When a heart contracts, its wall does more than pull in one direction. Muscle cells shorten while the wall becomes thicker and the blood-filled cavity gets smaller. Understanding that movement requires knowing how the cells are arranged in three dimensions. A slice through the tissue can reveal many cells, yet still conceal where their groups branch or join beyond the cut surface.
Robert S. Stephenson, Peter Agger and colleagues addressed that problem using high-resolution X-ray imaging. Their 2026 study in Scientific Reports compared heart tissue from a human, a sei whale, a giraffe, an Asian elephant, a pig and a rabbit. Across those very different mammals, the researchers found a shared principle: a connected, branching muscle mesh with several levels of organisation.
Following a structure beyond a single slice
The team examined one tissue sample from each species. Each crossed the full thickness of the left-ventricular wall, the muscular chamber wall that drives blood into the body’s circulation. The samples came from hearts obtained after death; the study did not image living hearts beating.
X-ray microtomography recorded the tissue from many directions and reconstructed a three-dimensional volume. Contrast treatment made the internal structures easier to distinguish. Researchers could then follow a group of muscle cells through successive virtual slices, rather than infer its shape from one two-dimensional section.
Their task was not simply to colour anything that looked like a separate piece. They used a consistent rule: trace an aggregate along its length, include its main branches and stop when it joined a larger component. This provided a practical way to identify comparable parts within tissue that was continuously connected.
That rule also explains why these reconstructed objects should not be imagined as separate muscles waiting to be pulled apart. Their boundaries mark a method for studying the network, while their branches keep them integrated with their neighbours.
Similar organisation, varied shapes
In every sample, individual heart-muscle cells formed chains. Connective tissue grouped those chains into aggregates, and spaces between aggregates provided possible surfaces along which neighbouring groups could move slightly relative to one another.
The aggregates were far from uniform. Those in the giraffe and rabbit samples tended to look flatter; those in the elephant and whale samples looked more tubular. Even within one piece of tissue, neighbouring groups differed in dimensions and orientation. Following the same group through the wall could reveal a changing shape.
This matters because anatomical language can quietly become a mechanical assumption. Calling every aggregate a “sheet” suggests a flat structure with fairly predictable boundaries. The authors preferred the more neutral term “aggregate” because the observed shapes were too varied for that simple picture.
The familiar broad change in muscle direction across the wall was still visible. But close inspection exposed abrupt local differences that a smooth, averaged description could miss. Some neighbouring aggregates pointed more than 45 degrees apart. A model can capture a general trend while overlooking substantial variation in the tissue that produces it.
A second level of branching
The scans also revealed larger bundles made from groups of aggregates. These stood out because their orientation differed sharply from the surrounding myocardium—the heart’s muscle tissue. Some angled across the wall, twisting and bending as they progressed.
They remained part of the same network. Branching connections joined them to adjacent aggregates, rather than isolating them into a separate band. The researchers observed this higher level of organisation in all six specimens.
The result helps fill a scale gap. Scientists can study individual muscle cells and can measure the overall heart’s contraction, but the arrangements linking those scales are harder to describe. The reconstructed aggregates and bundles show an intermediate architecture through which the pull of many cells can contribute to changes in the wall’s shape.
What the giraffe comparison can tell us
A giraffe is an informative member of this comparison because its heart works within a demanding high-pressure circulation. Yet the study’s main finding was shared architecture across mammals, not a special arrangement demonstrated to explain giraffe blood pressure.
There was only one sample per species, taken from one region of the heart, and image resolution varied with sample size. The observed contrast between flatter and more tubular aggregates is therefore a description of these specimens. It cannot establish how consistent that contrast is across all members of a species or throughout their hearts.
A different experimental approach shows why the distinction matters. Chang Liu and colleagues introduced giraffe-specific changes in FGFRL1 into mice and altered their response to a blood-pressure challenge. That experiment tested a genetic effect; Stephenson’s study mapped tissue structure. Neither supplies the missing causal link between that gene and the particular architecture seen here.
Building a more realistic picture of pumping
The authors propose that small movements between connected aggregates help the wall rearrange as its cavity narrows. Larger bundles angled through the wall could contribute forces in different directions. These are functional interpretations of the anatomy, rather than movements directly recorded in these preserved specimens.
The next useful comparison would connect detailed structure with movement: how do the same kinds of aggregates change orientation during contraction? Scanning larger portions of a heart at similarly fine resolution would also test whether the arrangements seen in small samples extend throughout the organ.
For now, the study gives heart researchers a firmer anatomical starting point. A giraffe’s ventricular wall, like those of the other mammals examined, is built from connected structures whose local variety matters. Understanding its pumping action requires keeping both features in view: the continuity of the mesh and the different directions in which its parts can pull.
Sources
Robert S Stephenson, John Partridge, Jonathan C Jarvis, Rajmund Mokso, Stephen Hall, Robert H Anderson, Peter Agger (2026). Three-dimensional imaging reveals a hierarchical organisation of the myocardial mesh in mammalian hearts. Scientific Reports. DOI: 10.1038/s41598-026-43337-7. Read the full paper.
