.

But in fact, the purpose of this structure is to provide the physical support and the chemical sensory necessary to essentially puncture foods in the environment and suck up their goodies.

And if you don't believe me, this is what it looks like.

When these cells, for example, encounter something they wanna eat, they puncture their targets and they use these arms, these straws, essentially, they've built, dozens of microns in length, to suck out all the goodies.

And then when they're done, you'll notice they eat for a while, like you see here.

And then when they've had their fill, at some point, they just throw a little leftover wrapper away, okay?

I think it's really funny because they're very, actually, they litter everywhere.

So like when my student, Maggie, grows these, we watch these.

You come back the next day and there's just wrappers of these carcasses everywhere, okay?

But what's remarkable, I mean, it's fun to look at these obviously, but these cells are essentially manufacturing microscopic straws that can detect targets, and upon contact, puncture and begin feeding them, okay?

So these structures are much longer than any individual protein.

And other cells that you might encounter in the environment can build extraordinary physical structures, physical objects like this.

So that earlier cell that I showed you swinging its sort of proboscis around to look for food.

Underneath that hood, there is an interwoven web of protein structures that sort of form this deformable mesh that allows these cells to stretch and bend this structure all through in their environment.

And these are structures that you can actually even see with your naked eye.

I mean, they're over a millimeter long.

They are very large and they're just made out of proteins.

So these are the kinds of microscopic manufacturing capabilities that essentially defy any sort of skills that we have in the laboratory to make ourselves.

So that's making stuff, and making stuff is great.

I was told not to walk around.

Okay, making stuff is great.

But another thing that cells can do that's quite remarkable is that they can actually compute for us.

And what I mean is just in the same way that you're listening to me and you're like, you might be thinking, "Ugh, when is this talk gonna be over?"

or whatever, your brain is making computations based on the inputs that you're receiving from me and the things that you're hearing, and it's making evaluations and choices and things like that.

And individual cells are able to do this as well, but they're able to do it without having a brain or anything like that, right?

So it's just a chemical system that's able to perform these kinds of computations.

So how do these work?

How can you compute using chemical substances instead of wires and transistors?

Well, I'll briefly introduce you to a very classic example from history for how this kind of computation is performed because I think it actually illustrates so nicely how many biochemical decision-making systems have almost direct parallels to the kinds of decision-making systems that you might use in an electronic system and the type of computational structures you would use there.

And so perhaps some of you have heard about this bacterium called E. coli or Escherichia coli.

These bugs are actually quite remarkable in terms of their computational capabilities.

And one of the decisions they make that I think people in Wisconsin will find a little distressing is these bugs love glucose as their food.

These bugs would like to eat glucose all day if they could.

It's their favorite food.

You wanna know what they don't like?

Lactose, the thing we use to make dairy products.

They would prefer never to eat this unless they were, like, absolutely starving, okay?

I don't understand why.

Somebody should give these bugs a curd, but... [audience laughing] But this is how their worldview is, okay?

And so how is this bacterium able to have this sort of picky preferences, okay?

'Cause it doesn't have a brain; it's just a little sack full of molecules, right?

And so we've actually known for almost 60 years now how these bacteria are able to make this decision.

And it's quite remarkable.

To me, it's even more remarkable that these two gentlemen were able to work this out almost 60 years ago, in a time period where it was perfectly acceptable to smoke in the lab and the lab looked like a kitchen and things like this.

But they're very committed to answering this question about why they don't like milk.

And what they discovered is actually perfectly analogous to a Boolean logic gate.

And I'll kind of take you through how this works.

What they found is that the bacteria in its genome has this region where it's gonna have all the genes it would need to eat lactose, okay?

If it doesn't make these genes, it can't eat lactose.

It's not gonna waste any time making lactose.

And right in front of those genes, there's two sets of DNA sequences which are called regulatory sequences.

These are sequences that don't specify how to make any protein, but they're gonna influence whether or not these proteins are gonna get made, okay?

And there's two kinds of regulatory sequences in here.

