How to Properly Freeze Cells for Cryopreservation

Freezing cells successfully comes down to controlling three things: the rate at which temperature drops, the concentration and type of cryoprotectant shielding the cells, and how quickly and carefully you warm them back up. Get any one of these wrong and ice crystals form inside the cell, membranes rupture, or toxic solute concentrations build up and kill the very cells you are trying to preserve. The standard recipe for most mammalian cell lines is straightforward, but the details matter more than most protocols let on, and the “standard” recipe is not universal.

Why Cells Die During Freezing

Understanding what actually kills cells during freezing helps explain every step of a good protocol. When ice begins to form in the liquid surrounding cells, it locks up pure water and leaves behind an increasingly concentrated solution of salts and other solutes. Cells sitting in those channels of concentrated fluid lose water osmotically, shrinking as the solution around them grows saltier. If cooling is slow enough, this dehydration is controlled and the cell survives in a shrunken state. If cooling is too fast, the cell does not lose water quickly enough, the interior remains supercooled, and ice crystals nucleate inside the cytoplasm.

Intracellular ice is almost always lethal. Research on the mechanism suggests that it is not simply the physical puncture of organelles by ice crystals. Instead, the rapid osmotic pressure difference across the cell membrane during freezing can cause the membrane itself to rupture, and once that barrier is breached, extracellular ice propagates inward.¹ Even without intracellular ice, cells that are cooled too slowly can be damaged by prolonged exposure to high solute concentrations or by excessive shrinkage. Each cell type has its own optimal cooling curve, a sweet spot between too fast and too slow.²

Choosing a Cryoprotectant

Cryoprotectants fall into two broad camps. Penetrating agents like DMSO and glycerol cross the cell membrane and work from the inside. Non-penetrating agents like trehalose and sucrose stay outside the cell and stabilize the membrane surface. In practice, DMSO at around 5 to 10 percent remains the default for most mammalian cell work. It depresses the freezing point of water and promotes the formation of a glass-like solid state rather than crystalline ice, a process called vitrification. Molecular simulations have shown that DMSO broadens the glass transition of water, making the resulting glassy state more thermodynamically stable and less likely to nucleate ice as the temperature drops.³

Glycerol was the original cryoprotectant, first shown to protect sperm cells during freezing in 1949, a discovery that launched the modern era of cryobiology.⁴ It is still widely used for bacteria and red blood cells but has largely been replaced by DMSO for most nucleated mammalian cells because DMSO penetrates membranes faster and works well at lower concentrations.

Sugar-based cryoprotectants like trehalose are attractive because they are nontoxic, but they face a practical hurdle: trehalose does not cross cell membranes easily on its own.⁵ Researchers have shown, however, that combining a sugar with a reduced amount of DMSO can match the performance of the standard 10 percent DMSO protocol. In one study on cord blood stem cells, 30 millimolar trehalose with just 2.5 percent DMSO produced post-thaw viability and clonogenic potential comparable to 10 percent DMSO alone, which is a quarter of the standard DMSO concentration.⁶ Reducing DMSO matters because DMSO is not benign.

DMSO Concentration and Toxicity

DMSO is effective, but it is also the source of most of the practical headaches in cryopreservation. It is toxic to cells at room temperature. Dose-response testing on cord blood showed that 40 percent DMSO eliminated all viable progenitor cells, while the optimal range for cryopreservation was 7.5 to 10 percent, with measurable harm outside that window.⁷ The same study found minimal toxic effect when freezing was delayed up to one hour after adding 10 percent DMSO, which gives you a practical working window. After thawing, washing out the DMSO was better than simple dilution when cells sat at room temperature for longer periods.

For clinical applications like stem cell transplants, DMSO toxicity in patients is a real concern. Patients receiving cryopreserved grafts can experience nausea, cardiovascular effects, and other reactions attributed to the DMSO in the infused product.⁸ This has driven a serious search for alternatives or ways to minimize the DMSO concentration while maintaining cell survival, including the sugar-combination approaches described above.

The practical takeaway: add DMSO to cells cold (keep everything on ice or at 4°C), work efficiently but without panic (you have roughly an hour), and after thawing, get the DMSO out as soon as your workflow allows.

Controlling the Cooling Rate

For most mammalian cells, the target cooling rate is about negative one degree Celsius per minute from roughly 0°C down to at least negative 40°C, ideally to negative 80°C, before transferring to long-term storage. This rate is slow enough for water to leave the cell osmotically but fast enough that the cell does not sit in damaging high-solute conditions for too long.

Controlled-rate freezers achieve this precisely using programmable liquid nitrogen injection. They are the gold standard in clinical and industrial settings, and measurements confirm they can hold the rate at about negative one degree per minute with high consistency.⁹ For most research labs, the more affordable option is an isopropanol-filled container (often called a “Mr. Frosty”) placed in a negative 80°C freezer. These containers approximate the one-degree-per-minute rate reasonably well, though the curve is less perfectly controlled, and the alcohol should be replaced regularly.

