DMSO Cryopreservation: The Protocol and Protective Mechanism

Dimethyl sulfoxide, universally known as DMSO, has been the dominant cryoprotectant for freezing living cells since the earliest days of cryobiology. Its protective mechanism works on two fronts: DMSO penetrates cells and disrupts the way water molecules organize into ice crystals, while also reducing the osmotic stress that would otherwise crush or burst cells during freezing. The standard protocol calls for suspending cells in roughly 5–10% DMSO, cooling them at a controlled rate of about −1°C per minute, and transferring them to ultra-cold storage. But DMSO is a compromise molecule. The same chemical properties that let it slip through cell membranes and interfere with ice also make it toxic to cells at higher concentrations and longer exposure times, a tension that drives much of the ongoing research around cryopreservation.

How DMSO Protects Cells From Freezing Damage

When cells freeze without any protection, two things kill them. First, ice crystals that form outside the cell draw water out through the membrane by osmosis, dehydrating and shrinking the cell to a lethal degree. Second, if cooling is too fast for water to leave gradually, ice crystals nucleate inside the cell, physically puncturing membranes and organelles. DMSO addresses both problems because it is one of the few molecules small and polar enough to cross the cell membrane freely, entering the cytoplasm and establishing a presence on both sides of the lipid bilayer.

At the molecular level, DMSO’s protective power comes from how it interacts with water. Water molecules in a pure solution naturally arrange themselves into a tetrahedral hydrogen-bond network, which is the structural scaffolding of ice. DMSO’s strong polar character and its lack of hydrogen-bond donors allow it to wedge between water molecules, disrupting that orderly network and inhibiting ice-crystal growth.¹ In practical terms, DMSO disorders the water in and around cells so thoroughly that, instead of forming large damaging crystals, the solution either forms only tiny, relatively harmless ice crystals or, at high enough concentrations, skips crystallization entirely and solidifies into a glass-like amorphous state.

That glassy transition happens at around −120°C (give or take about 10 degrees) for typical cryoprotectant solutions.² Below this glass transition temperature, molecular movement slows to a near standstill, essentially pausing biological time. This is why cryopreserved cells can remain viable for years in liquid nitrogen storage at −196°C.

DMSO also helps with osmotic balance. During freezing, as pure water crystallizes out of the solution, the remaining liquid becomes increasingly concentrated with salts and other solutes. This “freeze concentration” effect can generate enormous osmotic pressure that damages cell membranes. Because DMSO crosses into the cell, it balances solute concentration on both sides of the membrane, alleviating that osmotic crush.³

What DMSO Does to Cell Membranes

DMSO’s ability to cross the cell membrane is central to its usefulness, but the way it interacts with lipid bilayers is more complex than simply slipping through. At low concentrations, DMSO thins the membrane and makes its hydrophobic core more fluid. At higher concentrations, it can actually induce transient water pores in the membrane. Push the concentration higher still and individual lipid molecules start peeling away from the bilayer, eventually destroying it.⁴ This concentration-dependent behavior is why protocol design matters so much: you need enough DMSO to protect against ice, but not so much that the membrane itself is compromised.

These membrane effects are not uniform across all cell types. DMSO alters the mechanical properties of membranes in a way that depends on local lipid composition and structure, meaning a red blood cell membrane and a liver cell membrane may respond quite differently to the same DMSO concentration. Even at relatively low concentrations of about 3% by volume, DMSO can increase membrane permeability enough to let small molecules like ATP leak out.⁵ This variability is one reason there is no single perfect DMSO concentration for all applications. Different cell types require their own optimized protocols.

The Standard Freezing Protocol

Despite decades of incremental refinement, the basic architecture of a DMSO cryopreservation protocol has remained remarkably stable. The steps, in broad strokes, look like this:

DMSO addition: Cells are suspended in a cryopreservation medium containing DMSO, typically at a final concentration of 5–10%. The DMSO is usually added gradually and at low temperature (around 4°C) to minimize the osmotic shock and toxicity that come with sudden exposure.
Controlled cooling: The cell suspension is cooled at a rate of roughly −1°C per minute. This rate is slow enough for water to leave cells gradually by osmosis, preventing intracellular ice formation, but fast enough that cells are not exposed to concentrated extracellular solutes for too long.
Transfer to deep storage: Once the sample reaches about −80°C, it is typically transferred to liquid nitrogen (−196°C) for long-term storage. Some protocols hold samples at −80°C if only short-term storage is needed, though viability declines over time at that temperature.

Achieving that −1°C-per-minute cooling rate does not require expensive programmable freezers for every lab. A common low-cost approach uses a “box-in-box” device placed in a −80°C mechanical freezer, which produces cooling rates in the range of −1 to −3.5°C per minute depending on sample format.⁶ Samples in bags tend to cool more uniformly at around −1°C per minute, while samples in vials can cool somewhat faster, but both formats work in practice.

