Tiny Robots Helped Paralyzed Mice Walk Again

In this episode, we explore how researchers at ETH Zurich developed microscopic robots that combine stem cells with magnetoelectric nanoparticles to repair damaged spinal cords.

author avatar

18 Nov, 2026. 11 minutes read

This article was first published on

the-next-byte-wevolver.simplecast.com

 In this episode, we explore how researchers at ETH Zurich developed microscopic robots that combine stem cells with magnetoelectric nanoparticles to repair damaged spinal cords. By using external magnetic fields to stimulate nerve regeneration, scientists helped injured zebrafish regain nearly normal swimming behavior and significantly improved mobility in mice, offering a promising new approach to treating paralysis without implanted electrodes.



<style>p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px 'Avenir Next'; color: #000000} p.p2 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px 'Avenir Next'; color: #000000; min-height: 19.0px}</style>

Farbod 

Hey folks, welcome back The Next Byte podcast. And let me just ask you a question. What if repairing a spinal cord injury meant sending millions of microscopic robots loaded with stem cells directly to the damage and then turning them on with a magnet? It sounds super sci-fi, but actually researchers in Switzerland just got injured mice walking better using exactly that approach.

Friends, welcome back to the next bike podcast. As you heard today, we're going to be talking about spinal cord injuries. But before we get into that, let's talk about today's sponsor, Mouser Electronics. Now, if you've been rocking with us for a minute, you know we love Mouser. And one of the reasons we love Mouser is because, you know, being one of the world's biggest electronic distributors, they have great connections to academia. They have great connections to industry. They see trends happening before they reach the average person, and they write about it.


So we're going to be linking one of their resources in the show notes today, but it's all about advancements in medical technology. This article is kind of an overarching what is happening and the benefits of technology in medicine. They talk about robotics, but most relevant to today's episode, they talk about nanobots, right? In this article specifically, they talk about how nanobots can find damages in the body, repair it, how it can be used to deliver payloads for a cancer, cancer treatment.

And the the thing that kind of blew my mind is this was written in 2023 and we're about to talk about research coming out of ET Zurich, which is one of the what leading institutions in the world on advanced tech. And Mouser was ahead of the game by like what three years. It's so it's so incredible to see that being where they're positioned, they can have this incredible insight. but yeah, with that said, let's get into today's topic.

Spinal cords, ETH Zurich and University of Zurich. Now, before we talk about what they've been doing, let's talk about where we stand today with spinal cord injuries and the kind of treatments that exist. So this is a topic that directly impacts me. I have a family member who had a spinal cord injury from a car accident a couple of years back. They're actively their day-to-day, they struggle with issues that arise from that. And, you know, obviously I've always been interested in like what kind of solutions there are out there and how the medical world is advancing.

We've on the podcast talked about Neurolink before, where you can use a chip in the brain and a chip near the injury and have them talk to each other almost wirelessly to send the signals through. But the fundamental problem remains the same. When there's nerve damage, those cells often do not regrow back. And there's a variety of challenges that researchers have been trying to address. For example, the leading method of treatment right now is embedding electrodes in the person's body near the site of the damage and then injecting stem cells, hoping that via those electrodes you can stimulate the cells to grow the nerve path once again.

Daniel

Let's like take a second to appreciate how crazy that is. Just the way that stem cells work in general and the fact that we figured out that you can regrow with moderate levels of success. You can regrow nerve tissue by putting stem cells there or stem like cells there and then just zapping them with electricity, simulating nearby nerves, zapping them with electricity, and then they start to grow back. That like that blew my mind.

Farbod 

It's I mean, you know if you've been listening to the podcast, you know that Daniel and I love, love Reddit. ELI five, explain like I'm five is like kind of the precursor for what the show eventually ended up being. But there's one called Nature is Metal, which is all about how nature is like in insanely cool in some ways. How cool is it that your biology can write a blank check and be like, use it however you want. It'll just kind of be whatever you need it to be to best do whatever you want to do. That that's the beauty of stem cells, right? And that's the promise that they hold. You be you, you do your own thing. However, there's downsides with that, right? Like, first of all, the whole idea of like embedding electrodes in someone's body, especially in a region that's already compromised and sensitive, not super appealing. One, it's invasive, two, it can cause more issues down the road. Then there's the problem that stem cells being the blank check that they are. Scientists don't actually have a ton of control in what they turn into.

