Rare spindle-shaped neurons from deep inside the brain recorded for the first time

Losing the mysterious cells may lead to Alzheimer's, schizophrenia, or other neurological disorders

Burcin Ikiz


About five years ago, researchers from the Allen Institute for Brain Science in Seattle received a special donation: a piece of a live, rare brain tissue. It came from a very deep part of the brain neuroscientists usually can't access. The donated tissue contained a rare and mysterious type of brain cells called von Economo neurons (VENs) that are thought to be linked to social intelligence and several neurological diseases. 

The tissue was a byproduct of a surgery to remove a brain tumor from a patient in her 60s. The location of the tissue turned out to be in one of the deepest layers of the frontoinsular cortex, which is one of the few places where these rare neurons are found in the human brain. “This was one of the extremely rare chances that we received this tissue from a donor that had a tumor being removed from quite a deep [brain] structure,” said Rebecca Hodge, who is the co-first author of the study, published in Nature Communications on March 3rd. Hodge and her colleagues became the first scientists to record electrical spikes from these neurons. Further studies they did on these cells gave them clues about the VENs’ identity and function in the human brain.

Constantin von Economo


VENs are large, spindle-shaped neurons. They were first identified by the Ukrainian scientist Vladimir Betz more than a century ago. They were later named after the anatomist Constantin von Economo, who described their shape and distribution through the human cortex. Only humans and especially social animals with large brains, such as great apes, whales, dolphins, and elephants have VENs. It is hypothesized that the cells evolved independently in these animals. Since common lab animals with smaller brains, like mice and rats, don't have VENs, it is difficult to study them in a lab environment.

Past studies linked VENs to social engagement and cognitive health. An analysis of "SuperAger" brains from older people who don’t suffer from the memory loss of "normal" aging showed a greater number of VENs compared to their cognitively average-for-their-age peers. Loss of VENs, on the other hand, has been observed in brains of patients suffering from a neurodegenerative disease called behavioral variant frontotemporal dementia, as well as from several other neurological disorders, including schizophrenia, autism, and possibly Alzheimer’s disease. None of these studies, however, offered clues about VENs’ exact function or unique properties. Unraveling the mystery of these neurons can help find therapies for these disorders.

The team at the Allen Institute tackled this mystery with two parallel studies: The first aimed at understanding VENs’ electrical properties, while the second one focused on their genetic identity. They didn't go exactly as planned.

For the first study, neuroscientists Brian Kalmbach and Jonathan Ting, from the Allen Institute decided to capture VENs’ electrical activity using method called patch clamp. Patch clamp is a very delicate technique where a scientist carefully punctures a cell with a very thin piece of glass to record its electrical activity.

The scientists were experienced with this technique, but the VENs they received were much more fragile than what they were used to. Even a gentle touch would be enough to make them explode. In the end, the scientists were able to record from only three neurons. The neurons showed unique electrical properties compared to other neuron types. Though the sample size was small, it was still the first ever recording of VENs, and the data was promising.

spindle neurons

Spindle neurons


The second study provided more answers. Hodge and colleagues wanted to genetically identify VENs using a new genome sequencing technique they were trying to develop at the time called single nucleus RNA-sequencing. “We had the goal of turning that technique into kind of a big data pipeline,” Hodge said. "but we didn't have any methods for doing that in humans yet." 

The question was: how do VENs genetically differ from the other neurons in the same region?

One of the challenges with this study was how sparse VENs are in the human brain. They only account for a very small fraction, about 1.25 percent, of all neurons in the frontoinsular cortex, and about 500,000 neurons brain-wide. The group was only able to capture data from a handful of them. “So that's sort of a big challenge of working in human brain where you have you don't have nice transgenic tools like you have in mouse,” Hodge said. "You just have to see what you can get using the tools that you have available."

From the gene sequencing analysis the group was able to identify new marker genes for VENs that could be used to differentiate them from other neurons besides looking at their unique shape (which isn't always easy due to their rarity). But, the data set they got from the study was also too small to interpret the results accurately.

To tackle this problem, they compared these human cells to cell types that were already defined in mouse tissue to see if any of them matched using a computational mapping technique. Jeremy Miller, the other co-first author of the paper and the bioinformatician that performed the analysis, called the technique “a computational advance” that allowed them to predict what kind of class of cells VENs belonged to. The finding was surprising. Given the unique shape of VENs, the scientists expected them not to match with any other cell types, but they did. VENs’ genetic signatures looked very similar to those of neurons that send their axons from the cortex to deeper regions of the brain called extratelencephalic-projecting (ET) excitatory neurons.

Miller thought this finding could have interesting implications for neurodegenerative diseases, such as the behavioral variant frontotemporal dementia, where VENs are thought to be selectively vulnerable. Future studies might look to see whether it is all of the ET-type neurons that are lost in disease or whether it's only VENs. This information would help with deciding which cell types to target for therapies. The group is currently repeating the study using new sequencing methods that allow them to get tens of thousands to millions of cells in order to better understand the genetic properties of these neurons.

Comment Peer Commentary

We ask other scientists from our Consortium to respond to articles with commentary from their expert perspective.

Sahana Sitaraman

Neuroscience and Behavior

National Centre for Biological Sciences, India

This is the first I’m hearing of these neurons and your article has made me want to find out more about them. Given how difficult and rare it is for scientists to get a sample of VENs, did they try to culture them to get a sustainable source with higher numbers? I’m also curious about the similarity in the transcriptome of human VENs and mice ETs. Have there been studies to see if mice ETs have a subset which act as their VENs? Did the authors also compare their electrophysiology to see if they behaved similarly? That would open up a ton of possibilities. 

