Watch Episode Here
Listen to Episode Here
Show Notes
This is a ~1 hour talk by Jack Tuszynski (https://en.wikipedia.org/wiki/Jack_Tuszy%C5%84ski,https://apps.ualberta.ca/directory/person/jackt) on electrical and electromagnetic properties of microtubules.
CHAPTERS:
(00:00) Cellular electrodynamics overview
(35:14) Conductivity–capacitance experiments
(46:26) Microtubules as computation
(53:02) Modeling, SynBio, integration
PRODUCED BY:
SOCIAL LINKS:
Podcast Website: https://thoughtforms-life.aipodcast.ing
YouTube: https://www.youtube.com/channel/UC3pVafx6EZqXVI2V_Efu2uw
Apple Podcasts: https://podcasts.apple.com/us/podcast/thoughtforms-life/id1805908099
Spotify: https://open.spotify.com/show/7JCmtoeH53neYyZeOZ6ym5
Twitter: https://x.com/drmichaellevin
Blog: https://thoughtforms.life
The Levin Lab: https://drmichaellevin.org
Transcript
This transcript is automatically generated; we strive for accuracy, but errors in wording or speaker identification may occur. Please verify key details when needed.
[00:00] Jack Tuszynski: So Mike, I've been interested in things regarding living cells that not that many people, except for you and other people, are interested, namely the electrical and electromagnetic design involved in cells and the cell functioning. Much of it was done initially by computational or theoretical means. More recently, I've been doing experiments in my lab and with some collaborators. I'll try to cover this territory today in this presentation. Stop me at any point if you wish to get deeper into it or after. This is just the various agencies in Canada and beyond and companies that were providing support for this work. More coming up. There are two basic aspects, namely to understand the basic science behind electrical properties and electromagnetic properties of cells. Different types of cells have different properties. The second is applied aspects: how do we harness it for A, therapeutic, B, technological, and C, diagnostic aspects of our health? This is the outline. I don't know if I'll cover everything, but I'll try. My main interest for the past two decades in terms of the structural elements of the cell are microtubules, actin filaments, motor proteins. The machinery of the cell in terms of mechanics, morphological changes, and dynamics as well. Microtubules in particular and tubulin have been the central structures I've been focused on. We did some experiments on their electrical properties, electrophoretic mobility. We discovered in the process, as I'll try to report, that all of this depends really on the ambient conditions. Some of this was done in cells. Sometimes we use purified protein in solution. The type of solution that was used affected the results. One lesson I would like to convey before we even jump into the data is that in biology there is really no such thing as the object and the environment. It is always interacting with the environment. That's a main difference with physics. Even in physics, not quite true, but we can get away in physics with saying that this is the system and this is the environment. In biology, the system and the environment pretty much is one thing. I'll try to give you some examples of this later on. So then we measured conductivity and capacitance properties that physicists love. I'm a physicist, so I wanted to understand how that plays out. Then in the second half I will try to discuss other things such as electromagnetic fields. It's a big difference, and sometimes I have to spend a little extra effort to explain to non-physics audiences the difference between electric fields and electromagnetic fields. It's a major, major difference. In the past few years, we looked at quantum effects. I don't know if I'll have time to cover that towards the end, but there's also the enigmatic and highly intriguing and controversial aspect of whether biology and quantum physics have anything in common.
