Transformer: The Dep Chemistry of Life and Death, Nick Lane, c. 2022.
Epilogue: Self
"I think, therefore I am," said Descartes in one of the most celebrated lines ever written. But what am I, exactly? An artificial intelligence can think, too, by definition, and therefore "is." Yet few of us could agree whether AI is capable in principle of anything resembling human emotions, of love or hate, fear and joy, of spiritual yearnings for oneness or oblivion, or corporeal pangs of thirst and hunger. The problem is we don't know what emotions are: what is feeling in physical terms? How does a discharging neuron give rise to a feeling of anything at all? This is the "hard problem" of consciousness, the seeming duality of mind and matter, the physical makeup of our innermost self. We can understand in principle how an extremely sophisticated parallel processing system could be capable of wondrous feats of intelligence, but we can't answer in principle whether such a supreme intelligence would experience joy or melancholy. What is a quantum of solace?
Why would I even touch on such a question in a book on the Krebs cycle? The answer is that the flux of metabolism, moment by moment, decade after decade, has to correspond in some way to the stream of consciousness -- what else could animate our innermost being? In this book, we've explored the dynamic side of biochemistry, the continuous flow of energy and matter that makes us alive. I have argued that this flow began in deep-sea hydrothermal vents, electrochemical flow across barriers and membranes coaxed H2 and CO2 to react to form the Krebs-cycle intermediates at the heart of metabolism across life. These in turn gave rise to amino acids, fatty acids, sugars, nucleotides -- the building blocks of life. It might seem uncanny that whole metabolic pathways can spring into existence in this way, in the absence of genes and information, but this is what recent experiments are telling us. There iis something thermodynamically and kinetically favoured about the innermost chemistry of life. I find this unsettling, but that's how it is.
The power of this chemistry to self-organise, to grow, to form protocells animated by the same flux of gases into living matter, gave meaning and context to genes and information. For me, the first genes were random strings of a few letters of RNA, polymerising inside protocells growing in those deep-sea vents. From the beginning, genes copied themselves inside protocells, spreading when they promoted cell growth, regenerating what had come before faster and better. Genes never supplanted the deep chemistry of cells. They conserved it, and they built on it. Four billion years on, genes are still faithfully regenerating the deep chemistry of life in our own cells, at an unfathomable rate of billions of reactions per second. From the beginning, the flow of energy and matter through the Krebs cycle was bound to the electrical potential on membranes. Flux is movement. The electrical potential humming away on cell membranes is movement, too, dancing charge, electrons and protons, the elementary particles of life. Moving charge generates electromagnetic fields that permeate our being. And clearly, the flux of metabolism generates electromagnetic fields on cells. Could feelings somehow be related to this dance of charge, the ephemeral states of cells?
The idea is pleasing, but I wouldn't have give it any more thought but for a visit from a scientific seer. Luca Turin is a bio-physicist interested in quantum biology. He has led an unusual life, working for many years on scent and smell, and the possibility that we may be able to detect quantum vibrational states. When he visited me at UCL, that's what I thought he'd want to talk about. I knew little about it, and I did not feel able to comment. But that wasn't what he wanted to talk about at all: he had mitochondria on his mind. I could judge what he had to say about mitochondria, and it was thrilling. Turin is unafraid to confront the unknown in science, or for that matter the "known" (riling some), but combines his yen for vistas new with a rigorous understanding of some fundamental methods in biophysics, such as electron spin resonance. Rarely have I met someone who thinks in such clear lines. His papers convey this clarity, coupled with wry amusement. "Almost the only thing we know for sure about consciousness is that it is, so to speak, soluble in ether, chloroform and a variety of other solvents." Intriguingly, anaesthetics can dissolve consciousness reversibly not only in humans, but even in the simplest animals, as well as single-celled protists such as paramecium. Turin concludes from this that consciousness is not an emergent property of the complex nervous systems of higher animals, but is something more fundamental that works at the level of cells. This means we can study consciousness in some simple experimental models such as fruit flies. As Turin puts it, "While it is not known to what extent fruit flies are conscious, they are most definitely unconscious when exposed to chloroform or ether."
