My view of science | David Linden: Keep your research until the end of your scientific career

David Linden, a famous American neuroscientist and professor of the Department of Neuroscience at Johns Hopkins University School of Medicine.

17 months ago, after surgery to remove a large lump in my pericardial sac, the pathology report came out and the results were not good. I was diagnosed with cardiac synovial sarcoma and have only 6–18 months to live. Because the tumor is embedded in the wall of my heart, it is impossible for a surgeon to remove all the cancer cells without impairing my heart function. My tumor, which used to be the size of a soda canister, has shrunk to the size of an apricot. After the surgery, I underwent radiation therapy and chemotherapy, which was filled with all kinds of unpleasantness. The last dose of chemotherapy was given 9 months ago, and now the side effects have disappeared. A recent CT scan found that the part of the tumor that remained on my heart that could not be operated on had not grown further and had not metastasized.

By the time I lived to be 60, I found myself in a strange, cognitively uncoordinated state: terminally ill, but feeling good because it posed no immediate threat to my health. Just from the outside, you don't see anything wrong with me. Last week, my wife and I went to a jazz club, and today I'm in the lab talking to students about their experiments, and next week, I'll be on vacation with my family on a lake in Canada. Now, my life is wonderful and I am savoring its sweetness, trying to make the most of the rest of my life in various areas such as family, friends and work.

My colleagues are also in a state of cognitive discoordination. As I walked down the hallway, I would see strange expressions – everyone was polite, but I could tell you that they were thinking in the friendliest way possible, "Shouldn't he be dead now?" ”

For the past 30 years, I have managed a neuroscience lab at Johns Hopkins University School of Medicine. We use cell electrophysiology and to reveal the cellular processes behind the brain's ongoing response to experience. Among them, the range of experiences ranges from sensorimotor signals to brain damage to the long-term effects of drugs or hormones. Neuroplasticity includes subtle changes in the structure of neurons (and other cells in the brain, such as glial cells and vascular endothelial cells), as well as changes in the function of synapses and ion channels that give neurons electrical excitability. Neuroplasticity seems to be the basis for phenomena such as learning/memory, addiction, and functional recovery after brain injury.

This work has been, and will continue to be, a tremendous pleasure. I have this experience, mainly because of the smart, creative, kind people around me – my students and postdocs, but also colleagues and collaborators. I am very grateful to be given a job that follows my curiosity and can work on it in such a stimulating and supportive environment. Such intellectual freedom is a unique gift. In fact, the job also gave me a parallel career – writing books about brain function for the average reader, which made me even happier. By any reasonable standard, I've done well in science. At the same time, I am fully aware that, as a well-educated white person, I have gone more smoothly on this path than many others.

These days, my lab is still running, but the scale of our work is gradually decreasing. We no longer accept any new projects or personnel, nor do we write any new grant proposals. I hope I can live long enough to help this current group of outstanding students move on to the next stage of their careers. We have six ongoing projects that we hope to complete and publish and then our lab will stop. Having worked in the lab for 42 years (starting with my first undergraduate lab job in 1980), it's hard to tell. The idea that there will always be exciting new experiments to do and that I will have to stop is a challenge to my deepest sense of self: Who am I if I am not an experimental scientist?

As scientists, we have always struggled with a fundamental contradiction. We have to go deep into the present moment to really engage with people, data, and ideas. At the same time, we must always mentally predict the future – what happens if we get outcome X? What insights might arise from applying technology Y to our question? How does another lab report new discovery change our perception or experimental design? We can never be completely "here and now". We are always obligated to think about the next thing, but that's okay.

I wish I could tell you that my impending death has given me some new or surprising revelations about how best to have a happy, fulfilling, and productive life in science, but that's not the case. The secret is simple and well known: stay curious, embrace the uncertainty inherent in science, enjoy and respect your colleagues (especially your fellow students), and remember our ultimate goal to understand the natural world, transcend journals, grants, and awards, and try to live longer than anyone else.