One of them binds a protein that's called an activator, okay?

And this activator is required to be bound to the DNA in order for these genes ever to be transcribed.

And so if this protein is not bound, the genes will not be made into protein.

And in order for this protein to bind DNA, it has to have bound to it various by-products of starvation.

I'm not gonna tell you what they are 'cause they're not really important.

But the bottom line is, is that if the cell is not starving, this activator protein is not gonna bind to this piece of DNA.

And as a result, you're not gonna make any of these genes, okay?

So that's the first kind of input to this control scheme that the cell has, okay?

The other one is what we call a repressor protein.

And this is a protein that if this protein is bound to this piece of DNA, it will prevent these genes from being made into protein.

And it doesn't matter whether the activator is bound or not, in this case.

The binding of this repressor protein there completely stops the ability of these genes to be made.

And this repressor protein has evolved to sense the molecule lactose.

So if lactose is present in the cell, then this protein falls off and that frees up the possibility that these genes can be expressed.

And so we can think about this kind of for this cell's reasoning as saying, "There better be some lactose around if I'm gonna make these genes," okay?

And so what we have here are two regulatory sequences in front of this lactose metabolism gene.

And we're gonna think of 'em as taking two inputs.

One says, "Am I starving, yes, no?"

The other says, "Is there lactose around, yes, no?"

And the output to those two inputs is a decision, "Do I make these genes?

"Do I make the proteins that I need to eat this lactose," okay?

And for those of you who might be electrical engineers or anything like that in the audience, you might recognize that the conditions for making lactose, which are that both you must be starving and there must be lactose in the media, are exactly the same as the fundamental electronic logic AND gate that essentially a foundational component of all of our sort of computing devices, okay?

And essentially, you can then start to abstract this gene regulation that this bacteria is doing as a simple equivalent to an electronic circuit, a very simple AND gate where you need both lactose to be present and glucose to be absent.

And only if both are true, then do you make these genes.

And so this is pretty cool because this is basically Boolean logic like you might find in a transistor made out of silicone, and then we had to manufacture with various kinds of tech.

But it's built entirely out of those protein components that I told you about.

And instead of it being electricity that's being used to make this computation, it's biochemistry.

It's the binding of proteins to and from pieces of DNA and the ways in which they interact.

So that's pretty cool.

But an AND gate will only get you so far in life, right?

And this is where what becomes more interesting is that you can establish much, much more complicated control logic by stacking, just like you might in a computer, multiple kinds of logic gates together, connected in different ways.

And so if we go back to that little video I showed you before of that T cell that's interrogating the target cell, and it's trying to figure out, "What's going on inside?

"Should I kill this cell?

What should I do to it?"

You might notice that there's this very dynamic interface between the two, where it's sort of groping and grappling and shaking it around.

At that interface, there are hundreds of interactions between different molecules on the surfaces of these two cells, which are being used to inform the internal computational process about what sorts of decisions it should make.

And it's actually quite remarkable the number of choices that can be made.

You don't need to know any of those names, trust me.

They're all different possible inputs that provide different kinds of signals to the cell about what's going on.

"Should I kill this, should I not?

"Is this a cell that has a viral infection, "or is there bacteria in there?

"What types of things should I secrete depending on what I sense," okay?

And so by stacking hundreds of these on top of one another, you can build a system that can create much, much more complex decisions than a simple AND gate.

And in fact, many folks are beginning to start to model these kinds of networks as being almost like a biochemical version of the neural nets that computer scientists are using to process massive quantities of inputs and make different outputs.

And I will say, as biochemists and cell biologists, we still don't understand actually all the ways in which the computations are being performed by these kinds of networks as we go.

But it does suggest that it's being built up from these more simple things.

So everything I kind of said so far was about things where there's an analogy directly to a computing strategy we have in electronics, like logic gates and various kinds of neural nets.

But something that's very interesting that cells can do as well is they can also use very novel computational strategies that are almost unique to chemistry itself to do other kinds of computations, like, for example, solve a geometry problem.

