Optimization of slow freeze protocol for 3D cell structures using brain organoids and chondrospheres as models

Full Text

Background

Cryopreservation is a well-established method for the long-term storage of biomaterials. The underlying principle involves the freezing of cellular material at cryogenic temperatures ranging from −196 °C to −150 °C. The primary factors contributing to cell destruction during freezing and thawing are the formation of ice crystals from both intracellular and extracellular water, which cause the rupture of the cell membrane and an increase in salt concentrations in the non-frozen fraction. The general principle underlying the freezing of cell cultures, tissues, and organoids consists of supplementing the freezing medium with a cryoprotective agent. This agent penetrates the cell, binding intracellular water and reducing its crystallization, thereby maintaining a constant salt concentration. Dimethyl sulfoxide (DMSO, Me2SO), ethylene glycol, and propanediol are commonly used as cryoprotectants. In some cases, sucrose is also used. Notably, each of these cryoprotectants exhibits inherent cytotoxicity [1–3]. In this regard, the freezing protocol and the specific cell type need to be considered to choose the optimal cryoprotectant concentration. The concentration of toxic penetrating cryoprotectants, such as DMSO, ethylene glycol, and propanediol, can be reduced by using non-penetrating and less toxic cryoprotectants, such as disaccharides or macromolecules [1, 4].

All cryopreservation techniques can be broadly divided into two categories: slow freezing and vitrification, each with its own set of benefits and drawbacks. The process of slow freezing involves a gradual decrease in temperature at −0.3 °C to −1 °C per minute. Slow freeze protocols rely on relatively low concentrations of cryoprotectant agents, typically ranging from 7% to 10% by volume. Thawing is performed by adjusting the temperature to 36–37 °C within an average time of 3–5 minutes. Simplicity and practicality are among the key advantages associated with slow freezing. The primary risk factor associated with this procedure is osmotic shock experienced by cells during thawing. The addition of the Rho kinase inhibitor Y-27632 (ROCK inhibitor, Rho-associated protein kinase inhibitor, Ri) to the cryomedium has been shown to increase the post-thaw survival and viability of numerous cell types [5, 6].

The vitrification process is characterized by high concentrations of cryoprotectants (15%–30%), rapid cooling and thawing, with these processes occurring within seconds. During vitrification with rapid cooling, the highly concentrated cryoprotectant reaches a temperature at which it forms a glass-like structure and exhibits properties that allow it to avoid crystallization. This prevents osmotic shock and damage to the membrane of the frozen cell. Vitrification has been demonstrated to be a safer technique for frozen objects and is a widely used for freezing embryos. Among its notable drawbacks is its high labor intensity compared to slow freezing. The process involves the sequential transfer of frozen objects to various buffers containing increasing concentrations of cryoprotectants. The final step occurs with the immediate cooling to cryogenic temperatures after the objects are placed in the cryomedium [7].

Both cryopreservation techniques are applicable for freezing cell suspensions and more complex objects, such as tissue fragments and organoids. Organoids are three-dimensional multicellular structures that can be considered differentiated derivatives of stem cells or organ progenitor cells. Organoids are used for disease modeling, studies in regenerative medicine, and assessing various therapeutic agents [8, 9]. In contrast to cell suspensions, organoids have a number of unique characteristics that must be considered when developing a cryopreservation protocol. These include organized intercellular interactions, varying cell availability for solution components, and differences in cryoprotective and osmotic tolerance among different cell types within the organoid [10]. The optimization of cryopreservation conditions and the selection of a universal cryoprotective solution will ensure the long-term storage of organoids and significantly enhance their utility for research and clinical applications.

Currently, there is a paucity of available evidence regarding the cryopreservation of organoids. The slow freeze protocol has been successfully applied to intestinal and hepatic organoids derived from intestinal tissue and epithelial stem cells, respectively, and retinal organoids derived from induced pluripotent stem cells (iPSCs) [11, 12]. Additionally, there are documented studies focusing on the cryopreservation of cardiac organoids [13–15]. As demonstrated by numerous studies, vitrification emerges as a more promising approach for the cryopreservation of organoids compared to slow freeze techniques [3, 7, 16]. Such evidence has been collected for islets of Langerhans, spheroid cultures of multipotent mesenchymal stromal cells, and lung cancer organoids [7, 16, 17].