While this isn't "a magic trick," I'll offer a small but concrete piece of advice, or rather, repeat it, because before me, others have already said that it's to regularly take a break from the day-to-day experimental science challenges, lift your head, look around, and think deeply about what the scientific question you really want to answer is. It's an exercise that I try to try every few years, and I've achieved different results. I was doing the same thing in the last moments of my life, even though I couldn't turn those ideas into experiments.

So, if you allow me, I want to tell you a question I want to know, and if I can get my lab running for a few more years, it will be the top issue on the list I want to study.

When we look at the schematic of a neuron, we usually find a cell body in which multiple branching dendrites appear, which are the main information-receiving structures of the neurons, where synapses are formed, and a single straight axon, which is the main information-transmitting structure along which action potentials are transmitted. Axons are usually drawn as an unbranched line with a separate bulging portion at the end, the "final buckle," from which neurotransmitters are released.

However, since the time of neuroanatomist pioneers Camillo Golgi and Santiago Ramony Cajal, it has been known that axons can branch, sometimes repeatedly, dominating many different targets. More recently, the fusion of modern genetics and imaging techniques has enabled neuroanatomists to reconstruct entire axons and dendrites of many neurons in the brains of young mice (Winnubst et al., 2019). To me, one of the most interesting findings of these efforts is that it is very common for a single neuron in the brain to have a large number of branching axons that are widely distributed in different regions of the brain. In many cases, multiple targets of a single axon are not only anatologically far apart, but also functionally distinct, at least at our current level of understanding.

For some neurons, you may have expectations. For example, a single neuron in the blue spot has axons that are widely projected in the brain, and these axons primarily release the neurotransmitter norepinephrine, which has a corresponding slow neuromodulatory effect due to its effect on the slow G protein-coupled receptors, reflecting states such as wakefulness, alertness, and emotion. These signals are widely distributed in brain regions, which are not surprising, as they seem to convey information that changes more slowly, useful for setting the overall tone of calculations over a large area of the brain. In contrast, individual neurons that transmit sensory or motor signals work more often through fast receptors that are directly coupled to ion channels (such as certain receptors of the neurotransmitters glutamate, GABA, and glycine) and transmit rapidly changing information (in the range of milliseconds to tens of milliseconds). One might not necessarily expect this rapidly changing information to be useful for being widely distributed in the brain. However, modern neuroanatomy suggests that it does work.

For years, the mainstream view of axons has been that they have the same function as a slow wire — putting an electrical signal at one end and after a while, it comes out of the other end with essentially no change. If the wire is branched, the signal will propagate through the branching point, thus distributing it throughout the axonal trunk.

In recent years, scientists have put a lot of effort into mapping the axonal connections of the brains of various organisms. The basic assumption of this work is that electrical signals travel reliably through these neuronal wiring, so that, like a schematic diagram of a circuit, a brain wiring diagram will tell us where neural information is flowing. But is this assumption always correct?

There is a question: In a mature, intact, unanesthetized brain, does the action potential reliably propagate nerve signals through highly branched axons, or is it common for action potential failures at branch points to fail?

Of course, you'd think, this is something we already know, but we really haven't cracked it yet.