[04:23] Jack Tuszynski: I believe the answer is in the affirmative, but how much and how so are still open questions. At the end I want to talk about aspects of therapeutic advantages of understanding electromagnetic properties of cells. The big goal, the major objective of mine, is to uncover a blueprint for cellular electrodynamics. I know that you've been working on an electrical blueprint, especially for membranes and ion channels, and how that can be understood as an electrical circuit, and that's hugely important. But I think inside the cell there's another electrical system which also interacts with electromagnetic fields. As this slide illustrates, microtubules have always played a major role in cells because they play so many different functions. This is a hugely interesting structure with multiple functional roles in cells and living systems in general. They provide cell shape and rigidity, mechanical and morphological properties, and they are instrumental in dividing cells by forming mitotic spindles as shown in the middle. Actin is also important; it is illustrated in blue in the central panel. Motor proteins such as kinesin and dynein move processively on microtubules. That means they are moving in a given direction, kinesin to the plus end and dynein to the minus end. All of this is well known; great credit to cell biologists for uncovering these facts. But underneath all of this is an expanse of electrical properties, because, in addition to being structurally interesting, microtubules are formed from alpha-beta tubulin heterodimers, which polymerize to create these cylindrical structures, typically with 13 protofilaments or vertical strands in the illustration. Sometimes fewer, sometimes more. It's not always as symmetrical in cells; things can get messy and you may have defects. I don't want to get into this territory too much because that's a separate area. It's important to keep in mind that sometimes these beautiful illustrations need to be taken with a grain of salt. However, what should not be taken with a grain of salt is that microtubules and tubulin are highly charged electrostatically. This graph illustrates the special character of tubulin. Number one, the net charge on a tubulin dimer, which is the building block of microtubules (about 8 nanometers long and 5 by 6 nanometers in width and depth), is minus 50 uncompensated negative charges in vacuum. If you put it in a cellular environment, in an electrolytic solution, counterions congregate and screen it, some condensing on the surface, which cancels some charges but not all, and we'll come to it. That leads to a somewhat complicated situation, and it's one example where the environment and the system are tightly connected. The environment being the cytoplasm, or a buffer solution in which you do experiments. Because there's so much charge on the surface, you cannot separate it from the surroundings. The histogram shows the uniqueness of tubulin in terms of charge. When you screen most of the charge, what you have left is a dipole moment. On the right panel, you also see the histogram of all proteins.
[08:47] Jack Tuszynski: These are all proteins. Tubulin is an outlier with roughly 2500 Debye, which is a physical unit of the dipole moment. For comparison, a water molecule has about 1.8 Debye. Water is a polar environment. That also plays a role. You cannot abstract it out because water will attach to the hydrophilic surfaces; most proteins, including tubulin, have hydrophilic surfaces and will create a complex network. That's still not really investigated in great detail, but roughly speaking, the negative charges on the surface will attract the positive charge on the water molecule, which is the hydrogen. The oxygens stick outside of the surface and form a hexagonal-like lattice in terms of the orientations. So it's not random from the point of view of electrostatics as well, because these dipoles will create a lattice with a potential for ordered dipole moments on that surface. We'll come to it because all of it matters in terms of systems' response to electric and electromagnetic fields. All the charges, whether they are in the form of an uncompensated charge, which we call in physics a monopole, or dipoles or quadrupoles or more complex arrangements, feel electromagnetic and electric fields, and they will respond to them. The response function is called the susceptibility, the dielectric constant. So that's kind of the introduction to this work and conceptual framework. Because microtubules play such a major role in cell division and intracellular transport, they also modulate ion channels. We'll come to it. One specific aspect of the tubulin dimer: on the left-hand side, you can see the alpha-beta tubulin heterodimer and you see these C termini which are sticking out. C termini, which face the environment, the solution in which microtubules are located, contain about 40% of all the electric charges. They are flexible, so you can think about them as antennae. They will respond to electric fields and electromagnetic fields. Microtubules affect ion channel activity; C-termini in particular interact with voltage-dependent ion channels. They can open them or cause closing, et cetera. A more complicated aspect of these interactions is how electromagnetic fields interact. These are oscillating electric and magnetic fields propagating with the speed of light, roughly less in dielectric media, but comparable speeds. That opens a Pandora's box of possibilities. Electromagnetic fields span many orders of magnitude in frequency, from low single-digit hertz upward. If we want to talk about effects on the human body, we have to understand the processes in the human body occurring on similar time scales or frequencies; the inverse of time is frequency if the process is cyclical. When you look at processes which may couple to external electromagnetic fields, you should typically look for corresponding time scales or frequency scales. There will be vastly different responses.