However, the available data on the cryopreservation of neural organoids remains limited. A slow freeze protocol has been described as a viable option for the cryopreservation of glioblastoma organoids, with the culture medium containing 10% DMSO as a cryoprotectant [18]. However, tumor cells demonstrate a higher viability than healthy cells, thereby raising concerns about the efficacy of this technique for the cryopreservation of differentiated, untransformed cells. The use of the slow freeze protocol in a protein-free medium containing 10% DMSO has also been reported for aggregates of neural progenitors derived from pluripotent stem cells [19]. Recent studies have demonstrated the feasibility of slow cryopreservation of neurospheres derived from the enteric nervous system and neural organoids derived from embryonic stem cells (ESCs) in various media with 10% DMSO [20, 21]. DMSO exerts a dual effect, demonstrating both cryoprotective and cytotoxic behavior [22]. Current publications lack the evidence that would allow assessing the potential effects of a decrease in the DMSO concentration on the post-cryopreservation survival of organoids. A decrease in the DMSO concentration during the cryopreservation of certain cell types has been shown to have a positive effect on their post-thaw survival [23, 24]. Furthermore, even low concentrations of DMSO have demonstrated a cytotoxic effect on neural cells [25, 26]. Decreasing the concentration of the substance during cryopreservation could potentially enhance the post-thaw number and viability of organoids. The optimal stage of neural organoid differentiation for the cryopreservation remains to be elucidated. Mature neural organoids have been the prevailing choice for freezing in research [20, 21]. However, recent findings suggest that young organoids exhibit enhanced survival, potentially attributable to the presence of immature, actively proliferating cells [27, 28]. Finally, further research is required to refine the protocol for freezing organoids derived from neurally differentiated iPSCs. Midbrain organoids are of particular interest as a model system for the investigation of neurodegenerative diseases, such as Parkinson disease, amyotrophic lateral sclerosis, and Niemann–Pick disease type C [29–31].

Chondrospheres are spheroids originating from iPSC-derived chondrocytes, multipotent mesenchymal stromal cells, ESCs, or chondrocytes, and the secreted matrix. Chondrospheres may be used to repair severe cartilage defects [32]. The biobanking and transportation of chondrospheres to distant medical facilities require the long-term storage at low temperatures, as their shelf life is only two days [33]. The data on cryopreservation of chondrospheres remain limited. The slow freezing of native cartilage biopsy specimens has been previously reported [34]. The study demonstrated that the cryopreservation of biopsy specimens in fetal bovine serum (FBS) containing 10% DMSO had no significant impact on their viability. However, the expression of chondrocyte-specific markers or matrix composition was beyond the scope of the study.

Another study examined the cryopreservation of ESC-derived chondrospheres in six various commercial media [35]. The authors evaluated the viability of organoids at 4 hours after thawing, yet did not analyze the post-thaw expression of matrix proteins. High concentrations of DMSO are known to be toxic to articular cartilage; however, no studies have been conducted to explore the effects of decreasing DMSO concentrations for cartilage cryopreservation [36]. Consequently, there are several aspects of the cryopreservation of chondrospheres that require further elucidation.

Aim

The aim of this study was to optimize the slow cryopreservation protocols for neural organoids derived from iPSC and chondrospheres.

Methods

The experimental design is illustrated in Fig. 1.

 

Fig. 1. Experimental design: a, freezing of neural organoids on differentiation days 9, 16, 22, 29, and 43; during two weeks post-thaw and throughout the culture period, the micrographs were captured, and quantitative polymerase chain reaction and immunohistochemistry were performed; b, freezing of chondrospheres; after folding in V-shaped 96-well plates and incubation in the undiluted Matrigel matrix on day 29 from the formation of chondrospheres, the organoids were frozen in the 82% DMEM + 10% FBS + 8% DMSO + ROCK inhibitor medium; post-thaw micrographs were captured during two weeks, and then the expression of chondrocyte-specific markers was recorded and analyzed by immunohistochemistry. DMEM, Dulbecco’s modified eagle medium; FBS, fetal bovine serum; DMSO, dimethyl sulfoxide.