Axons can appear in many forms, they have different diameters associated with their conduction velocities, and axons wrapped in myelin from special glial cells, oligodendrocytes, can conduct action potentials more quickly. Typically, a single axon emerges from the cell body of a neuron, and its starting segment is specialized in a number of ways (most notably the high-density voltage-sensitive sodium channel in the plasma membrane) to produce a short, stereotyped electrical signal called an action potential (or spike). Once generated in axonal hillocks, the action potential travels from the cell body to the end of the axon (it also travels backward to the cell body and dendrites through different mechanisms, but this is later). As the action potential propagates, their amplitude does not decrease like the ripples of a stone falling in a stationary pond, but rather regenerates the peak amplitude as it moves. That's because, in a clever positive feedback loop, the depolarized open sodium channel allows sodium ions to flow in, resulting in more depolarization and spreading this depolarization to adjacent parts of the axonal membrane. When action potentials reach a specific region, the active region, they trigger a series of biochemical events that eventually lead to the probabilistic fusion of neurotransmitter-filled vesicles with the plasma membrane, releasing neurotransmitter molecules into the extracellular space, transmitting signals to other neurons (and some other non-neuronal cell types). These active regions can be located either at the very top of the axon, in the structure of the "final buckle", or in the swelling of vesicles-filled vesicles called axonal bulges distributed along the length of the axon. A single axon can have hundreds of such active zones.

So, what does the experiment show us about the action potential in the true branching axon? In invertebrate preparations, such as leeches (Yau, 1976) or crayfish (Smith, 1980), the answer is complex. Axonal branch failures sometimes occur and sometimes do not. It is more likely to occur during high-frequency bursts of action potentials, which is a common pattern of firing in the neurons studied.

In the mammalian brain, there are some experiments on brain tablets in the neocortex of juvenile rodents (Cox et al., 2000; Koester and Sakmann, 2000) or cerebellum (Foust et al., 2010) or in young neurons using disperse cultures (Mackenzie and Murphy, 1998), show reliable propagation of action potentials through axonal branch points. Other studies using mammalian hippocampal organotypic cultures have used paired somatic recordings to infer axonal branch point failure, although axons have not been directly documented (Debanne et al., 1997). But as far as I know, the experiment we really need to solve this problem, about the brain of a mature, unanesthetized mammal, measured with its anatomical and neuromodulatory integrity, is not yet complete.

Fortunately, contemporary tools allow for this type of study, at least in those parts located within 500 microns of the brain's surface (Broussard and Petreanu, 2021). The experiment that solves this problem involves a technique used to visualize the structure and function of neurons in the intact mammalian brain, called in vivo two-photon microscopy, and involves removing a portion of the skull to implant a glass window that provides optical access to the brain. It is now possible to use genetic tools to sparsely express fluorophores that dynamically report transmembrane voltage or internal free calcium concentrations in genetically typed neuronal axons (as slower agents of transmembrane voltage, with certain important considerations), and then make fast-delayed films in the wakeful, intact brains of mice carrying these reporters. As the action potentials propagate through axonal branch points, scanning two-photon imaging at a speed of 500 to 2000 Hz is enough to resolve the action potentials to determine whether they go off at the branch point, enter one subbranch, or enter two subbranches. However, another interesting possibility is that the shape of the action potential changes at the branch point (it is known that during high-frequency bursts, the action potential in the axonal endings also changes [Geiger and Jonas, 2000] and responds to certain neurotransmitters).

In my opinion, this is a crucial experiment in understanding how information flows through neural circuits. When we observe the contact between a neuron in region A and a neuron in region B, can we assume that the information flows reliably to the presynaptic side of the synaptic connection between A and B? I strongly suspect that, like most things in biology, the answer is "it depends". Perhaps the action potential can only propagate reliably through the branch points of larger diameter or myelin-sheathed axons, and not in smaller myelin-free axons. Perhaps the probability or pattern of branch point failures depends on the dynamics of the information being transmitted. One can imagine, for example, that during a high-frequency burst, only the first few action potentials propagate through the branch points, while subsequent action potentials fail in various ways. It is not unreasonable to assume that the recent activity history of axons (e.g., within the last 200 milliseconds) may affect branchpoint failure. Finally, it would not be surprising at all if axonal branch failure is altered by the overall state of the animal's interior (drowsiness, hunger, anxiety, pregnancy, etc.) as well as psychotropic drugs, brain temperature, or anesthesia.