And one of my students in my lab works on a system of proteins that can be found in bacteria whose function is to help this bacteria know where the middle of the cell is, okay?

They essentially need to be able to figure out where their midpoint is so that when they divide, they divide evenly into two cells.

They don't divide into a tiny little mini cell and a big cell over there.

And so the cell has to solve this geometry problem.

"Where is my midpoint," right?

And I think if we were humans, we'd pull out the ruler and we'd measure the two lengths, and then we'd say, divide by two or whatever.

So how do these bugs do it?

Well, they actually use a strategy that takes advantage of the chemical properties and the things that are sort of unique to a biochemical system, which is that these bugs will essentially create a wave of proteins that goes back and forth and back and forth and back and forth inside the cell.

And what they're doing is not trying to make you get excited about waves, although we, in my lab, are very excited about waves.

What they're actually doing is setting up a wave structure that has a node in it.

And that node based on the geometry of the cell will always occur at the midpoint of the cell.

So they set up a wave, and the sort of spatial constraints on how this wave propagates necessitates that a node will materialize in the middle of the cell.

And that is how they actually choose where their midpoint is.

So this is an example of a cell choosing a kind of method for computation that's not really a way we do, but is extremely efficient for them.

So briefly, we've now seen that cells can make stuff using their genomes, and those making proteins and making structures.

And we've seen how they can use proteins to do different kinds of computations.

So let's think a little bit about how you can put the two together to create a dynamic machine that can do more than just make and compute, but can kind of interact dynamically with this world like a tiny little robot.

And I'll give you two examples real briefly here.

One is a very classic one where I'm gonna show you this video.

It's from the '50s 'cause I like to pay my respects to the scientists of before.

The very famous video, I've seen it dozens of times in my life.

This is a neutrophil here chasing after a bacteria, okay?

So neutrophils are immune cells in your body.

Their purpose is to find things like bacteria and eat them.

And this movie is, again, almost 70 years old at this point, but I think it nicely shows very well how a cell is able to, you know, it's not just a bag of genomes and things like that.

This is something where it's sensing the location of where the bacteria is, and then it's modifying its structure so that it can grow towards it.

And it's going to keep doing this and modifying its structure and sensing and modifying until it catches up, okay?

And what this means is that this behavior that we're seeing here is actually sort of powered by this self-driving control loop, okay?

We have a process in which the location of that foreign bacterium is somehow detected, and the response is to try to change and build its structure so that it reorients towards where that bacteria is.

And it's gonna keep doing this, sensing, responding, sensing, responding sort of in this loop until it finally makes contact, right, with that bacteria.

And when it does, that's gonna trigger this engulfment response.

And if all goes well, you eliminate these bacteria, okay?

Now, I wanna emphasize, because I'm about to show you something else on the next slide.

This video is sped up about a 120 times.

So if you were looking under the scope, it would not look this cool, I'm afraid.

But if you speed it up, it looks really cool.

However, this same principle of self-driving control loops and sensing and responding can happen at exactly the same kind of timescales that our own lives are going on.

And they happen with that organism I showed you before where we have a simple, single-cell pond organism, and this is real-time.

So this is what it looks like if you look under the scope.

You'll see this little single cell is whipping this structure around.

And what is it doing?

It's looking for something to eat.

And right here, you'll see in a moment that when it finds it, it actually paralyzes those objects, and then you will see, it opens its mouth and it starts to eat them, okay?

But the point here is that, once again, the behavior of the cell is actually governed by the self-driving control loop.

And essentially, this cell has an algorithm that's implemented in it, which enables it to perform a random search of its local environment.

And at the tip of this structure where it finds these preys, it has all kinds of proteins loaded up in there that can sense what it's evolved to eat, okay?

So a repertoire of things that it likes to eat.

And when it contacts them, it will release toxins.

That takes about a couple milliseconds.

And then once the prey has been sort of paralyzed, then it can take its time to enjoy its meal, okay?