 

Differentiation of Neural Derivatives of Induced Pluripotent Stem Cells and Preparation of Midbrain Organoids

The neural derivatives were differentiated from the previously characterized IPSFD5S iPSC line carrying a female karyotype, obtained from fibroblasts of a healthy donor [37]. A 35-mm diameter Petri dish pretreated with the Matrigel matrix (Corning, USA) was inoculated with iPSCs to initiate the culture in the mTeSR1 medium (STEMCELL Technologies, Canada) with the HybriS-8 medium (PanEco, Russia) added at a ratio of 1:4. Midbrain organoids were obtained by following modifications to a previously published protocol [38]. As soon as the iPSCs achieved 90% confluence, the cells were removed using a 0.05% trypsin solution (PanEco, Russia) and folded into spheroids using AggreWell 800 plates (STEMCELL Technologies, Canada) in accordance with the manufacturer’s protocol. On the following day, the spheroids were transferred to bioreactors, which were prepared using an established protocol [38] in a medium comprising the following components:

• Advanced Dulbecco’s modified eagle medium (DMEM)/F12 (Gibco, Thermo Fisher Scientific, USA);
• 10 ng/mL basic fibroblast growth factor (bFGF) (Miltenyi Biotec, Germany);
• 2% NeuroMax supplement (PanEco, Russia);
• 1% N-2 (PanEco, Russia);
• 10 µM SB431542 (STEMCELL Technologies, Canada);
• 1% KnockOut Serum Replacement (Gibco, USA);
• 3 µM Dorsomorphine (Miltenyi Biotec, Germany);
• 0.1 µM LDN193189 (Miltenyi Biotec, Germany); and
• 1% Penicillin–Streptomycin (PanEco, Russia).

The spheroids were then cultured for seven days in a bioreactor on an orbital shaker at 70 rpm, 37 °C, and 5% CO2. The medium was changed on every second day. On day 7, the spheroids were transferred to the medium comprising:

• Advanced DMEM/F12 (Gibco, Thermo Fisher Scientific, USA);
• 10 ng/mL bFGF (Miltenyi Biotec, Germany);
• 2% NeuroMax supplement (PanEco, Russia);
• 1% N-2 (PanEco, Russia);
• 1% KnockOut Serum Replacement (Gibco, USA);
• 3 µM Purmorphamine (Miltenyi Biotec, Germany); and
• 1% Penicillin–Streptomycin (PanEco, Russia).

The spheroids were then cultured for seven days in a mini-bioreactor on an orbital shaker at 70 rpm, 37 °C, and 5% CO2. The medium was changed on every second day. On day 14, the spheroids were transferred to the medium comprising:

• Advanced DMEM/F12 (Gibco, Thermo Fisher Scientific, USA);
• 10 ng/mL bFGF (Miltenyi Biotec, Germany);
• 2% NeuroMax supplement (PanEco, Russia);
• 1% N-2 (PanEco, Russia);
• 1% KnockOut Serum Replacement (Gibco, USA);
• 20 ng/mL brain-derived neurotrophic factor (Miltenyi Biotec, Germany);
• 20 ng/mL glial cell line-derived neurotrophic factor (Miltenyi Biotec, Germany); and
• 1% Penicillin–Streptomycin (PanEco, Russia).

Thereafter, the organoids were cultured in a mini-bioreactor on an orbital shaker at 70 rpm, 37 °C, and 5% CO2. The medium was changed on every second or third day.

