“Three rarely-discussed modes of preservation retain features of organisms more lifelike than any other kinds of fossils: freezing, dehydration, and preservation in amber.” - (Grimaldi et al., 1994)
“It is manifest that flies, spiders, ants, and the like, that have accidentally been inclosed and buried in amber or even the gums of trees, never afterwards decay; though they are soft and tender bodies.” - Francis Bacon, The historie of life and death
“There’s a great future in plastics.” - Mr. McGuire in The Graduate
Embedding is the process of infiltrating biological tissue with a liquid medium that can then be solidified to provide structural support, enabling thin sectioning for microscopy. While the major purpose of embedding in electron microscopy is to allow for ultrathin sectioning and imaging, the embedding media themselves are potential options for long-term stabilization and storage of biological specimens for brain preservation.
Key considerations for embedding agents in brain preservation include:
We review several approaches, most importantly:
Overall, embedding has potential for brain preservation, but further research is needed to optimize techniques for maximizing long-term stability while minimizing damage, if someone decides to pursue this option. The major advantage of it is that it could potentially allow for very long-term structural preservation, hundreds or thousands of years, without any need for maintenance.
Some fruit-eating mutualism aside, it is not in the evolutionary interests of most plants to be eaten. As a result, plants have developed a wide set of techniques to protect themselves from herbivores, insects, and pathogens. One of these techniques is to secrete resins, which are highly viscous materials that can trap insects.
Under the right environmental conditions over long periods of time, some of these tree resins can vitrify. This process involves many chemical changes in the resin, including progressive free radical polymerization, cyclization (the formation of aromatic rings), and crosslinking (Pérez-Castañeda et al., 2014). As a result of this polymerization, amber chemically vitrifies into a glass (Pérez-Castañeda et al., 2014).
There are many different types of amber and they are created by a diverse set of trees. The chemicals responsible for the polymerization that produces amber depends on the type, but as far as I can tell, all ambers include terpenoids in their polymerization mixtures (Wolfe et al., 2009).
Terpenoids are built from the 5-carbon molecule isoprene, and the polymers that isoprene forms, which are called terpenes. Terpenoids can be classified based on the number of isoprene groups present. (Isoprene is also the main chemical constituent of natural rubber.)
The isoprene molecule in terpenoids contains alkene functional groups that can undergo free radical polymerization.
Baltic amber is the major source of amber in the world. Its chemical composition is a co-polymer of diterpenes – terpenoids with four isoprene groups – primarily communic acid and communol (Wolfe et al., 2009). Baltic amber also incorporates succinic acid, which likely acts as a cross-linking agent between the polymerized terpenoid units (Sonibare et al., 2014).
There is no clear-cut distinction for when a resin is said to have fully polymerized or fossilized, but clearly it takes many years. As the polymerization process continues, the resin becomes harder. Copal is the name of resinous substances found in an intermediate state of polymerization between the “gummy” initial resin state and the fully hardened amber. One classification system calls resins 250-5000 years old ‘ancient’, resins 5000 to 40,000 years ago ‘sub-fossil’, and older resins ‘amber’ (Stankiewicz et al., 1998).
Here is more on the chemical composition of amber and the distinction into classes:
“Resin is one of many biological substances secreted by plants (others include gum, wax, sap, latex, oil and mucilage) which polymerises over time into a sub-fossil form and then a fossil form – copal and amber, respectively (Lambert et al. 1993, 2015). The change from resin to copal to amber is a continuum, and although some authors have attempted to define the three substances based on age, chemistry or physical properties, there is no agreed demarcation between the three groups (Anderson 1996; Vavra 2009; Penney & Green 2010). In general, amber is harder and chemically more inert than either copal or resin, and copal is harder than resin but chemically very similar” - (McCoy et al., 2016)
The remnants of living organisms that can be preserved within amber are called “bioinclusions.” Bioinclusions are typically preserved plant or insect tissue. However tissue from vertebrates, such as a lizard foot in one instance, has also been found to be preserved within amber. Bioinclusions of animal and plant materials that happen to be present in the resin during the polymerization can become stuck and preserved inside of the amber for many millions of years.
The best known preservation occurs within Dominican and Baltic ambers, with Dominican amber likely the best. One source notes that “[i]nsects in Dominican amber are typically perfectly preserved, whereas Baltic amber specimens often have a milky coating on the body.” (Grimaldi et al., 1994). The features preserved by amber can include highly detailed structural morphology including clearly visible organelles such as mitochondria and endoplasmic reticulum (McCoy et al., 2016). In preserved insect flight muscle tissue, the expected sarcomeric repeat structures can be seen (i.e., Z-lines and M-lines between A-bands) (Grimaldi et al., 1994).
Brain morphology preservation in amber can also be remarkably high in some instances. For example, one study performed TEM of the protocerebrum of Proplebeia dominicana, an extinct stingless bee, preserved in Dominican amber and rehydrated prior to dehydration and embedding for TEM. The images show extensively laminated parallel membranes surrounding patches of cytoplasamic remains. These structures are suggestive of ensheathing glia cells surrounding large axons (Omoto et al., 2016).
DNA has been reported to be able to be extracted from biospecimens preserved in resin (Peris et al., 2020).
But if the free radical polymerization that causes the vitrification of amber takes such a long time, why do the bioinclusions not degrade in the meantime?
The answer seems to be that certain chemicals in the resinous substance that eventually forms amber can act to stabilize the structure of the organism and prevent the typical degradation processes.
One study used pyrolysis-gas chromatography/mass spectrometry on the insect tissue preserved in Dominican amber (Stankiewicz et al., 1998). They found that there were no traces of complete protein or chitin biomolecules, but there were abundant straight chain hydrocarbons that were likely derived from the insect tissue during the polymerization process. There was also large amount of bicyclic terpenoid structures in the insect tissue, even the internal tissue that was not in direct contact with the amber.
This suggests that the terpenoids in amber-forming resins do not just encase the bioinclusions but actually enter the tissue and form types of chemical crosslinks to stabilize biomolecules before the full polymerization of the amber is completed. The authors suggested that the crosslinks could could be formed by the sulfur or alkene groups of the terpenoids in the resin. This would explain how the fine detailed morphology of the internal tissues of the insects could be preserved.
So there are seems to be two steps for the preservation of bioinclusions in amber:
Overall, what amber shows us is that it is possible to preserve detailed biological structures for very long periods of time via chemical vitrification even at ambient temperature, if the right chemicals are used.
Google defines embedding as to fix an object firmly and deeply in a surrounding mass. Within biology, it is typically used to refer to the process of encapsulating tissue elements in a supporting medium (Glauert et al., 2014, ch. 5).
An alternative definition is to make something an integrated part of a whole, which is closer to what is performed during embedding of a biospecimen for electron microscopy (Echlin, 2009). The embedding media is typically a liquid. It infiltrates a permeable biospecimen in its liquid form before hardening to create a solid matrix. Often but not always, the embedding media is made up of a liquid polymer which can solidify by polymerization.
Sometimes “embedding” is used to describe surrounding the surface of a biospecimen with a support media without actually having the media enter the biospecimen and polymerize inside. By tightly surrounding the tissue, the tissue is prevented from moving during subsequent processing steps (Heinsen et al., 2000). This procedure is also sometimes called “mounting” or “simple embedding” (Toga et al., 2002).
Whereas true infiltration embedding requires the medium to diffuse into the tissue and interface with the tissue biomolecules. It is infiltration embedding that we are interested in for brain preservation. Notably, a brain hemisphere can take several weeks to be completely infiltrated with an embedding resin (Toga et al., 2002).
As a very basic brief history of embedding media, paraffin was the first embedding media used in histology and then resins were invented for TEM (Echlin, 2009):
Embedding for biological samples had its origins in light microscopy, in which warm paraffin wax was infiltrated into stabilized and stained specimens that were then sectioned as thin as 1–2 μm for histological and structural examination. Resins were developed for use in the same way for the TEM to enable sections as thin as 30 nm to be cut from suitably prepared material.