[13:11] Jack Tuszynski: In terms of physiology or cell biology, depending on the frequency, starting from single-digit hertz, which would correspond to the heartbeat, and then you have kilohertz, which would correspond to enzymes producing ATP, F-ATPases. This is kilohertz, milliseconds, and in between you'll have EEGs, which are the cyclical processes in the human brain from 10 to 40 hertz. And then you go up to hundreds of kilohertz, megahertz, gigahertz, terahertz, and then visible light, which is 10^15–10^16, and ultraviolet even higher. Physics kicks in because, from quantum mechanics, Planck's formula shows energy is proportional to frequency through the Planck constant. At high frequencies, you have quantum processes, photons, quanta of light. At low frequencies, you have largely classical slow processes. Now all of this interacts, and we examined some of it in terms of microtubules, how these interactions affect microtubules and uncovered some interesting behaviors. This is a different view of a microtubule. On the left you see, from many years ago, a study published in 2001, which was groundbreaking at the time, one of the most computationally intensive. That's not an artist's rendition. This is a computer representation of the electrostatic surface of microtubules with atomic resolution. Every atom was represented. Each tubulin dimer has about 10,000 atoms, so microtubules have millions of atoms, and each one contributes its electrostatic properties. You can see here red is negative charge and blue is positive, and in this case is symmetrically and beautifully represented. The two ends, the plus end and the minus end in panels B and C, are not identical. Biologists knew for decades that microtubules are asymmetrical: the plus end and the minus end are different. The same is true of actin filaments; they have the pointed end and the barbed end. For microtubules, the plus end is the growing end and the minus end is the shortening end. In the upper right-hand corner, the surface is a bit more nuanced and more detailed; the distribution of charge is much messier if you really zoom in. You can also see nanopores, black spaces between dimers where molecules such as paclitaxel, used for chemotherapy, penetrate and go into the microtubule lumen. A lot of interactions take place on the surface with other proteins, including motor proteins. This is electrostatically driven to a large degree, both in interactions between tubulins to form the microtubule and in surface interactions. The same authors later showed that, for example, for actin polymerization processes, if you were to turn off electrostatics (computers could do it but in the lab it's not so easy), the polymerization process would be reduced by roughly 30 to 40 percent. Electrostatics account for at least that much in terms of the rate of reactions. It's short range because it's typically screened.
[17:35] Jack Tuszynski: This is showing that something I mentioned before, namely polarizability. Polarizability means the response of the system. We are abstracting it slightly to an externally applied electromagnetic field, and polarizability can cause several distinct effects. One is, as you can see here in the middle, C-termini flail around and electromagnetic fields will affect their motion, which could be disruptive for motor transport, for example. The second thing is charges, whether individual charges or in the form of dipoles, will respond to the fields, and positive charge will move to the negative field direction and vice versa. Positive charges to negative will be attractive. So that's dipole stretching, charge movement. You also have reorientation of dipoles to align with the electric field. Stretching, reorientation, alignment. Finally, water molecules. At the bottom right-hand corner you see a water molecule with the charges and 1.8-Debye dipole moment. Water molecules will reorient themselves, but if the field is oscillating, they will oscillate with the field. That's polarizability. We were lucky enough to collaborate with Professor Dogaru from the University of Central Florida, where we measured for the first time very accurately several things, including polarizability as a function of pH. This is the first indication that the medium cannot be separated from the system, because if you change the pH of the medium, you change the polarizability of the system. This is shown as beta of tubulin, and you can see how that changes. It's even hysteretic; it depends on the change of pH and yields different values. In the right-hand panel, you see the dipole moment of tubulin ranging from 2000 to 1920. As you change pH from only 6.5 to 7.5, which is very physiological. Physiologically, pH even can change down to one in the gastrointestinal system. The net charge changes with pH. Because some of the amino acids can be protonated or deprotonated, the charge on the system can vary depending on the pH. That's the first experimental data showing that the system is composed of the environment as well. These are some illustrations of the microtubule drift with electric fields. I'll show it again.
