Brains beyond the wiring diagram

In the classic cyberpunk novel Neuromancer by William Gibson, characters can save their consciousness to a hard drive so that their personalities will live on inside a machine. When the novel was published in 1984, it may have felt far fetched that mapping a brain could allow computers to capture the personality within. But stories such as Gibson’s borrow from a real tenet of neuroscience: that by understanding how a brain is put together, researchers can understand and predict its activity patterns.

To work out how brains handle complex information, neuroscientists have put tremendous effort into mapping connections between cells and between brain regions. They have even completed wiring diagrams that show how each cell connects with others for simple brains of animals such as nematodes.

But researchers are starting to realize that these wiring diagrams, or connectomes, are incomplete. The brain isn’t just a bundle of anatomical connections. Researchers are increasingly observing the importance of diffuse chemical signals that float across tissues and flout the rules of those wiring diagrams. The next connectomes will be wireless.

Mapping and modeling connectome complexity

One of the first things a student of neuroscience learns is that most communication between brain cells known as neurons happens at synapses, small gaps where neurons almost, but don’t quite, touch. Information from one neuron passes to the next via the release of small chemicals called neurotransmitters into the synapse.

The latest estimates suggest that the human brain has 88 billion neurons (Proc. Natl. Acad. Sci. U.S.A. 2023, DOI: 10.1073/pnas.2303077120). As each neuron can make many synaptic connections, simple arithmetic helps explain why mapping a human connectome is fiendishly complex. Just to image a whole human brain at the scale needed to render every synapse would generate about a zettabyte of data—or about one-fifth as much data as the whole world generated on average each month in 2020, according to data company Statista.

To break the problem down to size, many researchers have turned to simpler brains, and they are making headway with model organisms. Collaborations like the Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative have invested in new tools and techniques to understand how neurons connect within different animal brains and within distinct parts of the human brain. Brain atlases, says John Ngai, director of the BRAIN Initiative at the US National Institutes of Health, provide important ground truth that researchers can refer to in studies of how brains direct behaviors. Meanwhile, in computational labs, researchers have taken information about those connections to simulate brain activity in computer models.

But recently, researchers who study the very smallest brains are realizing that they need to factor in other types of communication too.

Upgrading to wireless

New insights into how messages get communicated through the brain have unfolded in the community that studies the tiny nematode Caenorhabditis elegans, which has just over 300 neurons. “The worm is this great playground,” says Andrew Leifer, a systems neuroscientist at Princeton University. “It was the first organism to have a connectome, and it’s still the most mature and complete connectome.”

In Leifer’s lab, researchers study C. elegans to understand how brain activity generates behavior. They recently probed how activity propagates in living worms’ brains by stimulating individual neurons and monitoring which other neurons responded (Nature 2023, DOI: 10.1038/s41586-023-06683-4).

To do this work, the researchers rely on chemical tools that have become important in neuroscience: an optogenetic protein that can activate a neuron when exposed to light at a certain wavelength and a calcium sensor that lights up when neuronal activation occurs. In the study published last year, the team added this two-part system to all the neurons in the worm and then analyzed the strength of pairwise interactions between 186 neurons.

When the researchers fed their results into a computational model of the worm’s brain connections, they found that the wired connectome could not account for the activity they saw in a living brain. Instead, they observed that one neuron’s action often stimulated another neuron that was not connected to the first by synapses—nor even by a chain of several connecting cells.

“I was surprised at how challenging it was to go from anatomy to predict the kinds of measurements we were getting,” Leifer says.

The researchers started to solve the puzzle when they realized they could remove or reduce the links between unconnected neurons by blocking the release of molecules called neuropeptides.

Neuropeptides generally don’t act at synapses. While the molecules are stored and released from the same part of the neuron, they can diffuse much farther than neurotransmitters do, and they are recognized outside of synapses in the distant cells that receive their signal.

According to the BRAIN Initiative’s Ngai, the importance of these molecules has been known for decades. But because they act at a distance and are recognized by G protein–coupled receptors (GPCRs) that amplify signals, neuropeptides are somewhat difficult to study. “You can see effective transmission with a neuropeptide at very, very low concentration,” he says.

Unlike classical neurotransmitters, which mediate rapid-fire information transmission, neuropeptides have a reputation for being slow modulators that tweak the strength of synaptic connections. For this reason, Leifer says he was surprised to see that in C. elegans, some neuropeptides weren’t moving slowly at all. Some of them acted within half a second, the shortest time scale the researchers could measure.“In a small brain, like the worm [brain], where things can diffuse quickly,” Leifer says, “peptides could potentially be serving a role similar to classical neurotransmitters.”