And so this is just kind of another example of how combining those together can yield these sort of robo-like systems, okay?

So now, we've sort of projected the idea of cells onto various kinds of paradigms that we don't often think of as being biological, right?

Maybe some of you do if you're biologists, so whatever.

But many of us don't think about the computer and think, "Oh, the cell is computing and things like that."

And so what I wanna do now is show that that sort of abstraction and thinking about cells from the vantage point of different kinds of technologies can give us clues for how we might actually engineer cells to do new things for us.

And not only to build cells that do different things, but actually to potentially embed cells as a part of a technology in our everyday lives, where we're essentially harnessing the abilities that cells have to do new things that we could not before, okay?

And to start with this, we can just see that we just imagined cells as these kind of familiar analogs or analogs of these very familiar things.

So we thought about them as molecular factories that could manufacture huge quantities of substances.

We thought about them as chemical computing devices.

Here, I, again, pay my respects to the years of old in terms of what computer is.

And here are those very small robots that I was telling you about that deliver food to kids all around campus.

Quite impressive, right?

So we've seen that we can imagine cells as being similar to these kinds of things, but is that abstraction useful?

Can we actually repurpose living systems to build out these kinds of technologies in new ways?

And the analogy that I'd like you to think about is that if we could do this, it wouldn't really be the first time that we had done something like this as a human species.

Because many of the electronics that's powering our computers and everything like that, they never existed in the form that we have them today.

And they came from us looking at things like lightning strikes and other kinds of natural phenomena that inspired us and dazzled us.

And then we found ways to domesticate those kinds of forces of nature and make them bend to our will so that we could have new kinds of technologies available to us.

And so in the same way, we've domesticated these other forces of nature, can we potentially do the same to cells?

And why might you wanna do this?

I mean, don't we have factories?

My colleague, Vatsan, who's also a synthetic biologist is like, "Oh, people are always trying too hard in terms of selling these things."

But I would say that cells are completely unmatched in certain abilities they have.

And in particular, it is their ability to organize and manipulate energy and matter at the microscopic or nanoscopic length scales that really, there is no parallel on our planet to their capabilities.

And certainly, we have ways of manufacturing very tiny objects and making very tiny things.

I mean, we just talked about computer chips and things like that.

But to make those objects requires enormous investments in energy.

We have to use huge amounts of fuels or heat things to extraordinarily high temperatures, okay?

A cell does all of these things at room temperature, right?

My students, they grow 'em in the lab, it's like 30 degrees.

That's, like, what it takes.

They can also do all those things using simple chemical fuel sources as their supply, right?

So they can do it using glucose like we just talked about with the E. coli, or things like plants can harvest light and harvest energy from light to build their machines, okay?

And what I think is perhaps the most useful thing is that if you've ever fabricated, like, a PCB chip or whatever, you go to all these great lengths to do it, but you can only make one thing from all that effort, right?

The thing that you make, the wafer that you have, the chip that you use for that lithography, you can print many chips with it.

But if you wanna make a new chip, you're gonna have to make a new mold.

And so this kind of manufacturing, it's kind of scaled towards building one or two things from all this effort.

A cell, as I told you, is essentially infinitely reconfigurable.

From one genome, we make all the cells in our body, right?

And that means that you can start to imagine that cells are almost like some sort of living 3D printer where thousands and thousands of different designs can be installed into this living system.

The system can be shared with other people because you can grow it, and then hand it off to your friends.

I mean, my students have sent DNAs and cell lines and things like that they've made to other people.

It's an infinitely shareable medium.

And it can be loaded, preloaded with all kinds of extraordinary designs.

And all you need is, like, some sugar to get it going.

And I think that's pretty incredible to think about what you might be able to achieve if you could really find the best way to exploit that.

So let's kind of then think about some ways in which cells can be the different technologies that we thought about them in useful ways.

And I'll tell you some that are already used in our everyday lives.

And then we'll talk a little bit about some more futuristic ideas as well.

So the first thing we gotta think about is getting cells to make stuff for us, okay?