[21:59] Jack Tuszynski: You see how microtubules glide and drift. This is random at 0 field and then you increase the field to 50 volts per centimeter, which is not a lot. It's actually comparable to the fields in the cytoplasm. And they move with the field. There are micrographs from higher electric field experiments. In the presence of higher electric fields and fields with different frequencies, electric fields which are alternating, there are additional effects seen: realignments, reorientation of microtubules, and even formation of non-trivial structures. This is with frequencies going up to 2.5 megahertz. We explored even higher frequencies, as you will see later on. The intensity of the electric field or electromagnetic field is the first parameter that affects the results. The second is the frequency of the electromagnetic field. The third is orientation with respect to the microtubules and the cell plane in which the cells are present. The effects involve dielectrophoresis on microtubules, electroosmotic flows because there is some electrical gradient formation, and thermal conduction, small but non-zero. So these three are the main physical effects affecting the orientations and the formation of these structures. The next element of complexity is structure formation. External fields are now affecting internal structures. We start from tubulin, which is the building block of microtubules. Microtubules then form, as anybody who's taking cell biology knows, various structures, not just microtubules per se, but also centrosomes, axonemes, the various composite structures made of microtubules in various arrangements. The centrosome is probably the most intriguing and important because it's touted to be the eye of the cell or the processing unit in the cell. At different frequencies, you will have different mechanisms. I'm looking at resonance because all frequencies will contribute, but some sensitively and some not so sensitively. You have the range here and different structures. The rule of thumb in physics is this: the larger the structure, the slower the process is. The larger the structure, the lower the frequency that interacts with it. Small things like atoms will strongly interact with visible or near-visible, near-infrared. Molecules, maybe terahertz — water especially — and then the bigger it is, the lower the frequency that you should be looking at for significant interactions. The first group of dynamical experiments involved conduction. We suspected that ions will flow along microtubules.
[26:23] Jack Tuszynski: Microtubules can form conductive biological wires inside the cell. I think we proved to be right, as I'll show in a second. But we still don't know whether what's flowing along microtubules is just ions such as sodium, potassium, or protons, or proton hopping from water molecules, or electrons through the protein, which goes back to speculations of Szent-Györgyi in the 1950s that proton-filled proteins can be conductive electronically. I believe that may be the case under some specific conditions. We did experiments on microtubules and tubulin under various conditions, different ionic strengths and different tubulin concentrations. That was very eye-opening because we found a lot to our surprise. Before I show you the results that surprised us, microtubules may also interact with ions in different ways. Number one, they can impede the flow of ions because there's just very bulky obstacles in the otherwise conductive fluid. They can also attract and condense ions, which is known as the Manning theory of electrolytic solutions. They can also accelerate or conduct ions along their lengths. Finally, you have C-termini, which are very flexible and they can cause fluxes and even act as batteries. There were some experiments. This is a summary of different experiments. The values of conductivity of microtubules in Siemens per meter, as you can see in the literature, vary by six orders of magnitude. I suspect much of it is due to the experimental conditions, which were different. The system is part of the environment. If you're not careful, your measurement may not reflect the system per se, but the environment in which the experiment was conducted. We did experiments almost 20 years ago. These were the first surprising results published in Biophysical Journal with Horacio Cantiello, who did the experiments, and we did some computational explanations. It showed microtubules. In the micrograph, you see a single microtubule grabbed by two micropipettes with electrodes, electric signals were sent from one of the electrodes to the other, and the intensity was measured. We found, to our surprise, that the signal was increased. In other words, a microtubule acted as an amplifier. We call it transistor-like amplification of ion flows. That was interesting and intriguing, but there was a single microtubule and a signal or a pulse.
