High Accuracy Fluorescence Guided Focused Ion Beam Milling

This paper presents advanced fluorescence-guided cryo-FIB milling workflows that utilize distinct targeting strategies for different object sizes to overcome registration errors and enable the routine, high-precision capture of both large and small cellular structures for cryo-electron tomography.

The Gist

Imagine you are a master sculptor trying to carve a tiny, perfect statue out of a massive block of ice. Your goal is to slice off a thin, transparent sheet (a "lamella") that contains a specific, rare object hidden deep inside the block, so you can take a super-clear 3D photo of it later.

The problem is that the object you are looking for is often invisible to the naked eye and might be as small as a grain of sand or as rare as a needle in a haystack. If you guess where to cut, you might slice right past it, wasting your time and destroying the sample.

This paper introduces a new set of "GPS and laser-guided cutting tools" to solve this problem. Here is how they did it, broken down into two scenarios:

1. Finding the Big Targets (The "Map and Compass" Approach)
For objects that are relatively large (like a small house compared to a grain of sand), the team created a smart mapping system.

• The Problem: When you look through a microscope at something inside ice, the image can look "shifted" or blurry, like looking at a fish in a pond from above the water. This happens because light bends differently in ice than in air.
• The Solution: They built a system that acts like a GPS. First, they carve tiny, known markers (fiducials) into the ice. Then, they use a special mathematical rule to correct the "shift" caused by the ice. This allows them to draw a perfect line on their map and tell the cutting machine exactly where to slice, ensuring the big target ends up right in the middle of the thin sheet.

2. Finding the Tiny Targets (The "Laser Pointer" Approach)
For the really tiny or rare objects (like a single-celled organelle), the old map method wasn't precise enough.

• The Problem: These targets are so small that if you cut even a fraction of a millimeter too far, you slice right through them and destroy them.
• The Solution: They used a machine that is like a hybrid between a laser cutter and a high-powered flashlight. The machine has a built-in "flashlight" (fluorescence microscope) that makes the tiny target glow. As the machine cuts the ice, it watches the target glow in real-time. The moment the cutting blade gets close enough to start touching the glowing object, the machine stops instantly. It's like a car's automatic emergency braking system that stops the car the millisecond it sees a pedestrian, ensuring the pedestrian is safe.

The Result
By using these two methods, the team can now reliably carve out thin slices of ice that contain specific, hard-to-find structures—like centrioles (tiny cell engines) or cilia (hair-like cell parts)—that were previously too difficult to catch. They haven't just found a way to cut the ice; they've found a way to guarantee the treasure is inside the slice every time.

Technical Summary

Technical Summary: High Accuracy Fluorescence Guided Focused Ion Beam Milling

The Problem
Cryo-electron tomography (cryo-ET) represents a powerful methodology for visualizing macromolecular structures in their native cellular context. However, the widespread adoption of this technique is currently constrained by the significant challenge of reliably targeting specific structures for imaging. A critical bottleneck exists in the ability to capture small or rare objects within Focused Ion Beam (FIB)-milled lamellae. Without precise targeting, the probability of successfully imaging these specific, often scarce, cellular components remains low, limiting the scope of biological questions that can be addressed.

Methodology
To address these targeting limitations, the authors establish fluorescence-guided cryo-FIB milling workflows designed to minimize error sources and enable the routine capture of structures across a broad size spectrum. The approach is bifurcated based on the dimensions of the target:

• For Larger Objects (>500 nm): The authors developed a single-step registration-based targeting strategy. This method integrates FIB-milled fiducials with a physically grounded depth correction algorithm. This correction is essential to account for focal shifts caused by refractive index mismatches, ensuring that the milling plane aligns accurately with the target depth.
• For Smaller Targets (150–500 nm): Recognizing that static registration may be insufficient for minute structures, the team implemented a real-time fluorescence-guided milling protocol. This workflow utilizes a commercially available FIB SEM instrument equipped with an integrated cryo fluorescence microscope. This setup allows for the dynamic monitoring of the milling process, enabling the operator to precisely terminate milling immediately upon the onset of target ablation, thereby preventing the destruction of the structure of interest.

Key Contributions and Results
The primary contribution of this work is the establishment of robust workflows that overcome key sources of targeting error in cryo-FIB milling. By applying these strategies, the authors achieved consistent recovery of lamellae containing the targeted structures. The efficacy of the method was demonstrated through the successful capture of small, single-copy organelles, specifically centrioles and cilia, which represent the more challenging end of the size spectrum (150–500 nm). The integration of real-time monitoring for smaller targets and depth-corrected registration for larger ones provides a comprehensive solution for a wide range of cellular structures.

Significance
The paper claims that these workflows significantly expand the accessible target space for cryo-ET. By providing practical solutions for the reliable capture of structures that have previously been difficult to target, the work facilitates the study of a broader array of cellular architectures. The authors position these methods not as theoretical proposals, but as established, routine procedures that directly address the current bottleneck in targeting specific macromolecular structures within cells.