The major purpose of embedding in microscopy is to surround tissue in a structural support medium so that it can be sectioned and imaged under the electron microscope (Yeung et al., 2015a). From a brain preservation perspective, we are not interested in sectioning or imaging the tissue – at least not nearly as interested, although that could theoretically be a part of quality control metrics and even potential future revival procedures. However, the substances used as the support media in embedding procedures are potential options for long-term storage of the brain and/or body. As a result, embedding and the various embedding media become relevant to brain preservation.
One article describes the major goals of embedding for electron microscopy as being (Mollenhauer, 1993): - Making the tissue hard (but not too hard) to facilitate sectioning - Increasing the electron contrast in the electron beam - Increasing the stability of the tissue in the electron beam - Stable storage of tissue elements without deterioration
Sometimes these goals can conflict. For example, increasing the amount of crosslinking leads to more stability in the electron beam, however this also makes sectioning more difficult as the tissue can become too hard (Mollenhauer, 1993).
Here is another description of the goals of an embedding agent for EM:
Ideal qualities of embedding medium: 1. Easily available 2. Uniformity from one batch to another, no lot to lot variation 3. Solubility in dehydrating agents 4. Low viscosity as monomer for penetration 5. Uniform polymerization 6. Little volume change during polymerization 7. Good preservation of fine structure 8. Good sectioning quality that includes homogeneity, hardness, plasticity and elasticity 9. Resistance to heat generated by sectioning 10. Adequate specimen stainability 11. Stability in electron beam 12. Electron lucent
Here is another list (Echlin, 2009):
There are a large number of chemicals that may be used as embedding agents, and their suitability for a given sample must meet the following criteria: 1. The embedding material must have a sufficiently low viscosity to enable all exterior parts and, where appropriate, all interior parts of the sample to be infiltrated. 2. The embedding material must form a firm attachment to the sample. 3. The embedding material must preserve the fine structure of the sample. 4.The embedding material must retain the chemical identity of the sample. 5.The polymerized material must be stable in the electron beam. 6.The final embedded material must show no shrinkage and be sufficiently firm to allow post-embedding processes to be carried out. 7.Ideally, the embedding chemicals should be of consistent quality, inexpensive, and readily available. 8. The embedding process should not take a long time to complete.
Of these goals, sectioning quality, stainability, and electron lucency are not as important in brain preservation. On the other hand, uniform polymerization, little volume change during polymerization, and especially good preservation of fine structure clearly are important.
Another major difference between embedding for electron microscopy and the goals of brain preservation is that typically the biospecimens prepared for electron microscopy are extremely small. In a review of embedding methods, Kent McDonald notes that “it is common sense that pieces that are small will infiltrate quicker than larger pieces and we recommend cutting samples as much as possible before using these rapid resin infiltration techniques” (McDonald, 2014). Infilitration of larger biospecimens with the embedding agents described here can be quite difficult. Scaling from small samples to the whole brain or even the whole body this is a major barrier in directly applying these techniques to brain preservation.
A key exception to this is embedding via plastination, which is typically done on macroscopic-scale biospecimens. In plastination the opposite problem is more of an issue: while macroscopic-scale preservation is clearly possible, it is unclear how well the tissue is preserved on a microanatomic scale.
Allowing the tissue to be easily sectioned is probably the major goal of most embedding procedures. Most of the literature revolves around this question.
It is thought that hard tissue blocks will have less surface variation near the cut areas than soft blocks (Mollenhauer, 1993)]. Hard blocks are also thought to have less deformation as a result of the cutting process.
Sectioning quality is a key area where the goals of brain preservation differ from the goals of EM. In the EM literature, plasticizers are frequently added to prevent the tissue from becoming too brittle (Lascano et al., 2019), which would otherwise make it difficult to be sectioned. But plasticizers decrease viscosity and the glass transition temperature. For brain preservation, we ideally want the tissue to be vitrified to prevent long-term biomolecular motion and promote long-term structural stability.
Preventing displacement of molecules by the electron beam, both due to heat and radiation-induced bond breaking is another important goal of plastic embedding for electron microscopy. The energy of the electron beam is quite high and it is able to displace biomolecules and resin molecules and can even knock them out of the section (Mollenhauer, 1993). The displacement may be up to tens of nanometers.
Different embedding media have different tissue stabilities under the electron beam. For example, Glauert reported that displacement damage is thought to be minimal for the epoxy resin Araldite that contains an aromatic group while it is higher for certain acrylic resins such as the Lowicryls (as cited in (Mollenhauer, 1993)). Methacrylates are sometimes not used as plastic embedding media because they produce tissue sections that are unstable under the electron beam (Yeung et al., 2015b).
From the perspective of brain preservation, stability under the electron beam is not a directly relevant factor. However, it may be a proxy for stability of the tissue outside of the electron beam. You could potentially think of resilience to damage under the electron beam as a type of stress test or an accelerated aging test.
One source suggests that polymerization of the acrylate resin LR white performed at 0°C will only be partial which means that there will only be “partial gelation” and around 30% of the resin still in the liquid state (Newman et al., 1999). These blocks might be unstable in the electron beam, suggestive of a correspondence between stability in the electron beam and the type of solidification that will be useful for long-term brain preservation.
It is not clear whether electron beam damage is a good proxy for damage accumulated during storage over very long periods of time at ambient temperature. However, all things equal, stability under the electron beam seems like a good sign for stability over time.
From a chemical perspective, the aim of resin embedding is to replace the tissue solvent – whether that is the original water, the dehydrating agent, or an intermediate solvent – with a liquid resin monomer that can then be polymerized to form a solid tissue block (Glauert et al., 2014, ch. 5).
The first step in resin embedding is infiltration. During this process, the resin mixture enters the tissue, either by diffusion or through an active force such as convection or vacuum forced impregnation.
The rate of infilitration of a resin mixture depends on the viscosity of the resin components: low viscosity components penetrate tissue faster (Mollenhauer, 1993). If there are multiple components of a resin mixture, they need to have similar viscosities or somehow be distributed evenly across the tissue. If the resin components are unevenly distributed across the tissue there will be uneven polymerization during the curing process, which will lead to a less stable biospecimen that may be more prone to fractures.
Some resin components may begin to polymerize during the infiltration process, especially because lowering the temperature to prevent this will increase the viscosity of the resins and make infiltration even more difficult (Mollenhauer, 1993).
Some biomaterials are difficult to embed because the resin easily cannot penetrate them or bind tightly to them.
After the liquid embedding agent has replaced the tissue solvent by infiltration, the next major step is called curing. Curing refers to the hardening of a polymer-containing material by crosslinking of the individual monomers, oligomers, and chains into a large crosslinked network. The term curing is strongly associated with thermosetting polymers that irreversibly harden after curing by heat or radiation, which is the case for most embedding agents used in EM.
The biospecimen is often cured and allowed to solidify inside of a plastic mold. In microscopy, this can be helpful for subsequent storage and/or cutting of the material.
While it is relatively easier to understand the polymer chemistry of an embedding chemical in its pure form, it is often difficult to define the chemistry of the embedding polymer inside of a tissue, because tissues are made of a heterogenous mix of biomolecules and therefore more difficult to study (Mollenhauer, 1993). The properties of resin and tissue aggregates may be much different from the properties of pure resin polymers (Glauert et al., 2014, ch. 5).
Mollenhauer defines resin “glueing” as the process by which a resin copolymerizes with biomolecules within the tissue. For example, many epoxy resins are known to copolymerize with biomolecules such as amino acids and nucleic acids (Mollenhauer, 1993). The better tissue elements are glued together, the more heat stable they will be as well.
From an electron microscopy perspective, glueing leads to fewer sectioning-induced surface irregularities and it is thought to be why epoxy resins have fewer of these than acrylic resins (Mollenhauer, 1993). From a brain preservation perspective, glueing is somewhat of a double-edged sword as it may causes changes in biomolecular structure and position during the glueing process, but then makes likely makes the tissue structure more stable over long term storage.