One common mistake is leaving cryovials at negative 80°C indefinitely. That temperature is fine for a day or two as a waypoint, but for long-term storage, cells should be moved to liquid nitrogen storage (negative 196°C in liquid phase, or roughly negative 150°C and below in vapor phase). At negative 80°C, some molecular motion and ice recrystallization still occur, and viability slowly degrades over months.

Vitrification as an Alternative to Slow Freezing

Not all cell types freeze well with the slow-cooling approach. Oocytes and embryos, for instance, are large cells with high water content and are sensitive to ice crystal formation. Vitrification takes the opposite approach: instead of controlling the rate of ice formation, it eliminates ice formation entirely by cooling so fast that water solidifies into an amorphous glass. This requires very high concentrations of cryoprotectants and plunging samples directly into liquid nitrogen, achieving cooling rates of thousands of degrees per minute.

In reproductive medicine, vitrification has largely replaced slow freezing for oocytes and embryos. A comparison of the two methods for human cleavage-stage embryos found that vitrification produced a survival rate of about 97 percent compared to 83 percent for slow freezing, with clinical pregnancy rates roughly double in the vitrified group.¹⁰ A separate study comparing the two methods for oocytes found that DNA integrity was preserved by both, but vitrification delivered higher survival rates and better spindle preservation.¹¹

There is a tradeoff, though. The high cryoprotectant concentrations needed for vitrification can stress gametes and embryos, and the technique demands precise, fast handling that leaves less room for error.¹² For most standard cell culture work in research labs, slow freezing with DMSO remains simpler and adequate.

Storage Conditions and Contamination Risk

Long-term storage at temperatures below negative 150°C essentially stops all biological activity, including degradation. The two options are immersion in liquid nitrogen (negative 196°C) or storage in the vapor phase just above the liquid (typically negative 150°C to negative 190°C depending on the tank design and fill level).

Liquid-phase immersion keeps the temperature more stable, but it raises a contamination concern. If a cryovial has a hairline crack or an imperfect seal, liquid nitrogen can seep in during storage and potentially carry microbes between samples. Studies on microbial contamination in liquid nitrogen tanks have found that, in practice, liquid nitrogen is rarely the source of major contamination, but the risk is real enough that clinical facilities care about it.¹³ Vapor-phase storage avoids this by keeping samples out of direct contact with the liquid. Research on vitrified human oocytes stored in the vapor phase has shown that this approach maintains the oocytes’ developmental potential while preventing the cross-contamination risk.¹⁴

Ultra-low-temperature mechanical freezers (negative 80°C) are convenient but are not truly long-term storage solutions. Temperature fluctuations from door openings can be substantial, with peak variations of 10 to 15°C measured in the most exposed compartments. During a power failure, a negative 80°C freezer takes only a few hours to warm past negative 50°C.¹⁵ If your cells live in a negative 80°C freezer, an alarm system and a backup plan are not optional.

Thawing and Post-Thaw Handling

Thawing needs to be fast. The standard method is to swirl the cryovial in a 37°C water bath until a small sliver of ice remains, then immediately dilute and remove the cryoprotectant. The reasoning mirrors the freezing step: as ice melts, it releases pure water back into the surrounding solution, creating transient osmotic imbalances. A slow thaw prolongs the time cells spend at intermediate temperatures where ice can recrystallize into larger, more damaging crystals. Speed limits this exposure.

What you do in the first hours after thawing matters as much as the thaw itself. Post-thaw apoptosis is a well-documented phenomenon. Flow cytometry studies on T cells frozen with 5 or 10 percent DMSO showed roughly 20 percent early apoptotic cells at four hours after thawing, declining to about 10 percent by 24 hours.¹⁶ That wave of programmed cell death means the viability you measure 10 minutes after thawing will overestimate how many cells you actually have the next day. Accounting for this delayed loss when calculating how many vials to thaw can save a frustrating weekend of restarting cultures.

Viability itself should be measured with more than one method. Common assays include trypan blue exclusion (manual or automated), flow cytometry with dyes like 7-AAD or propidium iodide, and automated systems using acridine orange and propidium iodide staining.¹⁷ Trypan blue is the quickest and cheapest, but it only tells you whether the membrane is intact at that moment, not whether the cell is heading toward apoptosis. Flow cytometry with annexin V staining captures those early apoptotic cells that trypan blue misses.