[30:47] Jack Tuszynski: And we did some computational predictions. This is from the modeling representing microtubules as cables of these RLC networks shown here. A single dimer was represented as a combination of an inductor, two resistors, and a capacitor. A capacitor is shown with an arrow across, which means it's actually variable capacitance. Because capacitance is due to the formation of a double layer. The surface of a microtubule has a negative charge, as I told you, roughly. There's some positive, but mainly negative. The second layer is counter ions, positive potassium and sodium ions, forming the second layer separated by something called the Debye length, which under physiological conditions, the separation is about one nanometer, maybe 9 angstroms or so. If you know physics, electricity, magnetism, first year undergraduate, that will create a cylindrical capacitor. There's also resistance because it's a viscous fluid, so there's resistance along the flow and away from the microtubule. The inductive part is due to the geometry of microtubules: they are cylindrical, but the arrangements of tubulin dimers are helical. So you can create a solenoid, solenoidal flows and that's why there is an inductive part. So this is an equivalent electrical circuit and we predicted these values and waited not quite 20 years — maybe 15 to 17 years — to find some validation of these numbers which were predicted at that time. I'll just skip the models because they lead to some very interesting nonlinear differential equations, but they're not so important for your audience. These are the beautiful and complicated differential equations which we solved that describe the voltage pulse propagating along a microtubule in the form of this increasing amplitude solitary wave. That was a bit more detailed representation later on. Two years later, we published a Physical Review paper on a detailed representation of microtubules in cylindrical geometry with C term and I and so on. This is from you, Mike, and I show it a lot because it's a very powerful diagram showing how transmembrane potential changes depending on the cell type; normal versus cancer cells will have different potentials. This is one of the hallmarks of cancer. We published a review paper on electrical properties of cancer cells, and that's one of the electrical hallmarks of cancer cells. Another one would be a low pH, lower pH outside the cell and higher inside, and some other things. These are the charged properties of cells or constituents of cells. Inside you have the potassium, sodium, chloride, calcium, magnesium ions playing the main roles. Proteins that are most electrostatically important are tubulin and actin and so on. So all of this, and water, which is dipolar, forms a very complex electrical and electromagnetic structure that is dynamic in response to external fields. We know about voltage-dependent ion channels. You work on this and contributed tremendously to the field. But what happens inside is still, to a large degree, a mystery. In the next few minutes, I'll try to cover what we discovered in terms of microtubules. Here are the micrographs of our experiments with three different concentrations of tubulin. You can see microtubules labeled fluorescently and they show up as white streaks. It looks a bit like spaghetti, or some random networks.
[35:12] Michael Levin: This is in vitro. This is a cell-free system.
[35:14] Jack Tuszynski: This is in vitro, purified tubulins. But you have the scale. What's important to note is the numbers of microtubules, as you can see in the samples, increased by a factor of 10 from the left to the right, and then 100 again. Another 10 to the extreme right. You increase the numbers of these conductive, capacitive, and inductive units from the electrical point of view. You would expect corresponding or roughly corresponding changes in conductivity. We did these experiments. This is a schematic of the experiment with the semiconductor characterization device called Keithley. We found, first of all, frequency-dependent responses in conductivity. We compared tubulin to microtubules. In other words, unpolymerized protein to polymerized. There was a big difference: the same amount, but forming polymers increases conductance here. Green is microtubules. Also, with the number of microtubules that increases, but maybe not proportionally. There is also the aspect of capacitance, which we were able to measure as a function of frequency later on. This experiment was published five years ago, 2020. That's the setup. These are the results. Blue is microtubules and red is tubulin, unpolymerized protein. You can see the dramatic difference. The capacitance, which is the imaginary part of impedance, is higher for microtubules than free tubulin. Resistance in this experiment is greater for microtubules than for tubulin. They are more resistive. We did two sets of experiments with opposite results at low ionic concentration, which was 20 times lower, not physiological, something like 4 millimolar sodium. At low ionic concentration, microtubules are highly conductive but not highly capacitive. At physiological or close to 18 millimolar, the opposite happens. They are more resistive and more capacitive. In other words, microtubules change at low ionic concentration from being very efficient conductive wires of ions. At high ionic concentration, they trap ions and become capacitors. They store charges, counter ions.
[38:58] Jack Tuszynski: So you see, from static, from dynamic movement, conductivity is basically movement of charges, capacitance is trapping of charges. High concentration of ions, they trap them. Low concentration of ions, they conduct them. And the question is, at what point does it change? What's the crossover point? It's very important to keep this in mind because the cell may use it controllably to use microtubules for either conduction or trapping of ions or charged objects. These are some additional experiments. When you see this on the left-hand side, you see the concentration of tubulin being increased from 0.22 micromolar to 10 times that to 100 times that. Yet the capacitance of the system is virtually the same within a few percentage points, 10 to the minus 5 farads for the system. And then you have errors and so on, small errors actually. Resistance is also roughly the same because 9.74 times 10 to the 4th is approximately 1 times 10 to the 5th. So you see something interesting, still unexplained happens, that in spite of all of this, the networks themselves have some sort of stability of resistance and capacitance in spite of increasing the numbers of microtubules by large numbers. We were able to extrapolate these values to the 20 micrometer microtubule at the bottom in red, and they correspond quite favorably to our predictions from 15 years prior. We also measured electrophoretic mobility of tubulin, mobility with respect to an electric field applied as a function of pH. There's something very interesting happening here, too. Namely, at pH 5 or so, microtubules become neutral. Below, they actually turn to be positively charged. I kept telling you and your audience that they are negatively charged. At pH 7, if you lower pH, they become neutral and then positively charged. So that's another dial you can turn, and if you want to do experiments, change pH and properties change as well. Change the ionic concentration, they also change from conductive to capacitive at high ionic concentrations. That's what zeta potential shows.
