In a mouse Achilles tendon study, samples examined after reloading showed a sharper mineralization boundary at a new position, approximately 20 µm from the original tidemark, the term used for that boundary. Mineral between the old and new boundaries retained a nanoscale imprint of the unloaded state. Three-dimensional imaging showed a sharply delineated native mineralized fibrocartilage domain in controls, diffuse unbounded mineral extending beyond the original tidemark in unloading samples, and a sharper boundary at a new position in reloaded samples.
The comparison focused on the boundary
The study examined 18 12-week-old male C57BL/6J mice. Six were standard-housing controls, six underwent 14 days of hindlimb unloading, and six then had six days of free reambulation. The comparison examined loading-related nanoscale organization of the Achilles tendon enthesis.
The researchers combined pSHG and 2PF microscopy with scanning µXRD, XRD-CT, TexTOM and holotomography. Together, these methods measured matrix organization, mineral crystallography, texture and density contrast.
The matrix signal changed across the interface
Relative to controls, the unloading group showed a strongly reduced tidemark-associated 2PF peak and lower interface contrast, along with proteoglycan depletion in unmineralized fibrocartilage. The decline in collagen order across the interface was shorter and lower in amplitude. These measures partially recovered in reloaded samples. The researchers caution that 2PF is a broad signal from several non-collagenous proteins and has little sensitivity to proteoglycans, so it cannot identify a single molecule responsible for the signal.
The diffuse mineral carried a different signature
X-ray diffraction found a diffuse mineralization zone 20–30 µm wide in previously unmineralized fibrocartilage in the unloading specimen. Compared with native mineralized fibrocartilage, the zone had a higher T-parameter, an expanded c-axis lattice parameter, less ordered texture and lower average local misalignment. The T-parameter is a nanostructural proxy influenced by both mineral-particle dimensions and the degree of mineralization, so it should not be read as a direct measurement of particle size.
A broader scan found the same local pattern
A population-level µXRD analysis used five control specimens, six unloading specimens and seven reloaded specimens, drawn from four, four and five animals respectively. The unloading profile showed a roughly 20 µm diffuse-mineral zone, elevated values for the structural measures called the T-parameter and ACS, an extended c-axis plateau and lower mineral density. In reloaded samples, bulk values shifted toward controls, but a discontinuity remained at the new tidemark. Because the subset sizes differed by modality and no pooled effect estimate was provided, the findings should be read as evidence across the tested subsets, not as one population-wide estimate.
Three-dimensional scans sharpened the picture
Three-dimensional holotomography showed the same broad pattern. Controls had a sharply delineated native mineralized fibrocartilage (MFc) domain. Samples from the unloading group showed diffuse, unbounded mineral extending beyond the original tidemark, while samples from the reloaded group showed a sharper boundary at a new position. Within specimens, the reported density contrast with native MFc was approximately -0.24 for unloading diffuse mineral and -0.15 for newly formed MFc after reloading, indicating lower density in both comparisons. These figures are relative density contrasts within specimens, not absolute densities for comparison across specimens, because phase-retrieval offsets affected the estimates.
Supplementary within-specimen comparisons used fractional anisotropy (FA), a measure of directional organization. In reloaded samples, newly formed MFc had FA of 0.45 ± 0.08, compared with 0.53 ± 0.08 in old MFc; both reported p-value columns, Welch’s t-test and KS, were <0.0001. In unloading samples, diffuse mineral and native MFc had similar mean FA values, 0.50 ± 0.13 and 0.49 ± 0.09, but their distributions differed, with KS p = 0.01 and D = 0.20. Diffuse mineral also had lower mass density, 0.82 ± 0.26 versus 1.06 ± 0.23, with both test columns below 0.0001. The authors cautioned that the three-dimensional FA comparison was not statistically powered as a standalone result.
The images leave the mechanism unresolved
At the molecular level, the identity of the tidemark-associated compartment remains unresolved. The authors say the data show co-variation between matrix signal and mineral architecture rather than action of a specific molecule.
The six-day reloading period cannot establish whether the nanoscale imprint is permanent or whether longer reloading would restore native architecture. The evidence came from male C57BL/6J mice, so it does not establish human or clinical relevance.
The study’s boundaries are clear
Animal procedures received ethics approval and were reported to comply with French animal-care guidelines. The paper reports public DOI records for its two-dimensional µXRD, three-dimensional TexTOM/XRD-CT and three-dimensional holotomography datasets, and reports availability of the TexTOM analysis package. The authors declared no competing interests.
Paper data and sources
Original title: The Achilles tendon enthesis rebuilds its mineralization front on reloading but retains a nanoscale imprint of unloading
Authors: M. L. Stammer, C. Camy, M. Frewein et al.
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-24
DOI: Not available
Original paper · Full text