Special Considerations for Stem Cells

Human pluripotent stem cells, whether embryonic or induced, are notoriously fragile during cryopreservation. The problem is partly that dissociated single stem cells tend to undergo a type of programmed cell death triggered by loss of cell-cell contact, called anoikis. A widely adopted solution is the ROCK inhibitor Y-27632, which blocks the signaling pathway that leads to this dissociation-induced death. Research has shown that Y-27632 significantly improves both the recovery of cryopreserved human induced pluripotent stem cells and their subsequent growth.¹⁸

Timing of ROCK inhibitor exposure matters. Adding it only to the freezing solution is not enough. The real benefit comes when you include the inhibitor in both the culture medium before freezing and the medium the cells land in after thawing.¹⁹ Newer approaches are exploring ice recrystallization inhibitors, small molecules that limit ice crystal growth during thawing. One study found that an ice recrystallization inhibitor called 2FA reduced post-thaw apoptosis in iPSCs compared to a standard commercial cryopreservation medium, with the peak apoptosis difference appearing between 8 and 14 days after thawing, not immediately.²⁰

Freezing Immune Cells for Clinical Therapies

The explosion of cell-based immunotherapies, particularly CAR-T cell therapy, has put cryopreservation under intense clinical scrutiny. Manufacturing a patient’s CAR-T cells takes days to weeks, and the ability to freeze the starting material or the final product is essential for logistics. The question is whether freezing degrades the product.

Studies on CAR-T manufacturing have found that cryopreserving the starting blood cells does not significantly alter the key quality attributes of the final product, including how many times the cells expanded, how efficiently they were transduced with the CAR construct, or the ratio of CD4 to CD8 T cells.²¹ Clinical outcomes back this up. In one series of 118 treated patients, about 73 percent achieved an objective response overall, with no significant difference between patients whose CAR-T products started from fresh versus frozen blood cells.²²

That said, the picture is not entirely clean. Cryo-thawed CAR-T cells do show elevated markers of early apoptosis and upregulated pathways related to mitochondrial dysfunction compared to fresh counterparts.²¹ Whether this subtle molecular damage translates to meaningful clinical differences is still being studied. For now, cryopreservation is considered a viable manufacturing strategy for CAR-T therapy, but the field is actively working to understand and minimize these cryoinjury signatures.

Serum-Free Freezing Media

Traditional cryopreservation media contain fetal bovine serum (FBS) as a protein source that helps stabilize cell membranes during freezing. For clinical-grade and regulatory-compliant work, animal-derived serum is a problem because it introduces batch-to-batch variability and the risk of adventitious agents. Several serum-free formulations have been developed that perform comparably to FBS-based media.

Early work showed that cells cultured in serum-free conditions could be frozen in a medium based on their serum-free growth medium plus 10 percent DMSO, with or without a small amount of methylcellulose, and thawed with no measurable cell loss compared to the standard FBS-containing freeze medium.²³ More creative approaches have used the silk protein sericin as a serum substitute, producing a defined freezing medium of sericin, maltose, proline, glutamine, and 10 percent DMSO that preserved cells as effectively as the FBS control.²⁴ Fully GMP-compliant serum-free cryomedia that maintain viability, recovery, and antigen-specific T-cell responses have also been validated for clinical applications like immune monitoring.²⁵

Freezing Bacteria and Other Microbes

Microbial cryopreservation follows related principles but with different details. Most bacteria are frozen in glycerol (commonly 15 to 25 percent) rather than DMSO, stored at negative 80°C or in liquid nitrogen. Glycerol is cheap, effective, and well tolerated by most bacterial species. For anaerobic gut microbes, which are particularly sensitive to oxygen exposure during handling, adding sugars like sucrose or inulin alongside glycerol has been explored to improve survival.²⁶

An alternative for microbes is lyophilization (freeze-drying), which produces shelf-stable cultures that do not require freezer space. However, liquid nitrogen freezing has shown clear advantages over lyophilization for many species. A direct comparison with root nodule bacteria found that liquid nitrogen freezing preserved both viability and symbiotic function more reliably than freeze-drying, with less phage induction in lysogenic strains.²⁷

What Nature Figured Out First

Some of the most creative ideas in cryopreservation come from studying organisms that survive freezing in the wild. Wood frogs in subarctic Alaska endure body temperatures well below negative 10°C, with up to two-thirds of their body water turning to ice. They survive by flooding their cells with cryoprotectants, primarily glucose and urea, which limit intracellular ice formation and stabilize proteins during dehydration. Subarctic populations show dramatically higher freeze tolerance than frogs from more temperate areas, tolerating experimental freezing to temperatures at least 10 to 13 degrees below the lethal limits for their southern relatives.²⁸

This natural freeze tolerance has inspired research into antifreeze proteins and synthetic ice-controlling molecules. A range of materials have been explored, from antifreeze proteins to synthetic polymers, nanomaterials, and hydrogels, all aimed at inhibiting ice nucleation or recrystallization.²⁹ The results have been humbling in some cases. Different classes of antifreeze proteins with varying ice recrystallization inhibition activity produced very similar cell recovery rates (around 20 percent for red blood cells), suggesting that the ability to slow ice crystal growth alone is not enough to explain how these proteins protect cells.³⁰ The mechanisms are clearly more complex than just keeping ice crystals small, and figuring out what else these molecules do remains an active and genuinely open area of research.