[42:41] Jack Tuszynski: Change, and that'll probably be the end of my talk today. There is something called the Poisson-Boltzmann equation that, if anybody in your audience is interested in how to model these things, is an equation that solves for the electric potential around charged surfaces surrounded by ionic solutions. Because ions respond to the electric field of the fixed charges, in this case fixed on the protein surface. So we solved this for the cylindrical geometry of microtubules and the dielectric constant of the medium. From that, you can calculate the fraction of bound ions to the surface and how their concentration changes as you go into solution. Everything depends on the bulk concentration of the salt in which the microtubules are bathed. You see here the lumen, which is inside, how it depends on the distance to the outer surface of microtubules, and C-termini. This is a summary graph. If anybody wants to get into more detail and read the paper we published in Frontiers in Physiology, I believe it is. It shows you how everything depends on the bulk concentration and the surface from which you measure. Here's the bottom line. From that you can calculate the conductivity, and you can see here in the right-hand side panel. The red straight line is the conductivity of the solution itself as a function of potassium ion concentration. In green, you have the conductivity of microtubules, or around microtubules, as a function of bulk concentration. At low concentration, they have much higher conductivity than the medium. You can see it's actually 100 times or more, a few hundred times more conductive. There is a crossover point at about 100 millimolar, which is close to physiological, between 70 and 120 millimolar depending on the conditions. This is a tipping point where microtubules become either more conductive or less conductive than the cytoplasm. That's interesting because a cell can use it for whatever function. The explanation has to do with the Debye length, which I mentioned. The Debye length is the distance from a charged surface over which electrostatics dominate thermal fluctuations. Beyond the Debye length, the ionic solution doesn't feel the surface charges of microtubules. That value depends on the bulk concentration. In the case of microtubules, as I mentioned in the beginning, everybody says the Debye length is 9 angstroms. Yes, at physiological conditions. If you lower the ionic concentration to zero, it's infinity. When you lower it to 4 millimolar, it's about 100 nanometers. This number changes depending on the conditions. Again, the message of today: the environment is part of the system, and the system responds according to which environment it's placed in. I'll stop here and let you ask questions. There are some more results, but I think I've given you quite a lot to talk about.
[46:26] Michael Levin: Yeah, remarkable. One thing that I'm very interested in is let's assume, which I think seems very likely given what you've shown, that the microtubules implement an electrical network in addition to whatever biomechanical network they have. What we have then is an electrical computational network that is also continuously refactoring itself in terms of structure. This is a very different architecture than any kind of computational system that we make now. Much more relevant to the biology, which is also very plastic in terms of, for example, the bioelectricity that we study; cells are moving around all the time. How much work has been done so far in asking the question, if you make a somewhat biorealistic model of these things, what are the computational capacities of—I know you've done models of memory and things like that, but what are the computational properties of such networks? As they do start to move around, I think the cytoskeleton is a really good candidate for a memory mechanism, but the missing piece is to show despite all those things that a cell is constantly doing to rearrange and move around, refactor that cytoskeleton, how does information stay solid? And I don't think it needs to stay solid because I don't think that's what the biology is doing. I don't think it's locking down information with error correction codes and things like that. I think it's doing something quite different in terms of interpreting it on the fly. Nevertheless, there needs to be some computational modeling of what's actually possible with things like that. How much of that has been done?