Generally, the first step in the curing process is that the polymer will form into a gel. After this, there are three main ways that the tissue can solidify during the curing process: the tissue can vitrify, it can semi-crystallize, or it can crystallize. For most embedding agents used in EM, the available evidence suggests that the tissue generally vitrifies.
A key question for the form of solidification that the embedding will undergo is the glass transition temperature of the resin/tissue combination, which may be different for different areas of the tissue. For many embedding polymers, the glass transition temperature of the pure resin has not been studied in depth and may not be known. Even if it is known, the glass transition temperature of resin/tissue combinations will differ and is best studied empirically.
During the curing process, the glass transition temperature rises as the embedding agent polymerizes. Vitrification occurs when the glass transition temperature is equal to or greater than the temperature of the sample that is being cured. At this point, there is generally a dramatic slowing of the cure rate, as the viscosity of the material is so high that the polymerization reactions are very slow. This, in turn, slows the increase of the glass transition temperature. As a result, the glass transition temperature of the resulting polymerized resin is limited by the curing temperature.
One key exception to this is the possibility of an exothermic temperature rise driven by the curing reaction itself, which can raise the glass transition temperature of the epoxy system (Kroutilová et al., 2006). Another exception is the use of a post-cure step with a higher temperature, which can also raise the glass transition temperature by causing more complete reaction and polymerization (Lascano et al., 2019). Raising the curing pressure can also increase the glass transition temperature (Nakamae et al., 1991).
A major trade-off is that a higher crosslink density and glass transition temperature will tend to the material more brittle and fractures more likely at a given storage temperature (Utaloff et al., 2019). This is the same phenomenon as is seen in vitrification by cold temperatures, where storage further below the glass transition temperature tends to make fractures more common.
For some plastics, this is why plasticizing agents are used, which lower the glass transition temperature and make fractures less likely. For example, plasticizers can be added to plastic automobile parts to prevent fractures when a car is exposed to a cold temperature (Rutledge, 2018). For epoxy polymers, the increased brittleness with a higher glass transition temperature can be mitigated through the use of toughening agents even without affecting the glass transition temperature (Utaloff et al., 2019).
Another key trade-off is that a high polymerization temperatures, such as those >= 50-60°C, may cause a loss of biomolecular conformation due to denaturation (Mollenhauer, 1993). High polymerization temperatures may also cause morphological changes in the tissue.
Areas of the polymer can also crystallize, depending on the structure of the polymer such as the degree of straight chains. For many polymers, the degree to which it polymerizes or vitrifies under different conditions has not been studied in depth. However, it is unlikely that there is a significant degree of crystallization for most plastic embedding agents used for EM, because crystallization would likely alter the tissue morphology.
From an electron microscopy perspective, vitrification and storage longevity is not a widely considered topic, because as long as the embedded sample lasts long enough to be imaged in the electron microscope for the purposes of one study, it doesn’t matter very much how long it lasts after that.
From the perspective of brain preservation, the presence of vitrification is critical because this is what will inhibit molecular motion over long periods of time and allow for the slowing of biological time. However, this means that the cure temperature needs to be above the storage temperature, because the glass transition temperature is limited by the cure temperature.
A general lesson from experimentalists in resin embedding is that there are many trade-offs involved in the choice of infiltration and polymerization procedures and that multiple parameter combinations will work reasonably well (Mollenhauer, 1993).
Most of the resins used in electron microscopy are called synthetic resins. This is as opposed to natural resins, such as those produced by trees that eventually polymerize into amber.
One of the frustrating aspects of the resin embedding literature is that some of the chemicals used are trade names and the precise chemicals involved are not disclosed. For example, as of 1999 this was the case for the acrylate resin LR White (Newman et al., 1999).
Embedding is a tiny field so this is perhaps not surprising, but it is certainly a problem when one is trying to do iterative science and understand the differences between the resins. There are sometimes calls from academic scientists to firms to disclose what the chemical mixtures are (Gerrits et al., 1996).
Paraffin wax is typically a mixture of straight chain alkanes with hydrocarbon chain lengths of 20-40. Paraffin can have different melting points depending upon the composition. Generally, the longer the average length of the carbon chains, the higher the melting temperature (Akgün et al., 2007). For histology, paraffin typically has a melting point of approximately 57°C.
Paraffin wax is the most widely used agent for embedding in conventional neuroscience research, by far. It is used in both basic science labs and clinical neuropathology.
Paraffin was first created in Germany in 1830 and it represented an important advance in candlemaking as it burned cleanly and was cheap to produce. In 1869, Edwin Klebs introduced paraffin infiltration for embedding (An et al., 2003).
Paraffin wax embedding is also sometimes not used for EM because the tissue blocks it provides are not sufficiently hard to cut at the thin 80-100 nm thick slices needed for EM. Instead, paraffin-embedded sections used for light microscopy are typically around 3-6 um thick (An et al., 2003).
When paraffin is used as an embedding agent, the fixed tissue must first be dehydrated, typically with ethanol. The ethanol is then replaced by a clearing agent, typically xylene. Xylene is a lipid solvent that could theoretically remove substantial additional lipids from the fixed and dehydrated tissue. Although in one study xylene treatment was found to only remove 1-2.5% of residual lipids as the dehydration process likely already removed most of the soluble lipids (Leist et al., 1986).
Finally, paraffin wax is melted to above its melting point, typically 60°C-65°C, and then the molten paraffin replaces the xylene and infiltrates the tissue. As with any high temperature embedding method, heating the tissue can cause damage (Gerrits et al., 1996).
While it is not done in routine processing it is possible to avoid a clearing step altogether, for example using acetone or isopropanol as the dehydration agent and directly replacing this with paraffin wax. However, the infiltration may not be as complete.
When the tissue and paraffin cools, it solidifies. When the straight chain alkanes cooling they generally seem to crystallize (Katz et al., 1945), perhaps because the straight molecular chains are easily organized into crystals.
Paraffin has been reported to solidify into multiple types of crystals (Rhodes et al., 2002). Paraffin has also been reported to solidify into a smectic phase, which is a type of liquid crystal with soap-like properties (Machado et al., 1959).
Paraffin wax crystals have been reported to cause tissue deformations (Conkie, 1965). Uneven crystallization with areas of large and small crystals has also been reported to cause the tissue to harden unevenly.
It is unclear what happens to the tissue biomolecules during this solidification process. It seems unlikely that paraffin itself reacts with any biomolecules given that it lacks clearly reactive groups.
Perhaps in part because paraffin embedding does not allow for sectioning that is compatible with the thin slices needed for EM, there is a general impression that paraffin embedding is not compatible with fine structure preservation as seen under EM (Wang et al., 1987). However, studies that have evaluated this empirically have found that ultrastructural preservation in paraffin embedded tissues is actually quite reasonable (Wang et al., 1987). It has been reported to be mostly dependent upon the quality of the initial fixation, as is seemingly everything else in histology.
One study evaluated human tissue biopsy samples preserved with multiple different methods, including (a) formaldehyde fixation followed by paraffin embedding for light microscopy and (b) glutaraldehyde fixation followed by epoxy embedding for electron microscopy (Wang et al., 1987). They preserved 2-3 mm thick slices in a fixative volume at least 20x the volume of the tissue and immediately replaced the fixative if it became blood tinged or cloudy. They also deparaffinized paraffin embedded tissue and re-processed it for electron microscopy. This allowed them to determine the ultrastructural changes induced by the the process of paraffin embedding.
They found a slight improvement in fine structure preservation in tissues initially fixed in glutaraldehyde compared to formaldehyde, with better preservation of membranous structures and other loose components of the tissue, as expected. The main barrier to EM in this tissue seems to have been the lower quality formaldehyde fixation. However, they noted that the paraffin embedding itself seemed to make little on the ultrastructural quality. They found that all of the paraffin embedded tissues had “good, at at least passable, fine structure preservation.” However, they did note that poorly preserved tissues might be altered further as a result of paraffin embedding.
Notably, glutaraldehyde fixation can be used prior to paraffin embedding, to improve fine structure preservation, at the potential expense for antigenicity to certain antibodies (Pudney et al., 1995).