Matching signals to receptors

Neuroscientists are starting to get excited about the idea of a wireless connectome, and researchers in Leifer’s lab aren’t the only ones trying to map how neurons can connect without making contact. But to map a signaling network without direct connections, researchers need to know which cells are sending signals and which are receiving them. That requires matching peptide signals to receptors.

Neuropeptides outnumber the handful of neurotransmitters that act at synapses by at least an order of magnitude. That has slowed research into how neuropeptide signaling works.

C. elegans has roughly 150 GPCRs that researchers predict will recognize peptides. Until recently, around 80% of those GCPRs were orphan receptors without known ligands, says Isabel Beets, a neurobiologist at KU Leuven. Meanwhile, over 100 genes encode signaling peptides, many of which can produce more than one signaling molecule.

To try to resolve which peptides pair with which receptors, Beets’s team designed a reporter system that would detect GPCR activation for each of the receptors that the worm brain expresses. Then they systematically screened a library of C. elegans GPCRs against 344 nematode peptides and found over 450 peptide-GPCR interactions (Cell Rep. 2023, DOI: 10.1016/j.celrep.2023.113058).

Unexpectedly, Beets and her lab team observed that “it’s not always a one-to-one interaction,” she says. Instead, “many peptides can sometimes activate the same receptor, or the same receptor can have very different peptide ligands. There’s a lot of complexity in this biochemical language that the brain seems to use.”

Still, the study built a dataset for modeling neuropeptide signaling across the brain. It has proved invaluable for computational researchers trying to fit long-distance signaling into existing models of the brain.

Wireless diagram

Beets’s team worked with computational neuroscientists to produce the third in a triumvirate of papers published at the end of 2023 that underline neuropeptides’ importance. In that paper, researchers in William Schafer’s lab at the Medical Research Council Laboratory of Molecular Biology sketched out a connectome’s neuropeptide equivalent: a network map of how neurons across the worm’s brain might interact using neuropeptides (Neuron, DOI: 10.1016/j.neuron.2023.09.043).

To create this wireless connectome, the scientists combined the receptor-ligand pair data from Beets’s group with a single-cell transcriptomic dataset that describes which cells express which GPCRs and neuropeptides—and where those cells are in the worm’s brain.

According to the BRAIN Initiative’s Ngai, the team’s findings make a major advance in thinking about how neuropeptide signaling fits into and overlays anatomical circuits. He’s particularly impressed by the team’s approach for “inferring who is talking to whom via neuropeptide signaling,” he says.

With the whole wireless connectome in hand, Schafer says, his lab began to study how neuropeptide signaling circuits are organized. “The neuromodulators form a network that is equally—or actually, probably more complex than the synaptic network,” Schafer says. Compared with the more one-to-one relationship involved in synaptic signaling, he says, neuropeptide signaling involves “a lot more cases where each neuron is getting direct signals from lots of different other neurons.”

But many questions are still unanswered. Schafer’s team had to make their best guesses about a few variables that are not yet known, including how far they might expect a given peptide to diffuse and how strongly a peptide must activate a receptor to count as connecting two cells. New tools might help replace these estimates with data. One is a new generation of fluorescent probes that can detect neuropeptide signaling and help researchers uncover how that signaling works.

In a recent study, researchers reported on a generalizable, fast fluorescent method to track neuropeptide receptor activation in living brains and spot both when a ligand turns a receptor on and when it turns it off. According to Yulong Li, whose group at Peking University developed these sensors, they will be useful in studying what stimulates neuropeptide release, how the molecules spread across the brain, and what physiological stimuli they respond to.

And these and other new tools mean that work will soon expand beyond simple worms. “We’re at this kind of unique moment where new connectomes are becoming available for the first time,” Leifer says. That allows researchers to test how signals flow through each organism’s connectome and to ask new questions about how neuropeptide signaling works in concert with the better-understood synaptic connections.

It’s not clear how well these insights from nematodes will extend to brains such as our own, which are larger and have evolved in different contexts. But researchers say that decades of past neuroscience research have shown that principles observed in worms often repeat themselves in other branches of the evolutionary tree. Meanwhile, there are still new questions to explore in the small brain of the worm. “Somehow, each neuron has to decode these combinatorial signals that are coming through these different chemical messengers,” Schafer says. Decoding the new signaling diagram, he says, is an exciting prospect. “In terms of how this code works, that’s the next big question.”