The mechanism of general anaesthesia is one of the major unsolved problems in science, ranked as such by the journal Science alongside cancer, quantum gravity and high-temperature superconductivity. Our skill in manipulation often outstrips understanding, and in this case we know how to control the effects of anaesthetics with exquisite finesse, but next to nothing about how they actually work. The problem is the lack of relationship between molecular structure and biological activity: molecules of disparate sizes and shapes (precluding a common interaction with some receptor in a normal lock-and-key mechanism) all act as general anaesthetics. Perhaps most baffling is the gas xenon. As Turin points out, xenon has no "shape" (it is a perfect sphere of electron density) and no chemistry -- it is an inert gas. But it does have physics. It is capable of facilitating electron transfer between conductors. Just think about xenon lamps, which generate a white light similar to sunlight. So in principle xenon could induce anaesthesia by facilitating electron transfer. But why on earth would electron transfer induce anaestheisa?
One of the few things that general anaesthetics do have in common is that they are lipid soluble -- they accumulate in membranes, and the strength of anaesthesia depends on their concentration more than their structure. Anaesthetics accumulate in the mitochondrial membranes too. So, could they facilitate electron transfer to oxygen in cell respiration? Turin's work shows that this might be the case. Electron spin resonance gives a signal associated with oxygen, which shifts under anaesthesia (it's the only signal that does) but not in mutant flies resistant to anaesthesia. Even more intriguingly, Turin has detected a radiowave signal associated with electron transfer in respiration. Because all proteins are composed of chiral amino acids (which always have the left-handed form) the transfer of electrons from one protein to another in respiration locks them into the same spin phase, which can be detected as a radiowave signal when they relax upon reacting with oxygen. Don't worry about the details here. The point is that these radiowave signals increase when brain areas are active, and are suppressed by anaesthesia, again implying an effect on respiration. You can see why Turin has a reputation for being difficult. His science is right at the edge of the known. Even he admits that brains emitting radio waves sounds like the stuff of science fiction. But it seems they do.
Let's get back to xenon. All this suggests that xenon concentrates in the hydrophobic pockets of proteins sitting within the mitochondrial membranes, and flits respiratory electrons straight to oxygen. The effect must be subtle, for too much would kills us, and overdose is always a risk with anaesthetics. But suppose it's true. What next? Instead of electron transfer to oxygen being coupled to proton pumping and ATP synthesis, some proportion must hop on a xenon bridge straight to oxygen. That oxygen is presumable still bound to cytochrome oxidase at the end of the respiratory chain in the normal way, so the electrons are not escaping as free radicals. Even so, short-circuiting the respiratory chain must affect the electrical membrane potential, which should be measurable (though these are not easy measurements to make). So ... could it be that a change in mitochondrial membrane potential affects our conscious state?
I mentioned electromagnetic fields. We have long known that the brain generates electrical fields, which we measure in the EEG (electroencephalogram). As with general anaesthetics we know far more about how to interpret patterns in the EEG associated with epilepsy or sleep than we do about what generates them in the first place. According to the neuroscientist Michael Cohen, we know "shockingly little about where EEG signals come from and what they mean." Plainly the EEG is produced by changes in electrical voltage, and these changes are big enough to incriminate large networks of neurons firing in synchrony (rather than individual cells). But these neural networks are still composed of individual neurons, which behave in similar ways. The question is, at the cellular level, which electrical charges are involved? The glib assumption is that charges on the cell membrane (or action potentials) are responsible. But if Turin is right, then a big part of the answer might be mitochondrial membrane potentials. Not only is electron transfer to oxygen implicated in consciousness, but the mitochondrial membrane potential is twice that of neural cell membrane, and the convoluted folds of the mitochondrial inner membrane (the cristae) offer a much larger total surface area of charged membrane.
