$$\rightleftharpoonup{xx}$$
$$\longleftharp{xx}$$,
$$\longrightharp{xx}$$,
Cryo-electron tomography (cryo-ET) has emerged as a pivotal technique for visualizing macromolecular complexes and cellular architectures in their native, hydrated state1,2,3,4,5. By reconstructing three-dimensional volumes from tilt-series of projection images, cryo-ET bridges the gap between cellular context and near-atomic resolution, offering unprecedented insights into structural biology in situ6. While cryo‑electron tomography workflows for cellular specimens7 (including cryo‑FIB‑thinned cells in suspensions and adherent cells) are now well‑established, their direct extension to thick, architecturally intact tissues remains highly challenging8,9,10. This is especially true for high‑water‑content tissues such as brain11, where reliable vitrification and the preparation of electron‑transparent lamellae (typically <200 nm) continue to pose persistent bottlenecks12.
Achieving vitrification in tissues is challenging due to their high water content and limited heat transfer. While plunge freezing is effective for thin samples (≤20 µm)13, it fails for most tissues. High-pressure freezing (HPF) suppresses ice nucleation by applying high pressure, enabling vitrification of samples up to ~200 µm in theory. However, the efficiency of conventional HPF is often suboptimal for vitrifying tissues. To improve outcomes, recent efforts have employed strategies such as adding cryoprotectants14 and performing pre-sectioning15,16 (e.g., using a vibratome) to reduce the sample thickness mechanically prior to freezing11,17. While these approaches can enhance vitrification quality, they often entail prolonged ex vivo handling and the use of chemical additives that risk altering the native structures. To address these limitations, this study modified the “waffle” carrier strategy originally developed by Kelley et al.18 by incorporating a critical optimization: the brain tissue is physically compressed using two polished copper carriers prior to freezing, which pre-thins the sample while simultaneously securing it between a filmed grid and a copper foil. This design not only promotes rapid and uniform vitrification but also significantly enhances mechanical stability, thus effectively preventing grid breakage and sample loss during subsequent handling, establishing a robust foundation for obtaining intact, high-quality frozen tissue specimens.
Following vitrification, thinning remains a major barrier to both throughput and quality. As an early and widely adopted technique, cryo-ultramicrotomy enables numerous foundational in situ structural studies. Nevertheless, it frequently introduces compression and cutting artifacts19,20 that compromise structural integrity. In contrast, cryo-focused ion beam (cryo-FIB) milling21 offers superior precision and minimal distortion7,22. Initially, cryo‑FIB milling was applied directly to vitrified samples on grids to produce lamellae23; however, this in situ approach is primarily effective for thin specimens24,25 (e.g., cultured cells or peripheral tissue regions) and cannot access deeper structures within thick, bulky tissues. To overcome this depth limitation, lift-out strategies26 were developed, enabling the site-specific extraction of regions of interest (ROIs) from the interior of samples. Subsequent methodological advances, including serial lift-out27 and on-grid serialized sectioning (SOLIST)28, further enhanced throughput, lamellar stability, and precision of 3D correlative targeting in complex tissues. More recently, innovations have continued to advance the field, including the six‑sided attachment method by Tang et al.29, which enhances mechanical stabilization during lift‑out and thinning, as well as novel on‑grid milling approaches by Benjamin C. Creekmore et al.11 that enable deep‑structure access while preserving native tissue context. Despite these advances, protocols still vary in efficiency, reproducibility, and adaptability across different tissue types. A critical gap persists in the establishment of a standardized, easily replicable, and broadly applicable workflow that integrates these optimizations into a coherent pipeline for diverse thick tissues.
This gap is addressed here by synthesizing key technical details from established methodologies and refining them into a generalized, end-to-end pipeline. This protocol not only employs the enhanced HPF-waffle method for robust vitrification but also incorporates a series of optimized steps in cryo-FIB lift-out and thinning, such as systematic over-tilt compensation, standardized conductive coating, and calibrated milling sequences, that collectively enhance lamella uniformity, reduce artifacts (e.g., curtaining), and improve throughput. Unlike protocols tailored to specific cell types or model organisms, this workflow is designed with generalizability at its core, providing a versatile framework adaptable to a wide range of thick, complex tissues, including those that are relatively soft and prone to compression, beyond the demonstrated mouse hippocampus.
Demonstrated in the challenging model of the mouse hippocampus, notable for its susceptibility to ice damage and structural complexity, this integrated and generalized workflow reliably produces high-quality lamellae (120–200 nm) and supports efficient tilt-series acquisition (5–8 min/series). This consolidation of best practices into a coherent, accessible, and widely applicable protocol provides a reproducible framework that lowers the technical barrier to implementing cryo-ET in thick tissue research across diverse biological and biomedical contexts.