And this, again, gets back to this idea that the genome contains the instructions to make parts and the proteins are the things that they're gonna make.

So in this arm, we have this sort of manufacturing process going on here.

And the key thing that's really nice about cells is that because the information to make the proteins comes from here, in order to make cells make different things for us, we can simply modify their genomes because by modifying the genome of a cell, we can control what proteins it makes.

And those proteins might be its own proteins if we wanna maybe make the cell behave in a different way.

But we can also make them make proteins that are completely new.

Maybe they come from different organisms, maybe they're ones we designed on the computer, but the rules just say that the cell will make whatever we tell it to.

And so this is a tremendous capability to manipulate these organisms to make stuff for us.

And let's think about some things that they actually are already making for us to this day.

So I don't know if anyone has diabetes; my grandma did.

May she rest in peace.

But what used to be the case when she was a young woman was that when she would need her insulin, that insulin would come from a pig.

That is, they would take the pig and they would get the insulin out of the pig, so it's pig insulin, and she would take that insulin for her diabetes, okay?

And that's because there wasn't a good source of human insulin unless you're gonna try to, I don't think the ethical rules are appropriate to drain the blood of a human and take their insulin and use that instead.

So this was the best choice that they had, right?

And of course, this doesn't really work that well because even though it's kind of a substitute for our insulin, it's not the same protein as our own.

And as a result, you can have an immune response to it or it doesn't work as well.

It's really suboptimal.

So what happens in today's world is that most of the insulin that someone who's diabetic will take comes actually from what's called a recombinant source.

And what that means is that an organism, in fact, oftentimes the E. coli cell that I told you about with the lactose disdain for our good lactose, the genomes of those organisms are modified to basically include this human insulin gene.

And not just like any insulin gene, our insulin gene, the exact one that we would have for our proteins, okay?

So we can put those genes inside the cell, and then cause that cell then to manufacture huge quantities of human insulin for us, okay?

And the key thing is that by programming the bacteria to make human insulin, even though it was made in a bacteria, for all intents and purposes, if you give it to us, it is exactly the same as human insulin.

They made it exactly the same way.

It's the exact same sequence of amino acids.

It's the exact same protein, okay?

But we can make it at scale.

So imagine if you go to any of those breweries and you see those huge vats of yeast making delicious beer for us or whatever, there's another place somewhere where people are making huge vats of insulin from these bacteria, okay?

So instead of having to sap those poor pigs to get crummy insulin, we can tell cells to manufacture our own insulin for us.

That's pretty cool.

Now, of course, that's taking someone else's cells and getting them to make useful things for us.

But in the last few years, we've actually seen a different kind of paradigm in which our own cells become the factories to make a foreign material.

And so many of you may have taken an mRNA vaccine for COVID-19.

And what that vaccine was was it was a nanoparticle containing an RNA molecule, which contained the instructions for making large quantities of a COVID viral protein.

And this nanoparticle was taken up by your own cells, and your own cells then served as the factories to manufacture large quantities of this foreign material, in this case, the spike protein here that would serve as the antigens for your own immune system to have a response, okay?

So we don't have to make it that the manufacturing happens in the lab outside there.

The manufacturing can be performed directly inside you.

Okay, now that's a little scary sometimes too because maybe you gotta think about what you're manufacturing in your own body.

But it does create a new paradigm where instead of tedious manufacturing, purification of these protein products in the lab, separation, all those things, you can send the DNA materials into your body and get your own cells to do the work for you; pretty amazing.

Now, we don't have to stop, of course, at the single protein level.

There's active research going on all the time to use cells to build larger structures for us with unusual things.

So there's an interesting example where there's a lot of nanomaterials work done in which people actually use bacteria to make magnetic particles inside them, like these structures here, which are called magnetosomes.

And the regular magnetosomes aren't necessarily that interesting on their own, but these are perfect, defectless crystals because they're made by a cell using proteins to help structure and template this thing.