They're kind of hoping that by stimulating them in that region, they find the gaps and help with the recovery of the nerve tissue, but that's not always guaranteed. And then lastly, the transplanted stem cells do not always survive. And if they do, and they're actually growing into that nerve tissue, they don't grow properly either. So you have these three different buckets of problems that exist with today's therapy approaches and these folks at ETH  Zurich and the University of Zurich were like, can we do anything about this? Like if we take every issue at one at a time, is is there any novel method that we can find like a loophole around all this? So that's setting the ground for what's to come. 

Daniel

And the answer what they built is is like I I have to apologize here for the awful post that I made on Twitter when I saw this saw this article, but you know, like come on man. So we they basically start with like culture medium and stem cells and then they make them into like bionic hybrids. So they they put engineered nanoparticles on these stem cells and the result of what they build, they call NPC bots. I don't even remember what the actual NPC stands for in this scenario. I'm only thinking of non playable characters and the fact that my younger sister, if ever I don't know cultural reference, calls me an NPC or she calls me a bot. And then I'm seeing NPC bots right next to each other on a diagram about a cultural medium being applied, a culture medium being applied. I'm like, I'm the culture media. You're the NPC bot

Farbod 

That's really what it's about. Taking Gen Z slurs and turning it around and being like, It's a good thing. It's a badge of honor.

Daniel

Exactly. Now now please explain to me what NPC bot actually means and what it actually does.

Farbod 

Well, for me, it was actually more interesting, and this is crazy to say because a running joke on the podcast is that I am bad at biology and I did it was the one science that did Porleon. But what was interesting for me is what happens before the MPC bot. so researchers are actually just using ordinary cells and processing them into what is called induced pluripotent stem cells. I hope I didn't butcher that. But these IPS cells are kind of like generic stem cells that are ready to be programmed from normal human cells. And then these IPS cells are developed into neural, and once again I hope I'm I don't butcher this progenitor cells. And these are the precursor cells for becoming neurons. So before even like going into the body, they're trying to set the ground by saying you cells, you guys are going to be nerve tissue. So when they go into the body there's no like gamble of what will this stem cell become versus that become?

Daniel

Yeah, do I become muscle? Do I become flesh? Do I become bone?

Farbod

Exactly. So that that's resolves like one layer of the problem. The next one is once they have these progenitor cells, they cover them in this magnoelectric nanoparticle all around its surface. And the result of that, when covered with those nanoelectric nanoparticles, is the NPC bot. There's a really interesting diagram on the linked Wevolver our article if anyone wants to check it out, but it's like a nanofluidic one centimeter squared lab, like a lab on a chip that they're doing all this work on, which I don't know, as a visual learner, I liked it. But all that to say, that's what the manufacturing flow looks like. But what the audience probably wants to know, and what I was eager to find out is how does it actually work? Well, similar to the therapies that exist today, they take these NPC bots and they inject them at the site of the injury. Now, instead of using electrodes embedded in the patient, the researchers or the doctors are able to use an externally applied magnetic field to guide these particles in place because of that nano coating that's around them. What's interesting is that there's two functions of how those nanoparticles react. One is that they can detect the signal coming in, and then two, as a result of that signal coming in, they can create an electric pulse and propel them in a certain way or as you l find out later on.

What they can do is actually stimulate the growth of these neural pathways.

Daniel

Well, it's the same mechanism as what they were doing before with the externally applied electrodes, right? It's like let's use electricity to help stimulate these cells, which now because we're using and because we're using those neurally specialized cells, those NPCs, neuroprogenitor cells, they've got this tendency to think like, Okay, I'm a nerve, I'm a nervous system cell and then you start to apply electricity to them through these magnetoelectric nanoparticles. Basically, the idea is like instead of having electrodes everywhere having to attach electrodes next to the stem cells, the researchers basically just turn the stem cells themselves into these remote control electrical stimulation units that have the particles on them to stimulate them to become nerve cells.