[48:17] Jack Tuszynski: You're absolutely spot on in terms of understanding this as a dynamical computational system as opposed to silicon-based computers. It is a system that can be evolvable, erasable, reprogrammable. The architectures can be flexible and controllably rearranged. I'm sure the cells are doing that. We know because dividing cells need microtubules mainly for point-to-point transport and cell division—mitotic spindles, in which case they are oriented very precisely to the chromosomes. There is almost certainly electrical signals or pulses sent for synchronization of segregation of chromosomes. In neurons, it's a different story because they are stabilized; they form parallel bundles. These bundles can perform additional computational operations beyond action potentials, everything we know about neuronal computing. I call it a computer within a computer. That is, microtubules and neurons compute something and maybe they change the response function of the neuron to neurotransmitter release. How much? In one of the papers we did, we estimated how much information you can encode in a microtubule, and that varies depending on how many bits per tubulin, and it could be a lot. But the simplest version, which integrates all of this, namely electrical encoding and decoding, would be C-termini in two states: straight out or bent. We simulated that. If you assume that, then a neuron would have something like a gigabyte of memory through microtubules alone. Compare it to a one-bit switch off and on in classical neurophysiology. We go nine orders of magnitude higher. Each neuron can have at least one gigabyte of memory through microtubules. And since there are roughly 100 billion neurons in the human brain, then if that was the case, you could encode 10 to the 20th bits of information through microtubules. Again, that's nine orders of magnitude greater than just taking neurons as switches, which I think is an affirm to the neuron. That's the simplest. Just assuming something we know for a fact, there could be additional degrees of freedom. We wrote another paper with Travis Craddock and Stuart Hamerov on calmodulin kinase II; another level of encoding could be through phosphorylation, which is well-known signaling in biology. These enzymes phosphorylate tubulins, and then you could have vastly more information bits—at least 1000 times more—if you take this into account through microtubules, and we know that they are phosphorylated by this enzyme, CamK2. There could be more than one channel of information encoding, decoding, and information processing. One could be slow: enzymatic. Another could be fast: electrical. Another could be electromagnetic. That's the second part of this talk, which I didn't have time to talk about, but it exists, and we've done some experiments with Greg Scholes at Princeton and Arat Kalra, my former student, on quantum information storage of microtubules by using tryptophan residues; there are eight per tubulin, and they can be excited by electromagnetic waves. That will be another level of fast information processing and quantum state formation. It's a huge number of possibilities.
[53:02] Michael Levin: Amazing. I think there's a really great opportunity here to use these biorealistic models to then either make them a reservoir in the sense of reservoir computing or try some kind of more traditional connectionist architecture on them.
[53:25] Jack Tuszynski: We don't have a fully blown theory of this, even a model. But each glimpse gives another possibility of using it for various applications. I'm sure the cells over the past 2 billion years or so discovered how to do it.
[53:47] Michael Levin: Yeah.
[53:48] Jack Tuszynski: We're just trying to second guess the cells.
[53:51] Michael Levin: Amazing. That's before putting in all the SynBio stuff about changing the actual tubulins' workings. There's a whole other dimension here.
[54:02] Jack Tuszynski: I forgot to mention. The cells have a repertoire of tubulins. There are 20 genes encoding tubulin in the human cell. They can express different isoforms of tubulin differently. Each isoform has a slightly different electrical property. Most importantly, each isoform of tubulin has a different C-terminus with different numbers of charges and numbers of residues. The number of possibilities is astronomical. That can be technologically exploited if we were able to express tubulin recombinantly and make it into these computer architectures based on proteins.
[54:53] Michael Levin: Amazing. Super interesting. Thanks so much. That's fantastic. Really amazing. You've opened that whole aspect of biophysics, I think, is going to have a huge potential into the future.
[55:11] Jack Tuszynski: And I also want to mention that this is not an isolation because what you do with ion channels and membranes is a layer which interacts with this internal layer. And especially in neurons, that's obvious with action potentials. But not only in neurons, because if you modulate ion channels and change the ionic concentrations inside the cell, that's the environment that microtubules live in. It's a two-way communication. And that is the biophysics of the future. I think you can integrate the cellular, the cytoskeleton aspect with the membrane and ion channels into one interacting network.