One study reported on the author’s experience using paraffin embedded tissue for EM studies. They reported that ultrastructural morphology preservation has been found repeatedly to be acceptable in formalin fixed, paraffin embedded blocks, although it is poor in paraffin sections.
One source reported that they had paraffin embedded a whole human brain (Amunts et al., 2013). The brain was first fixed in formalin for 5 months. The whole brain was then embedded in paraffin and sectioned at 20 um thickness, creating 7404 total sections. It is unclear what was used for dehydration. Each section was then stained and imaged for histology. An image is available here.
One study also studied brain stems that were embedded in paraffin after being formalin fixed, dehydrated in isopropanol, and cleared with xylol (R. Quester et al., 1997). This study found that paraffin embedding led to a shrinkage of approximately 11-12% trasversely and 17% longitudinally, which was slightly less than the shrinkage values of 11–25% that had been previously reported for human cerebral tissue. So, paraffin embedding certainly does seem lead to some shrinkage of human brain tissue.
One study reported embedding a macaque brain hemisphere in paraffin and discussed the challenges involved in paraffin embedding large tissues. It is certainly part of the challenge that would be relevant if paraffin embedding were used in brain preservation. (Zhanmu et al., 2020).
Formalin fixed paraffin embedded (FFPE) tissue is very commonly used in pathology and histology. FFPE blocks are often stored at room temperature.
Because FFPE tissue storage is so common and this procedure has been around for so long, some studies have looked at older FFPE biospecimens to determine the tissue quality.
One study found that synapse immunoreactivity is preserved for up to around 25 years in FFPE human brain tissue (Liu et al., 1995 Jul-Aug).
Another study measured immunostaining of proteins in FFPE bladder tumor tissue blocks collected from the 1930s to the 2000s (Litlekalsoy et al., 2007). The FFPE blocks were stored at average ambient temperatures of 18-20°C (this was in Norway) at relatively high humidity due to the local climate. Overall, they found that the five nuclear and cytoplasmic markers that they measured were very well preserved over time and were detected as well in old FFPE blocks as in new FFPE blocks. The only antigen that decreased over time was p53, however the authors thought that this was more likely due to a shift in the malignancy potential of the bladder tumors collected over time rather than actual degradation. This study suggests that protein immunogenicity antigens can be preserved for at least 70 years in FFPE tissue at ambient temperature.
One study suggests that removal of residual water is crucial to prevent antigen degradation during long-term FFPE tissue sections (Xie et al., 2011). They found that storing FFPE tissue sections with desiccant protected against degradation of immunostaining over 3 months.
Regarding nucleic acids, there appears to be slightly more degradation during storage than proteins.
One study found that room temperature led to degradation of DNA and RNA in FFPE tissue blocks compared to those stored at 4°C over a period of nine years (Groelz et al., 2018).
One study found that there was no significant difference in RNA stored in FFPE tissue blocks at ambient temperature compared to 4°C at 6 months (Detmer et al., 2016). Tissue sections, on the other hand, were found to be more vulnerable to degradation, and RNA in tissue sections did demonstrate more degradation when stored at ambient temperature compared to 4°C over 6 months. This is likely because biomolecules in tissue sections are exposed to more oxygen, atmospheric water, and other environmental factors that can promote degradation (Xie et al., 2011).
Some nucleic acids have been found to survive in paraffin embedded tissue over the very long term. For example, the 1918 influenza virus was discovered through analysis of RNA fragments from formalin-fixed, paraffin-embedded autopsy tissue (Taubenberger et al., 2007).
One source reports that there is not an extensive degree of oxidation in FFPE preserved tissue (Davalieva et al., 2021):
Analysis of post-translational modifications which could be linked to the chemistry of formalin fixation in our dataset revealed significant increase of methylation and formylation of lysine at N-term and slight but not significant increase in hydroxylations on mainly basic amino acids such as asparagine, aspartic acid, proline or lysine in FFPE tissues. Lysine methylation is one of the most abundant variable modifications resulting from formaldehyde crosslinking, already recognized by a number of studies. On the other hand, formylation of lysine has been found both with no difference and with significant increase in FFPE tissues, although compared with that study, we found a lower percentage of peptides containing formylated lysine which might result from more efficient de-crosslinking in our study. In addition, we observed slightly higher number of methionine oxidations in FFPE samples, which is consistent with other published studies. But, in our dataset, the difference between fresh and FFPE samples did not reach statistical significance, in line with previous observations, suggesting that formalin fixation and storage do not result in an excessive degree of oxidation.
One source reports that proteomic data is not substantially changed after 7 years of storage of FFPE tissue (Jiang et al., 2007).
An important aspect of this study is that the FFPE tissues were “real” archival, diagnostic FFPE samples, instead of the experimental FFPE models used in many of the reported studies comparing FFPE with fresh/frozen material [11,19,47,48]. Moreover, our material had a considerable storage period of 7 years. In agreement with few published reports that analyzed archival FFPE samples in storage for up to 10 years [49,58], we independently validated the finding that high quality proteomic analysis is possible even after a prolonged period.
Overall, paraffin embedding may be overrated in the field of histology for path dependency reasons, because it was the first option used historically. Regardless, it does have major advantages for potential use in brain preservation in that it is widely available, relatively cheap, and there is an extensive database on its use showing relatively good histology preservation over time. It has also clearly shown the ability to preserve protein antigens for a very long time and be able to be performed on tissues the size of the human brain. There is still some concern about damage under electron microscopy, as the data for that is not extensive. There is no data available for connectomics, such as volume electron microscopy, which is among the most important types of evidence in evaluating brain preservation methods. If other alternative room temperature preservation methods are found to not be acceptable, paraffin embedded may be worthy of some consideration.
Polyester wax does not seem to be very commonly used in embedding, but it has some interesting properties. Polyester wax embedding has been found to preserve tissue structure better than paraffin, likely because polyester retains more lipids (Sidman et al., 1961). Polyester wax has a lower melting temperature (37°C compared to ~46-68°C for paraffin), which means that there is less heat/rapid dehydration-induced shrinkage of tissues during embedding (Merchant et al., 2006). This may be worthy of further study.
Celloidin is purified nitrocellulose, which is a polymer of nitrated cellulose. Nitrocellulose is also known as gun cotton due to its use as an explosive and replacement for gun powder.
Celloidin is often considered a better option than paraffin when working with larger and more fragile tissues (An et al., 2003). It has also been reported to allow for high quality morphologic preservation (Merchant et al., 2006).
The major advantage of celloidin embedding is that it bypasses the need for heating the sample at any stage of the process. This prevents heat-related artifacts, such as shrinkage, which is minimal. Celloidin embedding is also helpful at preserving dense tissue such as brain tissue and internal structures such as the layers of an organ tend to be better maintained compared to paraffin embedding. Finally, the absence of heat is likely to prevent heat-induced denaturation of biomolecules.
There are several downsides to celloidin as well. First, infiltration takes a very long time, reportedly as long as 3-6 months for large tissues (Block et al., 1982). Also, celloidin blocks must be maintained in liquid ethanol to prevent shrinkage, which is inconvenient and presents an additional failure point for the brain preservation procedure. Celloidin vapor is very flammable, c.f. its use as gun powder, which can be quite dangerous. As a result celloidin may not be the best option for brain preservation.
Epoxy resins were the first embedding medium used in electron microscopy. As Mollenhauer reports (Mollenhauer, 1993):
The widespread use of epoxy resins dates from the published reports of the Glauerts (primarily Audrey Glauert) in 1956 and 1957, although others reported the use of epoxy resins at essentially the same time (e.g., Maalere and Birch-Andersen, 1956). The Glauerts used a Ciba product called Araldite M, which was reacted with an anhydride hardener. The reasons for the widespread acceptance of epoxy resins for embedding were that they gave excellent structural preservation, were relatively stable in the beam, and could be sectioned easily. However, Araldite M was not immediately available in the United States and a similar product, Araldite 502, was substituted.
