Minimizing axial elongation of structures

I have a short tip for you this month, inspired by a conversation with one of my grad students.

In our lab, we use fluorescent beads to test the performance and calibration of our system and to measuring point spread functions (PSFs) to model the behaviour of light. The beads that we use are supplied by manufacturers who use strict controls to ensure that the beads can be considered to be spherical. My student was visualising some of these beads in 3D (reconstructed from 2D sections) using Volocity, and was confused because the beads appeared ovoid and not spherical, so she asked me what was happening. What I told her applies to structures within biological samples as well as to beads and so should be taken into account when interpreting 3D data. I thought it might be useful to share with my eager readers as well...

The reasons why objects might appear as ellipses instead of spheres are different depending on the size of the objects being observed:

Axial Elongation

Working with very small structures (< 10 µm)

Normally, the z (axial) resolution of a widefield/confocal microscope is 3 - 4 times less than its xy (lateral) resolution. In other words, the smallest object that can be resolved is ~ 700 nm in z and ~ 200 nm in x. When working with small objects, this effect is significant and the objects will appear to be 3 - 4 times elongated in z.

This elongation effect, caused by the difference in z resolution, is only significant on small structures around the resolution limit of the microscope. It is a fact of optical microscopy and should be taken into consideration when interpreting data.

Working with larger structures (> 10 µm)

If you are seeing elongation when working with larger structures then the cause is different. This time, it is due to a mismatching of refractive indices, either between the immersion medium of the objective and the sample medium, or between the sample medium and the sample itself. If the refractive indices are mismatched then the physical distance moved by the focus drive is not the same as the optical distance. This gets worse as you move deeper into the sample, away from the cover slip. Optical distance is less than focus drive distance and the result is the stretching in z that you observe.

To minimize this effect, your refractive indices should be matched as closely as possible. For example, if your specimen consists mainly of water and is mounted in an aqueous medium, then it would be preferable to use a water immersion lens of a slightly lower numerical aperture (NA) than an oil immersion lens of the same magnification and higher NA.

You should also be aware that spherical aberration from high NA lenses can also effect elongation. Most microscope manufacturers have special flat-field objectives that will reduce this effect. My tip is to consider evaluating lenses for spherical aberration before accepting the final example.

My student certainly found this information useful, and I hope you did too.