Moving charge necessarily generates an electromagnetic field, and the mitochondria clearly do so -- not only with the transfer of electrons to oxygen but even more dramatically in the circuit of protons across the membrane, looping from the respiratory complexes to the ATP synthase and back round. Doug Wallace is again at the forefront of the field, attempting to measure the strength of electromagnetic fields in individual mitochondria. But there are some broader principles too about the way that electromagnetic fields interact with each other -- separate fields can interfere with each other (cancelling out) or can link in phase to generate a stronger field, operating over a long distance. Parallel membranes, such as the cristae membranes of neuronal mitochondria, should generate stronger fields that amplify a signal and potentially interact with weaker fields on the cell membrane, modulating neural activity. Could this be what generates the EEG? I suspect so, although that would be immaterial if the EEG is no more than an epiphenomenon, a reflection of some underlying activity with no power to influence anything itself. But there's strong evidence that electrical fields can and do play a direct role in brain function. If you cut the axon of a neuron, for example, and separate the two cut ends by a fraction of a millimetre (which is too far for chemicals to cross quickly), an action potential can leap the gap as if it wasn't there. Electrical fiields can easily explain this behaviour. If so, the key point is that the electrical fields generated by neurons do have motive force. [Electrical fields] are not too weak to change things physically, as long assumed.
This kind of statement might have pushed the boundaries of respectable science until recently but the extraordinary work of the developmental biologist Michal Levin and others shows that electrical fields can control the development of small animals such as the flatworms known as planarians. I suspect that twenty-first-century biology will be the biology of fields. So, let's take it to be possible that the electrical fields generated by mitochondria do have motive force. What can that tell us about consciousness? Well, for a start, it might tell us why the brain is so hooked on glucose as a fuel. If you recall, calcium influx into the mitochondria from their associated membranes (MAMs) activates the enzyme pyruvate dehydrogenase, ramping up Krebs-cycle flux and ATP synthesis nearly exponentially. Plainly that powers work, but it also gives scope to the full dynamic range of mitochondrial membrane potential. To the full range of electrical fields. Too the full music of the orchestra. Until now, biology has tended to study the materials that make up the instruments. The time has to come close our eyes and listen to the music. I want to suggest to you that this music is the stuff of feeling, of emotion. Electrical fields are the unifying force that binds the disparate flowing molecules of a cell together to make a self with moods and feelings. Alzheimer's disease is the fading of that music as the fields fragment.
Let's put aside multicellular organisms with their nervous systems and think about protists such as paramecium, which also generate electrical fields on their mitochondrial membranes. Just watch their amazing behaviour under the microscope, the way they move around, explore, graze, give chase or flee from predators, struggle for their lives, or regenerate themselves, whirring parts and all, after some disastrous encounter. This behaviour is marvellous and sophisticated, and takes place in real time as we watch. What coordinates all of this? Do you think it is coordinated by lock-and-key receptor molecules, genetically specified interactions between proteins? What unifies the whole? What coordinates it as a "self"? Once you think about electrical fields, it is hard to imagine anything else. But then we are faced with another problem. As with our own nervous systems, I'm arguing that the electrical fields in protists are mostly generated by mitochondria, deep within the cell. So why would these internal electrical fields become associated with the self, the life, the potentialities, of an organism as a whole? They're only a bit of a cell, after all. Why would electrical fields iin mitochondria, generated by flux through the Krebs cycle, equate to the strivings of the self?
To understand why -- why the language of mitochondrial electrical fields become bound to the ephemeral states of cells -- you need to know that mitochondria were bacteria once. They were engulfed by other cells nearly two billion years ago. I explored the extraordinary ramifications of that relationship in The Vital Question. All we need to know right now is that the electrical potential on mitochondrial membranes inside our own cells is the same as the charge on the plasma membrane of bacteria -- the membrane bounding the cell, which separates and links the self that is a bacterium with the outside world. For the bacterial self, the plasma membrane is the threshold of the known universe. Everything else is shadow.