Epoxies are commonly used in a procedure that combines initial aldehyde fixation, osmium tetroxide processing to stabilize the lipids, and then embedding in Epon, a type of epoxy (Mollenhauer, 1993).
In addition to Epon (aka Epoxy Embedding Medium) other types of commercial epoxy resins that have been used are Spurr, Epofix, Epo-Kwick, Epo-Thin, Technovit 9100 New, Unicryl, and Vestopal W (Li et al., 2004). Arthur Spurr first reported what is now called Spurr’s media in 1969, reporting that it had low viscosity, and it has been commonly used (Spurr, 1969).
Epoxies tend to have relatively high viscosity necessitating relatively long, higher temperature infilitration procedures. This means that in order to use epoxy resins, it is also often necessary to use osmium, to prevent large amounts of biomolecular extraction (Newman et al., 1999).
Proteins have been reported to react with monomers of epoxy resins and co-polymerize with them (Yamashita et al., 2014).
One of the problems with epoxy embedding is that it can cause a loss of antigenicity:
The conventional osmium-Epon tissue processing appears a favorable approach, since natural sizes are well retained after this treatment (Griffiths et al., 1984; Konwinski et al., 1974; Tooze, 1964). However, in spite of occasional successful localization studies on Epon-embedded tissue (Bendayan and Zollinger, 1983), retention of antigenicity is often very limited. Therefore, aldehyde-fixed tissue, dehydrated and embedded in resins such as Lowicryls and London resins, has been introduced in immunocytochemistry. - (Mollenhauer, 1993)
Epoxies seem to generally cause poor antibody penetration, which makes combination immunostaining and electron microscopy difficult (Smith, 2018).
The choice of embedding resin chemistry also entails a tradeoff, with acrylic embedding resins (such as LR White and Lowicryls) offering much better antibody access for immunofluorescence while epoxy resins generally yield superior EM image quality
However, antigen retrieval is possible for many antigens embedded in epoxy resin, suggesting that the biomolecules are likely still present in the tissue, albeit inaccessible to immunostaining methods (Yamashita et al., 2014):
The present study clearly demonstrates that [heat-induced antigen retrieval] is effective for many antigens in tissues fixed with glutaraldehyde and post-fixed with osmium tetroxide and then embedded in epoxy resin, although the detection of antigens has been thought to be a special case in these specimens
The use of epoxies has been reported to quench biomolecular flourescence because of autofluorescence and high polymerization temperatures (Zhou et al., 2017):
Epoxy is usually used for sample embedding for electron microscopy owing to its high autofluorescence and high polymerization temperature, which quenches fluorescence irreversibly. Thus, it is not suitable for embedding fluorescent samples
Someone asks on Chemistry Stack Exchange: how do epoxies cross-link?. One study used molecular dynamics simulations to determine thermomechanical properties of cross-linking with epoxies, such as the glass transition temperature (Fu et al., 2017).
One source notes that gelation is considered the rubber form of an epoxy resin (Lange et al., 2000). Then the substance makes a rubber-glass transition when it vitrifies. So the first transition is liquid to rubber, then rubber to glass.
Epoxy resins are used in combinations with other chemicals to achieve the necessary embedding results (Mollenhauer, 1993):
The structure of the polymer is influenced by the ratio of epoxide groups to hardener groups, the amount of accelerator, and the initial temperatures at which the resin mixture is polymerized (Hayat, 1970; Luft, 1973). In addition, the resin will copolymerize with a number of tissue elements (e.g., amino acids and nucleic acids; Acetarin et al., 1987; Kellenberger et al., 1987), with the result that the polymer within a tissue block may be substantially different from that outside the tissue block.
One source investigates the curing process of epoxy resins, particularly when using amine hardeners, as this significantly influences the final product’s physical properties (Dyakonov et al., 1996). They investigate the relationship between crosslink density, stoichiometry, glass transition temperature (Tg), and thermal degradation onset temperatures, discovering that Tg increases with crosslink density and post-curing. However, they also note that higher Tg values do not necessarily correspond to enhanced thermal stability, as demonstrated by a few comparisons between different resin types.
One source has a simple explanation of the glass transition temperature (Tg) in epoxies:
The Tg is not a well-defined single temperature. The softening of the epoxy happens over a temperature range, typically of about 10 degrees. The diagram to the left shows a generalized version of the Tg transition point. The left side of the curve is the region of the more solid, glass-like cured epoxy. The more steeply sloped right side of the curve is the rubbery state of the cured epoxy. A single point, the Tg, is determined by extending the linear parts of the curve before and after the inflection point to arrive at a single point. On data sheets the Tg is typically reported as a single number, centrally located in the inflection of the curve.
Complete infiltration is important for epoxies as they need to surround the entire sample and form good attachments. Low viscosity helps with this. Investigators can also low vacuum infiltration can be used to ensure complete infiltration of porous samples (Echlin, 2009).
Epoxy resins used for embedding electron microscopy samples are composed of four key components (Echlin, 2009). The base epoxy resin contains epoxy groups, which are reactive and allow for cross-linking during polymerization. Three additives are mixed with the epoxy resin to control its properties: a hardener to initiate and control the speed of polymerization, a plasticizer to modify the flexibility and toughness of the cured resin, and an accelerator to further adjust the rate of the curing reaction.
Insufficient resin penetration can be a problem with epoxy infiltration, even in samples as thin as 1 mm slices (Ngo et al., 2019). Factors contributing to poor infiltration include insufficient immersion and perfusion time, inadequate dilution steps, and excessive resin viscosity. High-pressure vacuum infiltration (Kuo, 2007) and gradual polymerization under vacuum (Ngo et al., 2019) have been used to promote complete epoxy penetration.
Epoxies were used in Shawn Mikula’s BROPA (brain-wide reduced-osmium staining with pyrogallol-mediated amplification) method that was one of the contenders for the brain preservation prize in the mid 2010s (Mikula et al., 2015):
The epoxy formulation, a modification from Spurr’s, was: 10 g vinylcyclohexene dioxide (VCHD or ERL-4206), 20 g nadic methyl anhydride (NMA, Electron Microscopy Sciences) and 0.45 g dimethylaminoethanol (DMAE, Serva).
One study used a type of epoxy, Epon 812, in order to embed a 4x8 cm fish head (Anker et al., 1974):
Shrinking or swelling is slight during polymerization, staining, and mounting. The coherence of the tissues is preserved much better, and the histological details are at least as good, with the exception of undecalcified bone, which crumbles internally during cutting (Fig. 2). However, the epoxy resin encloses the bony elements so tightly that their shape and position are preserved very well, provided they are not too large.
The modification concerns the prolongation of the various steps to obtain better penetration of the fluids. The retardation of the polymerization also prevents temperature elevations inside the block (“hotspots”) and thus damage to the tissue. The temperature can be controlled by keeping the resin mixture at 4 C when an accelerator is used or by omitting the accelerator (Pool 1969). In the latter case the polymerization time is about four weeks at 60 C.
Mollenhauer has some good points on the viscocity and storage stability of an embedded epoxies (Mollenhauer, 1993):
Epoxy resins used in electron microscopy are not heat stable and soften significantly as their temperature is raised. For example, note differences in block hardness at room temperature compared to that in a polymerization oven set at 60°C. Even a few tens of degrees is sufficient to make some resin formulations soft and rubbery with the result that tissue elements may move within them. Such temperature increases are easily achieved even under very low levels of illumination - (Mollenhauer, 1993)
Polymerized epoxy resins act as viscous fluids (Mollenhauer, 1988, 1990). This is easy to demonstrate by looking at block faces that have been sectioned and then stored for long periods of time. A block face that is smooth immediately after sectioning often shows contours of tissue elements after periods of storage. The actual displacements of tissue elements are usually small, although a few tenths of a nanometer per hour have been calculated for extreme examples such as that illustrated in Figure 5. This problem may be more acute in sections and may set a useful limit to section life. This effect is very dependent on resin formulation. Another instability associated with periods of storage is illustrated in Figure 6A,B. In these examples, a loss of contrast and a propensity to form pepper (black specks; Mollenhauer, 1988) occurred in sections stored 1 month before they were poststained and examined. The causes of these effects are not known. - (Mollenhauer, 1993)
Epoxy resins are almost never maximally cross linked, and components of them commonly vaporize in the electron beam. Most also contain unreacted diluents such as dibutyl phthalate. The amount of vaporization depends on resin formulation but commonly varies from -7% (the best that can be achieved with commonly used resin mixtures) to -40% (Fig. 1; D.L. Ringo, personal communication). Moreover, resins are almost never fully polymerized, because they are not properly formulated, because they are removed from the oven too soon, or because not enough accelerator was used to assure full polymerization in the 24-48 hr usually allocated for the hardening process. Electron beam radiation may also break bonds within the resin and tissue causing fragments of them to be lost (A.M. Glauert, personal communication). This kind of damage is independent of degree of cross liking - (Mollenhauer, 1993)
Epoxies may be more stable if they are more fully polymerized and/or dehydrated, although these would be associated with more biomolecular alterations.