The question of whether 5% or 10% DMSO is better has been studied extensively, and the trend in recent years has been toward the lower concentration. A systematic review and meta-analysis of controlled clinical studies found that stem cells cryopreserved with 5% DMSO had greater post-thaw viability of key cell populations compared to 10% DMSO, and patients receiving the lower-concentration products experienced fewer adverse side effects.⁷ Work on porcine mesenchymal stem cells similarly found that 5% DMSO yielded survival comparable to unfrozen controls, with cell survival inversely proportional to DMSO concentration as it climbed to 10% and 20%.⁸

The Toxicity Problem

DMSO is paradoxical: the concentrations needed to protect cells from ice are also high enough to damage them through chemical toxicity. This is not a subtle theoretical concern; it plays out in real cellular damage. In cultured astrocytes (a type of brain cell), exposure to just 1% DMSO for 24 hours did not kill cells outright but caused mitochondrial swelling, impaired the mitochondrial membrane potential, triggered reactive oxygen species production, and set off an apoptotic cascade. At 5%, the effects were significantly worse, with clear inhibition of cell viability and promotion of programmed cell death.⁹

For hematopoietic stem cells, the picture is further complicated by the freeze-thaw cycle itself. Comparing the toxicity of 5% DMSO on cells that were never frozen versus cells that had been frozen and thawed, the post-thaw group showed greater sensitivity to DMSO damage. This likely reflects thermal instability of membrane-bound enzymes or increased membrane permeability from the freezing process, amplifying DMSO’s cytotoxic effects on cells that are already stressed.¹⁰ This is why speed matters during thawing and post-thaw handling: the longer thawed cells sit in DMSO-containing medium at room or body temperature, the more damage accumulates.

The toxicity profile scales with concentration, temperature, and exposure duration. For whole-organ cryopreservation, where DMSO must be perfused throughout a large tissue volume, these constraints become severe. The time needed to achieve adequate CPA penetration through every layer of tissue pushes up against the exposure window where toxicity becomes unacceptable.¹¹

Post-Thaw DMSO Removal

Because DMSO continues to damage cells after thawing, many clinical protocols include a washing step to remove it before cells are infused into a patient. The process typically involves diluting the thawed product with a compatible solution and centrifuging to separate the cells from the DMSO-laden supernatant. In one described protocol for autologous blood progenitor cells, the thawed bag contents were mixed with a hydroxyethyl starch and anticoagulant solution, centrifuged, and then the supernatant was removed, with the entire DMSO removal process taking about an hour per bag.¹²

Dilution itself addresses another problem: the extreme osmolarity that builds up in the frozen product. One study measured osmolarity jumping from about 330 mOsm/L before freezing to roughly 1,460 mOsm/L immediately after thawing, dropping to about 810 mOsm/L after a dilution step.¹³ That post-thaw spike in osmolarity, if left uncorrected, creates additional osmotic stress on already fragile cells.

Whether DMSO removal actually improves clinical outcomes is a separate question, and the evidence is mixed. A study comparing autologous stem cell transplant outcomes in patients whose grafts had DMSO removed versus those whose grafts were infused with DMSO still present found that the DMSO-containing grafts actually showed faster platelet and red blood cell engraftment. DMSO removal was not associated with fewer infusion-related adverse events.¹⁴ This could reflect cell losses during the washing process itself, or it could mean that for certain applications the side effects of infusing a moderate amount of DMSO are less damaging than the mechanical toll of an extra centrifugation step. The decision to wash or not remains institution-specific.

Clinical Side Effects of DMSO Infusion

When cryopreserved stem cell products are infused directly into patients without DMSO removal, side effects are common but usually mild. The most recognizable is a garlic-like taste and body odor that appears within minutes of infusion and can last a day or two, caused by DMSO metabolites exhaled through the lungs. Nausea, vomiting, flushing, and transient drops in blood pressure also occur. Serious reactions are rare, but anaphylaxis has been reported.¹⁵ Pre-medication with antihistamines and slow infusion rates are standard precautions.

Ice Recrystallization, the Damage DMSO Does Not Prevent

One common misconception is that DMSO provides complete protection against all forms of ice damage. It does not. DMSO and glycerol work through several mechanisms, including colligative depression of the freezing point and direct interaction with water, but at the concentrations used in standard protocols they do not effectively control ice recrystallization.¹⁶ Recrystallization is the process by which small, relatively harmless ice crystals merge into larger ones during storage or, especially, during warming. This can happen whenever a frozen sample briefly warms above a critical threshold and then refreezes, something that occurs, for example, when a rack of cryovials is temporarily pulled out of a liquid nitrogen dewar to retrieve a single sample. The inability to control recrystallization is a distinct vulnerability that researchers are now addressing with small-molecule ice recrystallization inhibitors as supplements to DMSO.

