Â鶹ÒùÔº

Researchers at EPFL have created the first 4D lipid atlas of vertebrate development, revealing how fats shape our bodies from embryo to organism.

We often think of embryonic development as a genetic ballet, choreographed entirely by DNA and proteins. But there's another cast member quietly shaping the scene: lipids. These fat molecules aren't just fuel; they play structural, signaling, and even patterning roles as embryos develop.

Despite advances in genomics, we still don't fully understand how metabolism is arranged in different parts of the body during development. A big part of the metabolism puzzle are lipids, which vary widely in structure and function and have been notoriously hard to map across entire organisms both in high resolution and across time.

Previous techniques only offered fragmented snapshots, but without a detailed atlas, scientists can't track where and when specific lipids appear in the developing body. This has limited our ability to understand not just basic biology, but also how metabolic disorders or congenital diseases might take root.

A team of researchers at EPFL have developed a new computational method that allowed them to build the first 4D lipid map of a vertebrate embryo—specifically, the zebrafish. "4D" refers to mapping lipids in the three dimensions of space plus the fourth dimension of time, which captures how lipid distributions change as the embryo develops. Using an innovative combination of imaging mass spectrometry and a new computational framework called uMAIA, they tracked more than 100 lipid types across space and time.

The research was led by Professors Gioele La Manno and Giovanni D'Angelo, working with the group of Andrew Oates. Their study is in Nature Methods.

The team used a technique called MALDI mass spectrometry imaging to scan "slices" of zebrafish embryos at different stages of development, allowing them to measure where different lipids are located in tissue slices. But processing this kind of data is not easy, as each zebrafish embryo generates huge amounts of spectrometry data, making an impossible puzzle.

To handle the data, the team developed uMAIA ("unified Mass Imaging Analyzer"), a powerful algorithm that extracts, aligns, and normalizes the data into coherent, accurate maps. It basically, turns a pile of noisy data into a clear movie of metabolic development. uMAIA applies adaptive image extraction, matching similar molecules across sections, and correcting technical noise. What emerges is a detailed, high-resolution atlas of how lipid distribution changes from early embryo to full-fledged fish.

The team found that lipids form highly organized patterns that match anatomical structures. For example, certain sphingolipids—which are important for cell membranes and signaling—accumulated in the swim bladder, a fish organ that is analogous to human lungs. Others concentrated in developing brain regions or bone-forming areas. These spatial patterns suggest lipids play key roles in shaping organ function and identity.

Knowing where and when lipids appear can help researchers understand developmental diseases, like congenital metabolic disorders. It could also inform regenerative medicine or tissue engineering. And because lipid metabolism is often disrupted in diseases like cancer or Alzheimer's, this atlas offers a baseline for comparison.

"From this effort emerges not only a powerful resource but a Swiss army knife for doing this kind of mapping again and again across other systems in health and disease," says Prof. La Manno.