Taken together, this source suggests that it cannot be assumed that just because tissue has been embedded in epoxy, that it will necessary be stable at any temperature for an indefinite amount of time. It would require some further investigation.
However, the practical storage longevity of epoxy stored specimens appears to be good. For example, one scientist reported that they were able to properly section and image human cartilage specimens that had been embedded in epoxy three decades prior.
Many histology, pathology, and hospital laboratories maintain archives of epoxy-embedded materials for morphological studies, with the expectation that these archives will provide valuable data for future research (Yamashita et al., 2014). Another source notes that once epoxies are embedded and polymerized into insect tissues, the tissue can be stored for years (Pernstich et al., 2003).
One study reported the successful embedding of whole mouse brains in epoxy resins, which were cured at room temperature (Becker et al., 2014). They used tetrahydrofuran as the dehydrating agent, which also cleared the tissue, resulting in lipid loss. Notably, the study showed that the epoxy-embedded brains preserved fluorescence when stored at room temperature for at least two years, which was the duration of their testing. This study is significant as it is relatively rare to find research focusing on the long-term preservation of epoxy-embedded tissues, and the results are promising for the use of epoxies in tissue preservation.
Acrylates are chemical derivatives of acrylic acid that are frequently used as monomers in the formation of polymer plastics. Acrylates embedding resins can infiltrate biological tissues and then be polymerized by a radical reaction, which can be initiated by ultraviolet light, benzoyl peroxide, or other sources (Fleck et al., 2019). This radical reaction can be stimulated at subzero temperatures. Many acrylate embedding agents can be used to preserve the structure of biomolecules, such as fluorescent proteins (Porrati et al., 2019).
In 1949, a combination of methyl and n-butyl methacrylate was first introduced as an embedding media and was found to be suitable for cutting sections for electron microscopy (1967). It had a good rate of penetration of tissues and allowed for thin sectioning. This was the embedding media of choice for approximately 10 years before it was replaced by epoxy resins in 1960.
Compared to epoxy resins, acrylates are somewhat unique in that they can tolerate small amounts of water, such as up to 10-12% by weight [https://link.springer.com/content/pdf/10.1023/A:1003850921869.pdf]. This means that dehydration does not always have to be as complete, which can be helpful for preserving tissue structures. However, from the perspective of brain preservation, this may lead to worse long-term maintenance of tissue quality during storage.
A key problem with acrylate resin embedding is that it is difficult to polymerize via UV light in large biospecimens. One source notes: “Only very small pieces of tissue can be embedded because they must be penetrable by ultra-violet light.” (Newman et al., 1999)
As opposed to epoxies, methacrylates have been reported to crosslink primarily with themselves and have little interaction with biological macromolecules. Therefore, methacrylate embedding has been described as “more like trapping the biological components in the meshwork of the polymer, like fishes in a net” (Fleck et al., 2019).
According to one source, acrylate resins are used in order to preserve antigenicity and fluorescence, while epoxy resins are generally used when the goal is ultrastructural preservation (Burel et al., 2018).
It’s important to note that acrylamide embedding is also a thing (Lai et al., 2016), but despite the similarity of their names, acrylates are not the same as acrylamides.
Three examples of acrylates that have been widely used as embedding resins and we will discuss here are glycol methacrylate, methyl methacrylate, and the acrylate mixture Lowicryl.
Glycol methacrylate, also known as 2-hydroxyethyl methacrylate (HEMA) (Yeung et al., 2015b), is a somewhat commonly used embedding resin for light microscopy. It is slightly hydrophobic, so a dehydration step is required prior to infiltration of the embedding agent, eg with ethanol or acetone (Glauert et al., 2014, ch. 7).
One source reports that glycol methacrylate has a glass transition temperature of 55°C, so a plasticizer is added to allow for sectioning at a lower temperature (Glauert et al., 2014, ch. 7). This is because blocks are ideally sectioned at a temperature close to the glass transition temperature.
Advantages:
Disadvantages:
In summary, glycol methacrylate is a suitable embedding medium for light microscopy, preserving antigenicity and enzyme activity in relatively large tissue samples. Its use in electron microscopy is limited due to poor ultrastructure preservation and sensitivity to the electron beam. These properties seem to make it less attractive for investigation in brain preservation as well. The long-term storage potential of GMA-embedded tissues at ambient temperatures remains uncertain.
Advantages:
Disadvantages:
Overall, methyl methacrylate is an interesting embedding agent that seems worthy of future research.
Lowicryl is a complex mixture of cross linked acrylate and methacrylate esters (Carlemalm et al., 1985). They are densely cross-linked and have low viscosity. They were developed in the 1980s to allow for allowing for embedding at low temperatures and thereby reduce the modification and extraction of cellular components during the embedding procedure (Glauert et al., 2014, ch. 7).
Lowicryl resins can be photopolymerized by UV light at low temperatures (preferred, (Newman et al., 1999)) or chemically polymerized at higher temperatures. There are many different compositions of Lowicryl, such as HM20 (Newman et al., 1999).
Among acrylate resins, Lowicryls have the lowest viscosity, allowing for fast and deep infiltration and therefore the best preservation of ultrastructure (Porrati et al., 2019).
Lowicryl HM20 can be embedded into samples at temperatures down to -70°C (Porrati et al., 2019). It is moderately hydrophobic, allowing for preservation of protein conformation and associated fluorescence and antigenicity. It allows for thin sectioning and can be used with FIB-SEM (Porrati et al., 2019).
It has been used for embedding whole mouse brains (Gong et al., 2018).
One study performs cryofixation followed by freeze substitution and then FIB-SEM of Lowicryl embedded brain tissue, with nice results (Porrati et al., 2019).
It is not entirely clear how stable the cured material would be at room temperature over the very long term. Several sources do report that this tissue can be stored at room temperature, 4°C, or -20°C, but they note that it needs to be stored in a dessicator as the resin is polar and can readily take up water (Hayat, 1989) (Glauert et al., 2014, ch. 7).
Other sources have reported good success with Lowicryl (Jones, 2016) (Oprins et al., 1994). One study using Lowicryl HM20 on 1.2 mm thick brain slices reported that the fine structures of neurite were preserved and that multiple antibodies and fluorescent tracers showed good results (Gang et al., 2017).
Overall, Lowicryl resins, especially Lowicryl HM20, allow for high quality in both cellular fine structure preservation and immunolabeling (Kiss, 1989). However, Lowicryl embedding can have relatively poor tissue infilitration (Kiss, 1989). This is especially true in the brain, because of its high hydrophobic lipid content, as Lowicryls are hydrophilic polar resins (Kiss, 1989). So there may be more research needed before this is a practical option for human whole brain preservation.
Plastination involves fixation, dehydration and clearing, and embedding of curing of polymers in order to preserve body parts or whole bodies. It was introduced by Gunther von Hagens in 1977 (Pashaei, 2010).
While plastination is similar to other tissue embedding procedures, it is a bit separated in the literature from the rest of these procedures, which may be due to the marketing efforts of von Hagens and their company Biodur (Pashaei, 2010).
