Summary
A bioRxiv preprint maps how cryogenic focused ion beam milling affects structural preservation at different depths. The apparent damage profile varied with the ion species, molecular reporter and lamella thickness.
A bioRxiv preprint reports that damage associated with cryogenic focused ion beam (cryo-FIB) milling varies substantially with the ion used, the molecular structure being examined and the thickness of the milled sample. The findings could help researchers decide how close to a milled surface high-resolution structural information can be interpreted in cellular cryo-electron tomography.
Cryo-FIB milling uses an ion beam to carve a thin section, or lamella, from a frozen biological specimen. The lamella can then be examined by cryo-electron tomography, which collects electron images at different angles to reconstruct structures inside the sample. Milling makes the specimen thin enough for electrons to pass through, but the process can affect material near the cut surface.
The study by researchers at St. Jude Children's Research Hospital and collaborating institutions used recombinant human apoferritin as a depth-sensitive molecular reporter. In this context, a reporter is a structure whose reconstruction can be compared at different distances from the milled surface. The authors obtained a 2.05 Å in situ reconstruction of apoferritin.
Damage depended on ion and molecular target
Under the workflows tested, apparent FIB-associated damage was most pronounced within the first 15 nanometres for xenon milling, 30 nm for gallium and 45 nm for argon. For oxygen-milled surfaces, the effect remained detectable at least 75 nm from the surface.
The measured profile also changed with the molecular target. Endogenous 70S ribosomes in the same xenon-milled tomograms showed a broader profile, spanning about 30–45 nm. By contrast, β-galactosidase showed a similarly narrow xenon-associated profile in a different specimen. These comparisons indicate that a single surface-exclusion distance may not describe how every molecular structure is preserved after milling.
The result is relevant because ribosomes are often used to assess the region near a milled surface in cellular cryo-electron tomography. The authors conclude that ribosome-based estimates may be overly conservative for some other molecular targets. That conclusion is specific to the reporters and workflows examined in the preprint; it is not a universal conversion table for all cryo-FIB experiments.
Lamella thickness also affected resolution
The researchers also compared lamellae of different thicknesses. Xenon-milled lamellae between 125 and 150 nm thick produced the highest resolution among the thickness groups examined. Both thinner and the thickest groups produced lower resolution.
This result shows that the usable structural information is shaped by two factors at once: damage associated with the milled surfaces and the overall thickness of the lamella. A lamella that is too thick may reduce the quality of the tomographic reconstruction, while making it thinner can bring more of the specimen into the surface-affected region.
The work is reported as a bioRxiv preprint rather than a peer-reviewed journal article. Its measured depth profiles should therefore be interpreted as results from the tested milling conditions and molecular reporters, with the apparent damage range expected to depend on the target and workflow.