So, you know, it's not a completely novel idea, and I'm sure some labs are already working on some aspects of it. Of course, it builds on the experimental and theoretical work of many others. I think this is a fundamental question of understanding neural circuits, and I hope you agree with it too. If I had the time, I would definitely follow this path of investigation, and I hope so will you.

[The author is David J. Linden, a famous American neuroscientist, professor of the Department of Neuroscience at Johns Hopkins University School of Medicine, editor-in-chief of the Journal of Neurophysiology, and author of best-selling books "The Evolved Brain" and "Looking for Cool Points". At the end of 2021, he wrote about his mixed feelings when he was first diagnosed with cardiac synovial sarcoma, as well as his perception of his situation in the face of death. 17 months after his diagnosis, he has a new understanding of the terminal illness and his own life: "The diagnosis of advanced cancer made me think about my scientific career, about the joys and surprises it brought me, and the experiments I would have done if I had more time." This article was first published on September 21, 2022 in Neuron, a top journal of neuroscience under Cell Press, with the original title of "A life in science, ending soon." The Paper was authorized by Cell Press to publish this article. This article was compiled by Cao Nianrun. 】

Bibliography:

1. Broussard, G.J., and Petreanu, L. (2021). Eavesdropping wires: recording activity in axons using genetically encoded calcium indicators. J. Neurosci. Methods 360, 109251. https://doi.org/10.1016/j.jneumeth.2021.109251.

2. Cox, C.L., Denk, W., Tank, D.W., and Svoboda, K. (2000). Action potentials reliably invade axonal arbors of rat neocortical neurons. Proc. Natl. Acad. Sci. USA 97, 9724–9728. https://doi.org/10.1073/pnas.170278697.

3. Debanne, D., Guerineau, N.C., Gahwiler, B.H., and Thompson, S.M. (1997). Action potential propagation gated by an axonal IA-like K+ conductance in hippocampus. Nature 389, 286–289. https://doi.org/10.1038/38502.

4. Foust, A., Popovic, M., Zecevic, D., and McCormick, D.A. (2010). Action potentials initiate in the axon initial segment and propagate through axon collaterals reliably in cerebellar Purkinje neurons. J. Neurosci. 30, 6891–6902. https://doi.org/10.1523/jneurosci.0552-10.2010.

5. Geiger, J.R.P., and Jonas, P. (2000). Dynamic control of presynaptic Ca2+ inflow by fast-inactivating K+ channels in hippocampal mossy fiber boutons. Neuron 28, 927–939. https://doi.org/10.1016/s0896-6273(00)00164-1.

6. Koester, H.J., and Sakmann, B. (2000). Calcium dynamics associated with action potentials in single nerve terminals of pyramidal cells in layer 2/3 of the young rat neocortex. J. Physiol. 529, 625–646. https://doi.org/10.1111/j.1469-7793.2000.00625.x.

7. Mackenzie, P.J., and Murphy, T.H. (1998). High Safety Factor for Action Potential Conduction Along Axons But Not Dendrites of Cultured Hippocampal and Cortical Neurons. J. Neurophysiol. 80, 2089–2101. https://doi.org/10.1152/jn.1998.80.4.2089.

8. Smith, D.O. (1980). Mechanisms of action potential propagation failure at sites of axon branching in the crayfish. J. Physiol. 301, 243–259. https://doi.org/10.1113/jphysiol.1980.sp013202.

9. Winnubst, J., Bas, E., Ferreira, T.A., Wu, Z., Economo, M.N., Edson, P., Arthur, B.J., Bruns, C., Rokicki, K., Schauder, D., et al. (2019). Reconstruction of 1, 000 projection neurons reveals new cell types and organization of long-range connectivity in the mouse brain. Cell 179, 268–281.e13. https://doi.org/10.1016/j.cell.2019.07.042.

10. Yau, K.W. (1976). Receptive fields, geometry and conduction block of sensory neurons in the central nervous system of the leech. J. Physiol. 263, 513–538. https://doi.org/10.1113/jphysiol.1976.sp011643.