And scientists are finding ways to dope them with interesting rare metals and things like that to create essentially remarkable sort of electronic magnetic systems that can be then isolated from the cells and used for different other kinds of things.

And in just one other example, you could even get cells to make pure biochemical machines for you similar to the structure, similar to those sort of single-celled dynamic structures that I showed you before.

Here, you can actually get cells to make chemical machines for you.

These are little flagella that have been isolated from living cells.

You can actually take them out of the cells, strip away all the membrane, all the cell, all the stuff.

You feed these a chemical fuel source, and they'll start swimming around and moving through in their environment and binding things and interacting.

And scientists are working to modify these with different kinds of detectors and things like that to really make these tiny little robots and things like that.

So you don't have to stop really at the single protein level.

You're really just limited by your imagination, I feel, in terms of how complex the structures that you create can be.

Now, in this last little part, I'm gonna tell you a little bit about how we can get cells to perform new computations and build towards getting cells to behave in new ways sort of as robots.

And so this time, we're gonna be thinking about this other arm in our kind of model of the cell where we have proteins doing work to then communicate some sort of computational result back to the cell to perform.

And since we've been sort of thematically building around this T cell interacting with the target, I wanted to tell you about some work that I've previously been involved with that I think really gets at the power of trying to project these computations onto other cells.

So what I showed you in the original video was a native T cell interrogating a regular cell in the body, but there's a great interest in engineering T cells to be able to recognize cancer cells that they encounter in the environment.

And why might you wanna do this?

Well, the regular way in which a T cell will recognize a foreign target is based on an interaction between its T cell receptor and antigens that are presented as MHC-peptide complexes.

Doesn't really matter for this audience what that is, but these are the interactions that tell the cell whether foreign material is present in this cell.

And of course, the challenge sometimes with cancer is that we have a cell in our body that we would like to eliminate, but oftentimes, the majority of that cell isn't foreign, right?

It's our own bodies that have become messed up in some way, okay?

And so that makes it hard for our immune system, which has evolved not to destroy itself to recognize these targets often.

And so scientists, myself included, have done work on the design of synthetic proteins called chimeric antigen receptors, whose purpose is to recognize new proteins, specific proteins that we choose, that are overrepresented on cancer cells, okay?

And the idea is that this synthetic receptor will emulate the behavior of this native receptor, such that when this recognition of some sort of cancer-specific marker on the surface here is recognized, it will trigger that same sort of downstream cascade of killing, and yeah, releasing toxins and killing the target cell, okay?

And this is a paradigm which has actually been used clinically, not my products, but it has been clinically used to treat various kinds of B-cell lymphomas that were previously thought to be untreatable.

And people have been, basically, their T cells are removed.

They are made to express this new kind of receptor, and then they're put back in the body.

And then those cells go out and they seek B-cell lymphoma cancer cells, and they destroy them.

And this has actually been the case where many people that basically were on the brink of death are now alive to this day; pretty amazing.

Now, you don't have to stop at a single input recognition.

So you can take that same idea that I told you about with the lactose, right, where you have a bacteria that's making a sort of Boolean logic decision about what's going on, and you can expand this kind of decision-making to your engineered cell products.

And so when I was in industry before starting my professorship, I worked on this strategy for expanding that kind of target cell recognition between an engineered T cell and a cancer cell to basically display more complex logic.

And what this was useful for, and I won't go into the details of how these receptors work, but a challenge for that paradigm I just told you is that cancer cells, as we discussed, are essentially our own cells, but with some problems, right?

And so a single thing that tells you, that you try to use to identify a cancer cell is often not sufficient to be able to not destroy other tissues in your body, right?

And in fact, many failures when people try these technologies clinically for the first time was they would kill the cancer, but then they would also go off and destroy other tissues in the patient.

So it looks like it's working, but then your heart gets vaporized by these cells.

And so by building these kind of more complex combinatorial logic gates, we can make a stricter requirement for what kinds of signals are required for a kill.

That is, you might need to, kind of like the old game of Guess Who, "Are you wearing a hat?