Supplementing DMSO With Sugars and Other Additives

Because DMSO alone does not provide full protection and has a toxicity ceiling, a major area of optimization involves combining it with non-penetrating cryoprotectants, particularly sugars. Trehalose, a disaccharide found naturally in organisms that survive extreme dehydration and freezing, has attracted the most attention. Trehalose is too large to cross cell membranes on its own, so it works from the outside, stabilizing cell-surface proteins and forming a protective glassy matrix in the extracellular space.¹⁷

The practical benefit of adding trehalose is that it can allow you to reduce the DMSO concentration while maintaining or improving viability. In one study on primary rat hepatocytes, a very low concentration of trehalose supplemented with DMSO yielded about 70% more viable cells after thawing compared to DMSO alone.¹⁸ Work on liver-derived cell lines found that adding saccharides like sucrose and trehalose to a DMSO-based medium enabled DMSO reduction, replaced the need for serum in the freezing medium, and improved the functional capacity of the thawed cells.¹⁹

Beyond improved survival numbers, multi-component cryopreservation solutions may also have benefits at the genomic level. Mesenchymal stromal cells cryopreserved in optimized multicomponent solutions showed different patterns of DNA hydroxymethylation compared to cells frozen in DMSO alone, along with upregulation of genes involved in cytoprotection and extracellular matrix signaling.²⁰ In other words, the freezing medium you choose does not just affect whether cells survive, but may influence how they behave after thawing.

Pushing Toward DMSO-Free Protocols

Given DMSO’s toxicity limitations, there is active research into eliminating it entirely for certain applications. One approach draws inspiration from nature: some cold-tolerant organisms produce antifreeze proteins that do not prevent all ice formation but instead control ice crystal size and shape to minimize damage. Synthetic mimics of these proteins, particularly polyvinyl alcohol (PVA), have been tested as DMSO replacements. A recent study used PVA alone to cryopreserve both mouse and human oocytes, controlling ice growth and morphology rather than trying to suppress ice formation entirely.²¹

However, for many cell types, DMSO remains hard to beat. In mouse oocyte vitrification experiments, removing DMSO from the vitrification solution significantly reduced cryo-survival rates and blastocyst development compared to protocols that included it, even though early embryo development looked similar between the groups.²² This suggests that for reproductive cells, DMSO provides a specific benefit in maintaining developmental competence that current alternatives have not fully replicated.

The Whole-Organ Frontier

Cryopreserving individual cells in suspension is one thing. Preserving an intact organ like a kidney or heart is an entirely different challenge. The goal for whole-organ cryopreservation is vitrification: loading enough cryoprotectant into every cell in the organ to prevent any ice formation at all. This requires perfusing the organ with very high concentrations of CPA through its vasculature, reaching every tissue layer uniformly before the toxicity clock runs out.²³

The fundamental problem is scale. A suspension of stem cells can be mixed with DMSO and cooled in seconds to minutes. An organ has thick tissue layers, varying cell types, and a vascular tree that must be used to deliver the cryoprotectant. The perfusion time needed to achieve adequate concentration throughout all tissue is long enough that outer layers are already suffering toxicity damage before inner layers are adequately protected. Researchers are using computational modeling to optimize perfusion protocols, stepping up CPA concentration gradually while keeping the organ cold to slow toxic effects, but no one has yet demonstrated reliable vitrification and successful rewarming of a transplantable human organ.

How Long Can Cells Be Stored?

In liquid nitrogen at −196°C, cryopreserved cells can theoretically remain viable indefinitely because molecular activity essentially stops. In practice, most clinical and research storage uses liquid nitrogen and achieves excellent long-term results. But some settings rely on −80°C mechanical freezers, which are cheaper and more widely available, and here the story is less encouraging.

A study tracking hematopoietic progenitor cells stored at −80°C with 5% DMSO found a progressive decline in viability over time. At six months, viability was still around 80% and colony-forming capacity was reasonably preserved. But viability dropped to about 32% by 31 months, and the ability to form colonies disappeared entirely after 24 months.²⁴ The glass transition temperature for typical DMSO solutions sits near −120°C, meaning that at −80°C the sample is actually above the glass transition: there is still residual molecular mobility, slow ice recrystallization can occur, and cells gradually deteriorate. For anything beyond short-term holding, liquid nitrogen storage is strongly preferred.

Quality Control Beyond Viability

A common pitfall in cryopreservation work is equating “cell viability” (the percentage of cells that exclude a dye or show membrane integrity) with actual functional quality. A cell can pass a viability assay and still have compromised mitochondria, altered gene expression, or reduced engraftment potential. Quality control for clinical-grade cryopreserved products goes well beyond counting live cells. Microbiological testing for bacterial, fungal, mycoplasma, and endotoxin contamination is a critical step before clinical application, as contamination is responsible for a significant share of adverse reactions.²⁵

Regulatory frameworks for cell therapy products generally require documentation of the cryopreservation protocol, the DMSO concentration used, the cooling and storage conditions, post-thaw viability data, sterility testing results, and in many cases functional assays specific to the cell type. The lack of a single universal standard reflects the diversity of cell types and clinical applications: what counts as adequate quality for a bone marrow transplant product is different from what matters for a research-grade cell line or a reproductive cell bank. Labs navigating this space need to validate their specific protocol for their specific cell type, rather than assuming a generic DMSO recipe will work across the board.