Freeze Substitution

Introduction

Freeze-substitution involves the replacement of ice in a frozen specimen with an organic solvent at a higher temperature than that at which the specimen was frozen. Fixation continues to occur at a low temperature during the substitution process, resulting in less extraction. The solvent is then replaced by a resin for specimen embedding.

In general, ultrastructure preservation is improved by slow dehydration and the low temperature embedding. Depending upon your choice of resin, the sample can then be sectioned for morphology in epoxy resin, or immunolabeled as easily as cryosections after embedding in Lowicryl, LR White or other specially formulated resin.

The substitution is usually carried out at -80°C to -95°C, and can be performed on either aldehyde fixed specimens or on rapidly frozen, unfixed material. Any warming of the specimen or contamination by frost condensation before, during, or after transfer of frozen specimens into the substitution medium should be avoided.

Aldehyde-fixed material is first cryoprotected by infiltration in 2.1M to 2.3M sucrose, mounted onto metal specimen pins and frozen by immersion in liquid nitrogen. Rapidly frozen material can be prepared either by jet, immersion or slam freezing, taking care that vitrification occurs.

Many solvents have been used as a freeze substitution medium and are covered in the referecnces below. However, we have found that methanol has two advantages which makes it a simple, easily used medium for novices.

First, it is better capable of dissolving ice at low temperatures (supposedly it takes only a few hours).
Second, methanol can tolerate a higher water content (up to 10%) than others (in particular acetone, which can only accomodate only 1% water).

The addition 0.5% to 1% uranyl acetate to the methanol also improves cell structure, and seems not to affect immunolabeling. Other additives are also possible (glutaraldehyde, osmium, etc.) and should be tested for each different cell type or tissue.

Go here to see an example of a sectioned pancreas, freeze substituted in methanol containing 1% osmium tetroxide, embedded in HM23 and immunolabeled.

Freeze substituted specimens can be embedded in any resin. However, the hydrophobic Lowicryl resins (HM20 and HM23) cut almost as easily as Epon yet can be easily immunolabeled. HM23 will polymerize at lower temperatures (down to roughly -90°C) using ultra-violet light but takes up to 6 days to polymerize. The HM20 will polymerize at -60°C in a few days.

These resins are methacrylate based and should be handled very carefully. Full handling instructions and safety sheets are provided by the manufacturers.

Go here for details on simple freeze substitution protocol that does not require the use of specialized equipment and that can be easily performed in any laboratory.

References

Van Genderen et al. (1991)

Subcellular Distribution of Forssman glycolipid in epithelial MDCK cells by immuno-electron-microscopy after freeze-substitution. J. Cell Biol. 115:1009-1020.

This is a good demonstration of the technique as applied to MDCK cells. One of the interesting elements is the remarkable labeling of lipid the authors achieved. The resin embedding held the antigen in place whereas the same antigen in cryosections either redistributed during the antibody labeing steps or was washed away.

M.A. Hayat (ed) Principles and Techniques of Electron Microscopy, 3rd Edition. CRC Press. (1989)

Freeze substitution methods and theory are covered in pages 395-398.

A.J. Verkleij, and J.L.M. Leunissen (eds) Immuno-Gold Labeling in Cell Biology. CRC Press. (1989).

See chapter 7 by B Humbel and H Schwarz, which covers freeze substitution for immunochemistry.

Finally, do not forget to read through the various publications and pamphlets written about and provided with the resins. They have practical information on safe handling and use of the resins. Do not hesitate to contact the suppliers for help with using these resins. They can provide helpful tips and information.