One difference is that the plastination literature tends to be more focused on the macroscopic anatomy of large biospecimens, such as organs or whole bodies. Whereas much of the rest of the plastic embedding literature tends to be focused on the microscopic anatomy of smaller biospecimens, such as dissected blocks of tissue from a particular organ.
In brain preservation, we are interested in the microscopic anatomy of larger biospecimens, which means that the optimal procedures may draw upon aspects from both plastination and plastic embedding for electron microscopy. Plastination techniques deserve consideration in brain preservation, as they have clearly shown the ability to scale to the size of the whole body.
Plastination has come under some criticism for whether the people whose bodies were preserved had been fully consented about the details of the procedure prior to their death (Riederer, 2014). Informed consent is critical in order to perform any procedure on someone’s body in an ethical manner, including brain preservation (this will be discussed further in a later essay).
The procedure of plastination is basically four steps (Waters, 2010):
1: Fixation of an organ or the whole body, sometimes via perfusion.
Generally formaldehyde is the key component of the fixative. Potassium nitrate and potassium acetate may be used as well in Kaiserling-I solution (Pashaei, 2010). This is also sometimes called the embalming step.
2: Dehydration and clearing (lipid removal or “degreasing”).
In plastination, the dehydration step is often done with acetone at low temperatures of approximately -25°C. Performing dehydration at this subzero temperature tends to reduce tissue shrinkage, although it is also expected to lead to ice crystal formation. This is an example of how plastination procedures focus on macroscopic rather than microscopic anatomy preservation. It is also possible to perform the dehydration step at room temperature, although this is non-standard (Zheng et al., 1998) (Starchik et al., 2019).
For a whole human brain, the acetone dehydration procedure is reported to take approximately 3 weeks (Riederer, 2014).
Clearing is often done by increasing the temperature of the acetone-filled specimen to room temperature, since cold acetone does not efficiently remove lipids (Robert W. Henry et al., 2019).
3: Polymer infilitration.
In polymer infilitration, the acetone-filled specimen is first placed in a bath of the liquid polymer at a cold temperature of approximately -25°C. The system is then placed under a vacuum, which decreases the pressure of the system. The pressure is lowered slowly so that the acetone boils slowly and the specimen does not collapse or shrink without solvent (Riederer, 2014). For example, one study gradually decreased the pressure from 75 mmHg to 5 mmHg in increments of 10 mmHg (Asadi et al., 2013) (note that atmospheric pressure is 760 mmHg).
As the acetone is removed from the biospecimen via vaporization, the liquid polymer, which has a low enough vapor pressure that it does not vaporize, infilitrates the tissue and replaces the acetone. Once all of the acetone has been removed by vaporization, the liquid polymer solvent will have fully replaced it.
Acetone has a high enough vapor pressure to boil even at lower temperatures of around -25°C, so this vacuum infiltration procedure is thought to only be practical with acetone, not ethanol (R. W. Henry, 1998).
Polymer infiltration by vaporization of acetone has been described as the key to plastination and it is the step that has been attempted to be protected by patent (Bickley et al., 1987).
(Note that this is often called “impregnation” or “forced impregnation” in the plastination literature, but I use infilitration because it seems more descriptive to me and it seems to be the more common term in the embedding literature in general.)
For the most common silicone embedding method (called the S10 silicone technique), polymer infiltration is actually done with both the silicone polymer and a catalyst/chain extender (which is called S3). Even with the catalyst present, reaction proceeds slowly during the infiltration at -25°C, so viscosity of the solution remains low, allowing infilitration to proceed (Robert W. Henry et al., 2019).
Once infiltration is complete and the specimen is raised to room temperature, the reaction will begin to occur, which can take several weeks depending on the size of the specimen. The curing reaction is speeded by putting the specimen in a container along with yet another chemical, called S6, that vaporizes and infilitrates into the specimen as a gas. The S6 vapor will first contact the surface of the specimen, so this part will cure faster than the inside of the specimen.
For specimens embedded with epoxy polymers, curing is done by heating the temperature of the specimen to 45°C (Ravi et al., 2011). For specimens embedded with polyester polymers, curing is done by UV light exposure followed by heating at 45°C (Ravi et al., 2011).
Vacuum infiltration of embedding polymers is certainly not limited to plastination procedures. Many studies have used vacuum infiltration of embedding polymers prior to the invention of plastination (for example, (Inman, 1968), (Kimmel et al., 1975)) and many studies have used it since.
Vacuum infilitration is generally thought to lead to a more uniform penetration of embedding agents than simple immersion infilitration. It helps to ensure that both small structures and large structure such as the ventricles will be filled (Horton et al., 1980) [[https://www.tandfonline.com/doi/pdf/10.3109/10520298009067891 (Pernstich et al., 2003). By varying the pressure used, it is possible to vary the force of embedding agent infilitration.
In plastination, many of the polymers are sold by von Hagens’s company, Biodur. The main three types of polymers that are sold by Biodur are silicone rubber, epoxy resins, and polyester resins. Like many polymers used for embedding, their precise chemical composition is not publicly known, as far as I know (Srisuwatanasagul et al., 2010).
In the ]2016 Biodur plastination catalogue](http://www.biodur.de/assets/biodur_catalogue_usb_2016.pdf), epoxy resins are only recommended for use with “sheet plastination” of tissue sections. The polyester resin is only recommended for brain slice sheet plastination.
However other groups have reported using the plastination procedure with polymers other than silicone, such as epoxy or acrylic resins, for embedding large tissues (Pandit et al., 2015).
One of the major advantages of embedding in silicone rubber as compared to epoxy is that the biospecimens are easier to handle (Grondin et al., 1994).
Silicone polymers are made of siloxane functional groups with Si–O–Si linkages. An example is polydimethylsiloxane (PDMS).
Si–O–Si bond angles are bigger than bond angles in C-C or C-H bonds, making the polymer chains more elastic and flexible at low temperatures.
Polydimethylsiloxane (PDMS) has a low glass transition temperature of around -115°C, so it will not be close to vitrification at room temperature (Lazurkin et al., 1958). Instead, the polymer appears to crystallize first, with a melting temperature of -40°C. Surprisingly, chemically crosslinking the ends of polymer chains tends to increase the crystallization tendency (Dollase et al., 2002).
However, these results tend to be for pure PDMS. It is unclear how the silicone polymers would solidify when combined in a heterogenous tissue. When silica particles are added to PDMS, crystallization tendency decreases, suggesting that when embedded in a heterogenous tissue, silicone polymers may have much different thermodynamic behavior (Fragiadakis et al., 2005). Further, while PDMS is a model silicone polymer, it not the same as the Biodur S10 silicone (Marks et al., 2008).
In one study, epoxy embedding of biospecimens was found to lead to less shrinkage of rat tissues than polydimethylsiloxane embedding (Marks et al., 2008). This difference was attributed to the epoxy monomer being less viscous and so diffusing better and more uniformly during infiltration.
Many studies have used plastination to preserve brain tissue in the dry state, which is primarily used for for neuroanatomy education (Weiglein, 1997).
My impression of the literature is that most studies perform “sheet plastination,” where they slice the brain prior to embedding it into one of the resins (Taghipour et al., 2016). While this is a useful technique, it makes plastination less unique compared to other methods for embedding brain tissue.
Silicone embedding of whole human brains has certainly been performed, however. In one study, gross anatomical features such as grey-white matter distiction and nuclei were preserved, although the microanatomy was not queried (Baeres et al., 2001).
In another study, it was reported that gross anatomical features such as grey-white matter contrast were not well preserved after silicone embedding of whole human brains (Wadood et al., 2001). This may have been due to incomplete infiltration. They did report that using the plastination procedure with embedding in a polyester polymer led to better gross anatomic preservation.
When performing plastination of brain tissue, there tends to be shrinkage. One study reported that there was often a shrinkage of about 20% of brain tissue mass during plastination (Riederer, 2014). This has been reported to be one of the major issues with plastination of brain tissue. Another study performing sheet plastination of donkey brain tissue found that shrinkage was 8% (Mohamed et al., 2019).