"Do you have glasses?

What color is your hair?"

If you can build up a profile for your cancer cell, then you can encode that target for your thing and achieve more selectivity.

It's pretty cool.

Okay, so with the last five minutes, I wanna tell you a little bit about how we can integrate those two parts of manufacturing and computing together in one last arc to kind of think about creating cells that can act and do things for us that are more like robots.

And my students are here, so they might be mad about this next part because one thing that we're trying to do in my group is program cells to actually be able to be the scientist for us.

That is, to make measurements about themselves and report back to us, okay?

You can imagine this cell being like, "I'm gonna do this thing to myself and have your results back for you quickly."

And so, my joke is now you have to pull it back.

But it was my plan to replace my graduate students with cells.

No, of course not, because they're the engine that is creating all of this results.

But that's one idea we have is, can we turn the cells into the scientists for us?

And an area that we focused on quite a lot is actually this little text bubble here 'cause how do you get a cell to tell you what the results of its computations are in real time?

How do you get it to communicate to you what's going on or what exciting result it has?

Sometimes I have a hard time getting my students to tell me what exciting results they have.

So how do you get a cell to do it?

Well, this is where we can start to connect back to some of those earlier slides I showed you.

'Cause remember how I told you that those bacteria perform that geometric computation by creating this kind of wave structure like I showed you here?

Well, it turns out that we can make cells, human cells, manufacture this wave generation system for us.

And when we put those into human cells, they create these stunningly powerful and beautiful oscillations.

So these are human cells, where you're seeing protein oscillations traveling through them.

It's some beautiful psychedelia, for sure.

My dad is a child of the '60s, so he loves seeing these videos.

But I wanna emphasize that it's not just about creating cool visuals because it turns out that the oscillations that we can produce are extremely powerful as a communications media for us to go back and forth between what the cell is measuring and what's going on.

And that's because these oscillations actually enable us to turn cells into the objects that are almost like radio stations, where each cell will produce some sort of oscillating data structure, which we can encode all of its calculations and computations in.

And the idea is quite simple.

We have a beautiful image of these cells and you could stare at it all day.

And my dad just says, "Keep 'em coming, keep 'em coming."

And I'll put my Grateful Dead records on.

But the power of this structure is that if you actually analyze the intensity of these oscillations over time, they'll produce this astonishingly regular oscillation inside a human cell.

These are the kind of oscillations that never exist inside these cells on their own.

And that means that what we have is a unique sort of high frequency signal running inside these cells just like a radio station.

And so we can use techniques from electrical engineering.

Those would be things that would be pretty much undergraduate level electrical engineering tools, where we use things like Fourier transforms, et cetera, to isolate the specific cellular radio signal of interest and project it back onto the cell, okay?

And what this enables us to do is start with data like these, where it's great for my dad, but it's not necessarily scientifically useful on its own.

And then begin with these analysis tools to pull out quantitative, where here, I've color-coded them by different kinds of data that we're collecting, the information that's going on inside here, okay?

And because each cell is sending this data to us on this very unique high frequency structure, it actually provides us an incredibly straightforward way to access this kind of hidden information going on inside the cells.

So it's actually been super useful for us in the group because we can then extrapolate these circuits to do really interesting kinds of computations.

So if the cell has received some kind of signal from its environment that turns something on.

In this case, it's a molecule called PKA.

It doesn't really matter for the purposes of this talk.

We can make it so that when the PKA activity is on, we can see data appearing in some sort of oscillation, okay?

And we can then get cells to encode the results of their computation for us on this little radio.

And it really does look like a radio signal because when we analyze it, we have a carrier signal that's always being broadcast to us from inside the cell.

And then the data that we're interested in, like whether that kinase was on or whether some protein is active or whatever, they appear to us as modulation of the waves that are produced, okay?

And although it seems crazy, it really is just taking the same concepts that are used to process radio data or telecommunications data, and then reencoding them in a sort of biochemical fashion, okay?

