A lipid map of the mouse brain reveals 539 hidden chemical zones

Scientists in Switzerland have produced the first comprehensive chemical map of lipids across the mouse brain, dividing the organ into 539 distinct territories that do not always match the boundaries drawn by a century of anatomy and genetics. The study, published this week in Nature, comes from a team at the École Polytechnique Fédérale de Lausanne (EPFL) and adds a dimension largely missing from existing brain atlases, most of which are built from gene activity or cell shape rather than the fatty molecules that make up cell membranes and insulate nerve fibres.

The researchers call the new reference work the Lipid Brain Atlas, and they have made it freely searchable online. It is intended as a resource other labs can use immediately, in the way earlier gene-expression atlases have been used for two decades.

A map built from seven million data points

To build the atlas, the team used mass spectrometry imaging, a technique that fires a laser across a thin slice of tissue and records, point by point, which molecules are present at each spot. Applied to 109 brain slices taken from 11 mice, the method measured the distribution of 172 different lipids, generating roughly seven million individual measurements. New machine-learning software, built specifically for the project, was needed to stitch those flat, two-dimensional readings into a single three-dimensional model of the whole brain.

Sorting that data produced 539 chemically distinct regions, which the authors named "lipizones." Many line up with structures neuroscientists already recognise — particular nuclei, layers of the cortex, specific cell types. Others do not. The atlas picked up previously unmapped chemical boundaries inside the choroid plexus, the tissue that produces cerebrospinal fluid, and in the linings of the brain's internal cavities. According to the coverage of the findings, the lipid signatures were distinctive enough that the researchers could work out, from chemistry alone, roughly where in the brain an unlabelled tissue sample had come from.

Why lipids, and not just genes

Most modern brain atlases are organised around gene expression: which genes are switched on in which cells, and where those cells sit. That approach has been powerful, but it has a blind spot. A single gene can be active in cells that look and behave very differently depending on where their fibres reach, and genetic profiles alone often cannot capture the physical, chemical relationship between a cell body and the distant brain region its long axon connects to.

Lipids turn out to pick up some of that signal. Giovanni D'Angelo, one of the study's senior authors, said the molecules effectively mark connected but physically separate regions with a shared chemical label.

"In effect, lipids work like postal codes, giving distant but related parts of the brain the same chemical address," said Giovanni D'Angelo, senior author of the study at EPFL.

One of the clearer illustrations of this came from white matter — the brain's wiring, made largely of fatty myelin sheaths wrapped around nerve fibres, and long treated by researchers as a relatively uniform substance compared with the more structurally varied grey matter. The lipid data told a different story. First author Luca Fusar Bassini said the chemical patchwork inside white matter suggests the cells that build myelin are more varied than previously assumed.

"This shows that the cells which make myelin are far more varied in their chemistry than their genes alone would suggest," Fusar Bassini said.

What the atlas is, and is not, telling researchers

The work is a mouse study, and its authors are careful about what it does and does not establish. It does not test any disease directly. What it provides is a detailed baseline of what a healthy adult mouse brain's lipid chemistry looks like, against which abnormal patterns — in ageing, injury or neurological disease — can eventually be compared, both in mice and, if the method translates, in humans.

The team did, however, run one biological test with the new tool: tracking how the lipid map changes during pregnancy. The shifts were larger than those typically seen between male and female mice. The cortex was the most heavily reshaped region, and the concentration of galactosyl ceramide, a lipid that is a core structural component of myelin, rose sharply across much of the white matter.

Gioele La Manno, the paper's other senior author, said the pattern points to the adult brain actively adjusting its own physical wiring in response to major physiological change, rather than simply ageing or degrading over time.

• 172 lipids measured across the whole mouse brain
• 109 tissue slices drawn from 11 adult mice
• About 7 million individual chemical measurements
• 539 lipid-defined "lipizones" identified

"Together, these shifts suggest the brain retunes its own wiring to meet the demands of pregnancy," La Manno said, according to reporting on the study.

Who the atlas is for, and what comes next

The immediate audience is other neuroscience labs. Because mass spectrometry imaging and the accompanying software are both demanding and expensive to set up from scratch, a public, pre-built reference lets smaller groups compare their own tissue samples against an established lipid map rather than building one independently. The EPFL team, led by La Manno's laboratory, has also published its analysis pipeline so other researchers can apply the same methods to their own data.

Longer term, the authors frame this as a proof of concept for a method rather than a finished map of anything beyond the mouse. Fusar Bassini was explicit on that point: the mouse atlas cannot simply be overlaid onto a human brain, because anatomy and lipid chemistry differ between species, and the mapping technique itself currently requires tissue samples rather than working non-invasively in a living subject.

"What transfers to humans isn't the map, it's the method, and the fact that we now know that a healthy brain has a very organized lipid map at all," Fusar Bassini said.

That distinction matters for how the work is likely to be used. Researchers studying conditions in which lipid metabolism is already implicated — including Alzheimer's disease, demyelinating disorders, and some psychiatric conditions linked to changes in mood-regulating brain fat — now have a documented, searchable baseline to test whether similar chemical reorganisation occurs in disease models. None of those links have been demonstrated by this paper itself; they are motivations for the next round of studies, which the EPFL group and others are expected to pursue using the same imaging and computational pipeline, including attempts to apply comparable methods to post-mortem human brain tissue.

For now, the Lipid Brain Atlas stands as a reference tool, built and released publicly, with its authors framing its lasting value less in any single finding than in demonstrating that a systematic chemical map of the brain, distinct from its genetic one, is achievable at all.