There have been some reports that using variations on the plastination procedure may lead to improvements in preservation of brain tissue. For example, one study found that performing the vacuum-driven silicone infiltration at room temperature led to superior outcomes for brain preservation, including better maintenance of original weight (10% loss of weight in room temperature compared to 50% loss at -25°C) and better maintenance of morphologic features (Owolabi et al., 2019).
In a sign of the accessibility of the technique, people can buy apparently plastinated animal brain tissue online, such as a whole plastinated sheep brain.
One of the major goals of plastination is to retain the natural color of the biospecimen, as this is helpful for education and display purposes (Pandit et al., 2015). However, this is not a major goal in brain preservation except to the extent that the biomolecules providing color to the tissue are related to memories or something else that is desirable to preserve.
Another goal in the plastination literature is to make the tissue flexible so that it can be put in a particular position for display purposes. This is not a goal of brain preservation and in general it is probably better for the tissue to be less flexible as this may be more stable over the long term, although also not so brittle that the tissue is liable to fracture.
Many articles about plastination claim that tissue microanatomy is well preserved by the procedure. But they frequently cite small, old studies that I cannot find (Bickley et al., 1987) (Ravi et al., 2011). It seems that the evidence base for this claim is relatively small and inconclusive.
One of the tricky things in studying the microanatomy of silicone-plastinated tissue is doing so requires the use of sodium ions to cause depolymerization of silicone rubber, in a procedure known as deplastination (Ravi et al., 2011). This is necessary when one desires to re-embed the tissue in a resin more compatible with electron microscopy, such as an epoxy resin.
One study used used two methods of preservation for oral tissue biopsies fixed in formalin: standard paraffin embedding and plastination by von Hagens’s S10 silicone method (Rahul et al., 2020). The silicone plastinated tissue was left to cure for 3 months in a cardboard box and was then deplastinated for 2 days in a 5% solution of sodium methoxide in 90% methanol. The deplastinated tissue was then paraffin embedded. Tissue processed via both techniques were then H&E stained and evaluated by blinded evaluators for tissue architecture, staining quality, and intracellular structure. The tissue that had been plastinated and then deplastinated was found to have significantly lower scores on all three of these metrics. Some of the difference between the groups was attributed to incomplete deplastination, especially on larger tissue specimens. It is unclear the degree to which the microanatomy would have been preserved in the absence of deplastination.
One study evaluated pancreas and spleen tissue that had been fixed and then either plastinated or embedded in Epon (an epoxy resin) for electron microscopy (Grondin et al., 1994). This study found that ultrastructure was retained in pancreas and spleen tissue embedded in silicon rubber after initial fixation with a formaldehyde- and glutaraldehyde-containing fixative. They had difficulty with deplastination and re-embedding of pancreas tissue for ultrastructure studies, but deplastination was successfully performed on the spleen tissue.
The major variable in this study was the fixation method: the fixative that contained 0.56% glutaraldehyde in addition to 2% formaldehyde led to successful ultrastructural preservation regardless of the embedding procedure, whereas the fixative that contained formaldehyde alone did not lead to sufficient ultrastructure preservation. The authors concluded, “clearly, plastinated tissue retains its ultrastructure.”
One study used epoxy sheet plastination of fixed human brain tissue that had initially been embedded in gelatin. They found that the arachnoid membrane was preserved well enough with this method to visualize it under the dissecting microscope to analyze blood vessels (it seems that the SEM data in this study was not plastinated, or at least it is not clear) (Zhang et al., 2000).
It has been claimed that plastinated specimens “do not deteriorate with time”, however it is unclear how long of a timespan this claim is relevant for.
One review notes that plastinated specimens have “semi-permanence”, so that preservation procedures only need to be performed only 10 years or so rather than every few years with other preservation techniques (Estai et al., 2016), however they do not provide any citations with evidence for this.
Maintenance of microanatomy and biomolecules over time depends on the polymer used. It is unclear how the most common polymer used in plastination, silicone, solidifies in a heterogenous tissue. Given the state of the evidence, it seems difficult to make certain claims can be made about the maintenance of microstructure over time in silicone embedded plastinated tissue during long-term storage at ambient temperature.
Mechanistically, we can make some guesses. Certainly there will not be any hydrolysis as long as the water has been fully removed. Diffusion should not be significant if the embedded silicone polymer has truly solidifed via crystallization. Given that they both seem to solidify via crystallization in their pure forms, silicone embedded tissue may degrade over time in a similar way to paraffin embedded tissue. So likely relatively slowly, given the good data on preservation of paraffin embedded tissues. However, this is speculative extrapolation and requires further study.
| Embedding Agent | Advantages | Disadvantages |
|---|---|---|
| Paraffin | - Widely available and relatively cheap - Extensive database on use and preservation over time - Can preserve protein antigens for a long time - With a moderate amount of sectioning, can be performed on the whole human brain |
- Requires lipid removal - Questionable degree of ultrastructure preservation - No data available for connectomics |
| Polyester Wax | - Preserves tissue structure better than paraffin - Lower melting temperature, less heat/dehydration-induced shrinkage |
- Not commonly used - Requires further study |
| Celloidin | - Bypasses need for heating - Minimal shrinkage - Good for dense tissue like brain |
- Very long infiltration times - Blocks must be maintained in liquid ethanol - Flammable vapor |
| Epoxies | - Excellent ultrastructure preservation - Stable storage of embedded samples |
- High viscosity, long/high temp infiltration - Often requires the use of osmium for infiltration - Can cause loss of antigenicity - No protocols for large specimens the size of the human brain without sectioning |
| Acrylates (General) | - Can tolerate small amounts of water - Can be polymerized at low temperatures |
- Difficult to polymerize via UV in large samples - Generally worse ultrastructure preservation compared to epoxies |
| Glycol Methacrylate | - Good tissue infiltration, even for large samples - Better morphological preservation than paraffin, celloidin, or frozen sections - Can be polymerized at cold/ambient temps - Adaptable to different types of histochemical stains |
- Poor ultrastructure preservation - Causes considerable shrinkage and lipid extraction - Unclear long-term stability at ambient temperature |
| Methyl Methacrylate | - Good tissue infiltration - Better morphological preservation than paraffin - Can be used for EM - Can polymerize at low temps |
- Can result in excessive heat during polymerization - May require toxic catalyst - Questions about degree of ultrastructure preservation |
| Lowicryl | - Can be photopolymerized by UV at low temps - Low viscosity allowing fast/deep infiltration - Moderate hydrophobicity preserving proteins/antigens - High quality ultrastructure and immunolabeling |
- Relatively poor tissue infiltration, especially in lipid-rich
brain - Needs to be stored in desiccator |
| Plastination | - Can preserve whole organs/bodies - Silicone widely used, but epoxy/polyester possible - Gross anatomy well preserved |
- Microanatomy preservation not well characterized - Significant shrinkage of brain tissue - Long-term microanatomy stability uncertain |
Currently, discussing embedding options for brain preservation is mostly stamp collecting, without too much underlying theory. Paraffin, the most widely used embedding medium, has advantages such as low cost and long-term preservation of protein antigens, but require harsh processing conditions and therefore may not provide sufficient ultrastructure preservation. Epoxy resins excel in ultrastructure preservation but can be difficult to infiltrate into large samples and may compromise antigenicity. Acrylate resins, like glycol methacrylate, methyl methacrylate, and Lowicryl, offer benefits such as low-temperature polymerization and improved biomolecular preservation, but each has its own drawbacks, such as poor ultrastructure preservation or difficulty in infiltrating large, lipid-rich samples like the brain. Plastination techniques, which can preserve whole organs or bodies, show promise for embedding large specimens but require further research to characterize and optimize the quality of microanatomy preservation. Overall, while embedding approaches has potential for brain preservation, more research is needed to optimize techniques that maximize long-term stability while minimizing damage. The most practical option available today would probably be paraffin embedding, but it would require sectioning the brain before the embedding procedure could be performed.