And using this approach, we're actually kind of essentially creating these little cellular scientists that can report out on all kinds of interesting stuff.

You, again, don't need to know what PKA or ERK is or mTOR.

They're things that we care about as scientists, and then we send them into, these cells into different environments, and we can start to learn about what's going on in these individuals as they form different layers or as they develop into organoids or even as they potentially develop into a large animal.

So every cell is gonna tell us what's going on.

They're gonna do the experiments for us.

And then my students are gonna put their feet up on their desks and be like, "Ah, perfect."

And it'll be worth it because they did all the work to make the machines do that.

Okay, so with one last thing, I just wanna emphasize that we don't only use these kinds of tools to create data structures.

We can also take advantage of their properties to manipulate the spatial organization of cells in new ways.

And the key thing that enables us to do that is that that pair of proteins that I just talked about, they form what is sort of an old-timey chemical structure called a reaction diffusion system.

And again, you don't need to know what the partial differential equations mean.

All you need to know is that the systems under certain conditions are actually what gives rise to beautiful, beautiful patterns in animals that we see in nature.

So for example, this beautiful fish pattern over here is the result of this combination of molecules reacting and diffusion together.

And they call these Turing patterns.

I should have capitalized the T because this concept was developed by famous mathematician, Alan Turing.

And so a student in my group, Eden, has taken these kinds of components that we've been working with, and he's started manipulating them in new ways to get human cells to actually produce beautiful, stunning patterns inside themselves that are essentially identical to those kinds of patterns that we can create... That were predicted to be created in animal scales.

So this is like a single human cell that now has all these beautiful spots.

And while it's, of course, beautiful, one of the things we're using these kinds of patterns to do is to organize components in the cell in new ways.

So for example, getting them to template the location of receptors for other molecules and things like that.

But it does bring me to one last thing that I believe that we should be able to project onto cells as something they can be, and that's works of art.

Because I really think that these are truly beautiful things that my students have created that are a product of their understanding of how the cell works and the plasticity and manipulability of those systems.

Okay, and so cells don't have to only be a technology.

They can also be a canvas where our understanding of biochemical principles and cell biology could be manipulated to create sort of astonishing displays.

So long term, of course, our goal is to, of course, build new kinds of technologies and products out of cells, where the idea is to emulate those key features of cell biology to enable them to then perform new kinds of tasks.

These might be therapeutic tasks like I talked about with the T cells, but these could also be manufacturing or computational tasks where perhaps cells might be more well-suited for performing certain kinds of computations for us than the traditional computing paradigms we have to this day.

So to just sort of sum up where we're at, I told you about how I like to think about cells in the simplest way.

And of course, I don't want to imply that the cell isn't extremely complex.

It's extremely complex.

But by abstracting a little bit, we can see that there's some key design principles underlying their form and function.

And those design principles can be really understood in terms of things that are familiar to us, like making things or computing or building machines.

And that thinking helps us see ways to repurpose those living technologies or those living cells as technologies where potentially, we can start to reimagine how we might create nanoscopic devices or machines, and ultimately have a future that's not so reliant on burning all this energy and operating in high temperatures, but a world in which our technology is sort of at the same temperature and length scales as ourselves.

And so with that, I just wanna say thank you very much for the opportunity to kind of give you my philosophical worldview on cell biology.

If you're interested in learning more, you can visit our website and read the different things that we've been up to.

I wanna thank the funders who gave us the money to be able to do all those kinds of fun experiments in the lab.

And I, of course, really wanna emphasize, and maybe they'll stand up.

My students, can you stand up?

My students are here and they're embarrassed.

But I really wanna say... [audience applauding] It is my students.... [audience applauding] It is my students who generated everything you saw today.

So they sat in my office and I said, "You gotta do this, and you gotta do that."

And then they went out there and they produced all of this extraordinary science.

And so they are the real heroes, I think.

And when I think about this sort of biologically-powered future, they really inspire me and give me tremendous hope for that vision.

And with that, I'm happy to take questions from you guys about anything that you might have.

[audience applauding]