Let me give you an example of how important this membrane potential is to bacteria. In the ocean there are about ten times as many viruses that attack bacteria (phages) as there are bacteria. You might have seen pictures of these viruses attaching themselves in droves to bacteria. Phages are remarkable, mechanical structures, like miniature lunar landeers, with the malevolence of H. G. Well's Martian tripods, anchoring themselves to the surface and injecting their DNA at high pressure into the soft body of the bacterium. Scores of them can line up like an invasion from outer space. The poor bacterial cell doesn't stand much chance, but it does have defences -- defences that we have recently learnt to exploit, called CRISPR, which allow for sophisticated gene editing. If the bacterium (or indeed its ancestors) has had an earlier exposure to the virus, it can recognise the viral DNA and marshal a counterattack, chopping up the DNA into bits before the virus can copy itself. But the time window is short. If there are too many phages, then there's only one way out for the bacterium: die, fast, for the good of its kin. How does it do this? It yanks open pores in its cell membrane, collapsing the electrical potential. It dies almost immediately, before the virus has a chance to copy itself and infect its sister cells. As a result of this sacrifice, at least some of its genes may live on in those sister cells -- the bacterial equivalent of us laying down our lives for family.
I have long wondered if that collapsing membrane potential "feels" like something to a bacterium. More than anything else, the humming electrical potential on the membrane betokens the living force. And if it feels like something for a bacterium to die, its living force sucked away, what about other modulations in the electric fields generated by membrane potential? Viruses are not the only things that kill bacteria. They can die by wear and tear too -- damage called by overexposure to bright light, or too little iron for the photosynthetic systems to work properly or toxins squirted out by neighboring cells. Each of these can trigger waves of death through marine blooms of cyanobacteria. All operate through much the same mechanisms, collapsing electrical membrane potential to induce death. Presumably, there must also be some "pre-death" state, where the living processes are tenuous. Beyond that, membrane potential is needed for far more than the basics oof ATP synthesis and CO2 fixation. It powers the bacterial flagellum, allowing cells too move around and seek better conditions, as well as pumping all manner of things in and out of cells, maintaining their homeostasis. Most strikingly, bacteria need their membrane potential to find their own midpoint, to divide in two and generate offspring. Nothing in biology is more sacred than reproduction, and the simplest form of reproduction does not happen without an electrical charge on the membrane. All these states of living and dying are linked with electromagnetic fields. Do they all feel different? How could they not? Metabolism and electromagnetic fields on the membranes bounding cells are intimately entwined and intrinsically meaningful. These are the living states of cells, the stream of consciousness in its most elementary form.
Shrink yourself down to the size of a molecule in the Krebs cycle. Succinate. The cell you're part of is the size of a city, a metropolis on the scale of London, Tokyo or New York. What connects you with another molecule of succinate across the city, twenty miles away? You won't even be succinate for long. In a tiny fraction of a second, you will transform into malate, then oxaloacetate, or perhaps an amino acid or sugar. You are a fleeting moment in the kaleidoscope of metabolism, shape-changing a billion times a second. There's no sense in which information binds you. Yet you are still part of an entity, a self. Your flux through the Krebs cycle is linked, moment by moment, with the balance of metabolism, with how much there is of you compared with the next molecule. You are bound too the flow of electrons, to the pumping of electrons, to the electrical charge on the membrane, to the genes switched on and off. Protons scuttle instantaneously around the membrane, equalising the charge at each location, cohering an electromagnetic field that exerts its force everywhere in the cell. The water bound to surfaces inside the cell oscillates in phase, uniting the molecules of metabolism in the symphony. Changes in the outside world -- in food, electrons, protons, oxygen, heat or light -- all are converted through metabolic flux into dancing electromagnetic fields, shifting the mood, the living states of a bacterium. You have just been part of something magic, the flow of life through a living cell on this restless planet of ours, the rush of change that forges the oneness of self. You are a moment in a life.
No comments:
Post a Comment