Burcin Ikiz responds:

To my knowledge, VENs have never been successfully cultured and there are no existing protocols on them. My guess that it is because they are very rare in humans and do not exist in rodent models. Otherwise, I agree, that culturing would serve as a great source to get higher yield of cells to study them further.

Mice ETs have similar transcriptomic profile as human VENs. They are also similar in the way they project deep into the brain’s cortical structures. However, I don’t think the authors have compared their electrophysiological properties to test how similar they are functionally to one another. There are still a lot of questions to be answered about these mysterious neurons. I think that is partially why they are so fascinating!

Marnie Willman


University of Manitoba Bannatyne and National Microbiology Laboratory

This reminds me how far neuroscience has come as a field. From large images of the brain and autopsies being the only information available only 100 years ago, to now noticing these small subsets of specific cells that likely have some contribution to devastating diseases like Alzheimer’s and schizophrenia. I’m looking forward to seeing when we learn next about VENs… It seems to be a relatively novel frontier.

I may have missed this in the article, but can you study VENs when the subject is still alive? Or are they buried so deeply in the brain that you have to dissect brain tissue in order to reach them? That image of a VEN from Twitter was very interesting, and made me wonder how they captured it.

Burcin Ikiz responds:

Neuroscience has indeed come a long way in the last century - we now know so much about how the brain functions, what it is made of and how neurons communicate with one another. That being said, there is still so much to learn about the brain, especially in terms of how a neurological/psychiatric disease progresses and how much the brain is connected to the body.

You asked a great question! While the technology for in vivo brain imaging in animals and humans is progressing relatively well, scientists are still not able to record from such deep brain structures at least in humans. That is why studies like this, where scientists were able to  receive a live donated tissue from a brain tumor patient, are so valuable. To capture the VEN in the image from Twitter, the scientists kept the donated tissue alive in a dish for just enough time to record it.

Lauri Elsilä

Neuroscience and Psychopharmacology

University of Helsinki

I usually shun the idea of a certain population of neurons being responsible for some form of behavior, like you shortly mentioned in the story citing and interview of John Allman, who has postulated a very specific role for these cells. Usually this is an over-simplified view and grossly ignores the complexity of the brain’s connectivity and dynamics. This said, with a unique cell-type like this, with such a restricted localization in rather small cortical structures, such a pivotal role could actually be possible: if the VENs constitute a bulk of projections to certain specific brain regions, their part in producing certain behaviors might be surprisingly large.

I assume that analogous neurons can be found, for example, in rats, but there must be biological reasons that caused the VENs to evolve in the first place, something that makes a difference between them and their analogues in other mammal species. And to know all this, naturally, the kind of research you showcased here is essential.

Burcin Ikiz responds:

Yes, I agree with you wholeheartedly! I also think that the notion that a group of neurons or even a specific part of the brain is responsible for only certain functions/behaviors (like the functional imaging studies on human subjects used to conclude) is very limiting and over-simplifies how our brains work. Today, we know that neurons make connections with each other all throughout the brain, as well as with other body parts, such as the gut, that play a role in our behaviors and even in neurodegenerative diseases like Parkinson’s disease. That being said, since VENs are so rare and are very specifically localized, I believe that their role in complex brain functions such as emotional intelligence is highly plausible. I am curious to see how the research on these neurons will progress and what it will discover.

Simon Spichak


This is absolutely fascinating!  

Do we know what kind of neurotransmitters these neurons produce? That might tell us something about what they do! What I find fascinating with diseases and disorders of the brain are the sex differences. Perhaps these neurons might be related to why we might see sex differences in humans but not in certain animal models of disease.

I would imagine these cells are quite hard to isolate. It’s  incredible that the researchers managed to capture them. It would be really cool to find out if there are specialized astrocytes or glia that support these VENs!

Burcin Ikiz responds:

I don’t think the researchers know which neurotransmitters VENs produce yet, but in this study they predict that VENs are a form of extratelencephalic-projecting excitatory neurons. So, they most likely  release excitatory neurotransmitters. I agree that knowing the exact neurotransmitter could give us a clue about what they do.

What an interesting theory about VEN’s relation to sex differences in diseases. It would be very exciting to pursue this question. I agree that it would be very valuable to learn whether there are certain types of glial cells that support them. Given their size and function, I would assume that there are several types of cells that  provide support to them.

Kathryn Vaillancourt


McGill University

I’ve been calling myself a neuroscientist for almost ten years and I’d never heard about VENs until now!  

As someone who works with human brain tissue (in my case, post-mortem samples from a brain bank) I want to emphasize how rare it is to have access to surgical samples at all, let alone from a rare type of cells. Also, it’s interesting how the researchers used single-cell RNA-seq (scRNAseq) in a different way than we usually see in brain research.

Typically, I’ve seen researchers use scRNAseq on tissues to identify which cells are present, and in what proportion, in a specific brain region but this study did the inverse. They used it to understand how the VENs might be related to other types of brain cells that we already know about.

This does make me wonder, though… we know that cells can change which genes they express as they go about “behaving” and responding to their  environments, and the cells in this study were taken from tissue next to a brain tumor. How confident are we that the genes they express are  representative of their normal, healthy functioning and not in response to the tumor?

Burcin Ikiz responds:

I have also never heard of VENs before until I read the paper. That is why I found the study so interesting and wanted to write about it.

Your concern about the cells behaving differently due to their proximity to the tumor is a very valid one. The group used the surgical sample for electrophysiology and used postmortem tissue for the genetic studies. So, their gene expression results were not affected by the existence of a tumor. Either way, I am glad you brought up this  important point, because it is definitely something to keep in mind for future studies using surgical samples.