Farbod 

Exactly. So that now removes like pretty much all three of the buckets of issues that we talked about earlier, right? Existing issue existing approaches required those embedded electrodes. They had limited control over what the stem cells do, and the transplanted cells didn't do what you wanted them to do. So now by guiding them to where you want them to be and then stimulating them by alternating this magnetic field externally, you can pretty much increase your chances and almost guarantee that they will behave exactly the way you want them to. Which, like, in theory, all sounds perfect, right? Right? So, like, what's next? What actually happened here? Does this work? These are all the fantastic questions that we want answered. Let's talk about zebrafish. We talked about them in the past. Long story short, zebrafish are able to heal themselves. They have regressive regenerative abilities. However, they still wanted to analyze what it looks like on this subset of species because if it worked here, that could be great evidence to step it up a notch. They tested against the zebrafish, and within three days after causing a spinal cord injury, the zebrafish were pretty much swimming normally all over again. They had completely normal behavior, and the integration of the implanted cells was almost complete. Good sign, right? Well, the next step up was mice. This  sounds sad and it in my opinion it kinda is sad, but they but I guess that's that's the cost for not testing on human beings and figuring out if something works or not. But they severed the spinal cords of a group of mice for testing. And similar to humans, and unlike zebrafish, mice do not have regenerative abilities. So after this treatment, the NPC bots that were injected were localized around the injury site successfully. The neuron differentiation increased, so they started to see the pathway regrow. Nerve cells, nerve cells started to reconnect around the damaged area. And over time, the gait, the stride length, the general exploratory behavior of these mice improved. And within four weeks, they had significant motor function improvements. So they went from being completely immobilized to walking around and exploring almost normally within four weeks.

Daniel

And not just that, like they didn't just note that they the neural tissue healed and grew back, which is obviously an amazing improvement all on its own, but they also tracked that the mice over the course of those four weeks improved their gait, improved their stride length, regained ability to have coordination in those limbs even after the spinal cord injury, and started to like naturally walk around and explore as opposed to like laying around and being sick and healing from their traumatic spinal cord injury. So the interesting part of that is like no immune reaction, no ma major negative effects. You achieve the basic job, which is like regrow the nervous tissue, but then also it seems to be functioning and improving as well.

Farbod

Yeah, so all all of that is incredibly promising, right? But I think everyone wants to know what's next. Unfortunately, there's no more insight that they've provided us past this. But one thing that was notable is that not only were they able to see like this like healthy response, the bodies of the mice that are rejected, and that sets a pretty good precedent for the next step, which would naturally be human testing, right? So without the explicit mention of that from the researching team at ETH Zurich, the hope and our best hunch is that that is what's gonna happen. And if I were to say, like, I'd love to see this happening sometime in the next year or two, if possible, you know, think about the ramifications it can have for the population impacted by this. Again, I'm biased. I have family members that suffer from this, but it could be pretty incredible.

Daniel

Yeah, for sure. And especially thinking about the alternative, right, which is slice you open to not just put these stem cells near your spinal cord or the site of the injury, right? Slice you open, but we also have to embed these electrodes and these wires into your body to be able to stimulate these sp stem cells. And by the way, it may not work. Versus higher likelihood of success, less invasive. And at the same time more scalable in terms of the specialization and manufacturing process. To me that all sounds like a win win win across the board. Obviously those effects have to be tested in humans before we can go say like, let's start shooting everyone up with NPC bots. But it's very exciting.

Farbod 

Right, let's do it. Folks, reach researchers at ETH Zurich and the University of Zurich have created microscopic robots called NPC bots that could one day help repair spinal cord injuries. They're essentially stem cells covered in special nanoparticles, and doctors could guide them toward an injury with magnetic fields, then use those same fields to create tiny electrical signals that encourage the stem cells to become neurons and reconnect damaged tissue.

In zebrafish, the injured animals were swimming almost normally within three days. But more importantly, in mice, which are more similar to humans, the completely severed spinal cord showed significant improvements in movement within four weeks. This is still animal research, but it could turn stem cell therapy into something doctors can actually guide and control from outside the body and make this kind of therapy more accessible to the masses.

24,000+ Subscribers

Stay Cutting Edge

Join thousands of innovators, engineers, and tech enthusiasts who rely on our newsletter for the latest breakthroughs in the Engineering Community.

By subscribing, you agree to ourPrivacy Policy.You can unsubscribe at any time.