Preparation of Chondrospheres

Chondrocytes were isolated from the biopsy (surgical) specimens as previously described [39]. After the chondrocytes achieved 70% confluence, the cells were extracted from 6-well plates using a 0.05% trypsin solution (PanEco, Russia). After a rapid wash in DMEM medium (PanEco, Russia) with 10% FBS (Gibco, Thermo Fisher Scientific, USA), the cells were centrifuged for five minutes at 200 g. Thereafter, the cells were transferred to 96-well plates coated with 1.5% agarose at 100,000 cells/well. The cells were subjected to 1–3-days culturing in 150 µL complete chondrocyte medium containing DMEM/F12 (PanEco, Russia), 2 mM GlutaMAX (Thermo Fisher Scientific, USA), 1% Penicillin–Streptomycin (PanEco, Russia), and 10% FBS (HiMedia, India). Using a 3 mL Pasteur pipette, the spheroids were transferred from the wells to a test tube. Following a 2–3-minute period of settling, the supernatant was collected. Thereafter, the spheroids were immersed in a freshly thawed undiluted Matrigel matrix kept at 4 °C. Thirty minutes later, the Matrigel matrix was removed by passive precipitation of the spheroids in a test tube or by centrifugation for one minute at 100 g. The spheroids were transferred to mini-bioreactors [38] and placed on an orbital shaker (Infors, Switzerland) in a CO2 incubator (37 °C, 5% CO2, 70 rpm). The medium was changed twice a week without centrifugation. The spheroids were allowed to settle down by gravity in a test tube. Then, the supernatant was removed, and an equivalent volume of the fresh medium was added to the tube. Finally, the organoids were returned to the mini-bioreactors.

Freezing of Organoids

For the initial experiment, neural organoids were cryopreserved on days 8, 10, 15, 18, 20, and 22 from the start of differentiation using media 1, 2, and 3 (see Table 1). In the final experiment, the midbrain organoids were cryopreserved on days 9, 14, 22, 29, and 43 after the process of folding into spheroids and the initiation of differentiation. Cryopreservation media 2 and 4 were prepared (Table 1). On the scheduled days, non-frozen organoids served as controls.

 

Table 1. Composition of cryopreservation media for organoids

Medium No.	Composition
1	90% FBS, 10% DMSO, 10 µM Ri Y27632
2	92% FBS, 8% DMSO, 10 µM Ri
3	80% DMEM, 10% FBS, 10% DMSO, 10 µM Ri
4	82% DMEM, 10% FBS, 8% DMSO, 10 µM Ri

Note: FBS, fetal bovine serum (HiMedia, India); DMSO, dimethyl sulfoxide (PanEco, Russia); ROCK-inhibitor Y27632, Rho-associated protein kinase inhibitor (Miltenyi Biotec, USA).

 

At each time point, 10–15 organoids were carefully selected using a 3.5 mL Pasteur pipette and transferred to a 1.5 mL test tube. At 2–3 minutes after the precipitation of the organoids, the residual culture medium was removed, 500 µL of the cryomedium were added, and the organoids were transferred to the cryovial. The vial was maintained at 4 °C for 30 minutes, then cooled in CoolCell SV2 Freezing Container (Corning, USA) to –80 °C at –1 °C/minute. On the following day, the vial was transferred to liquid nitrogen.

Similar to the neurospheres, the chondrospheres were frozen in medium 4 (82% DMEM + 10% FBS + 8% DMSO + 10 µM Ri) on culture day 29 from the moment of folding.

Thawing and Culturing of Organoids

DMEM/F12 with 1% Penicillin–Streptomycin (PanEco, Russia) was used as a thawing medium used for washing the organoids from cryoprotectant. After the removal from liquid nitrogen, the cryovial was warmed to 37 °C for two minutes. Using a Pasteur pipette, the vial was transferred to 4 mL of DMEM/F12 at room temperature and gently mixed. After the organoids were settled down by gravity, the supernatant was removed almost completely, and the organoids were resuspended in a medium suitable for the specific cell type and stage of differentiation. The organoids were transferred to the mini-bioreactors and kept in an incubator on an orbital shaker at 70 rpm. The medium was changed every 3 or 4 days.

Assessment of Organoids for Integrity and Size

The organoid images were captured using Olympus IX53 Microscope with U-LS30-3 Camera (Olympus, Japan) on the day of thawing, the following day, one and two weeks post-thaw. The size of the organoids was analyzed using ImageJ software [40]. The photographic documentation of the non-frozen organoids was obtained on the cryopreservation days. These images were used as a reference. The organoids were assessed for the structural integrity and the presence of debris using optical microscopy. The following criteria were established for the assessment of the structural integrity in the present experiment:

1) The organoid’s shape was almost spherical.

2) The organoid demonstrated clearly defined boundaries.

3) The optical density of the organoid matched the reference value on the specified differentiation day.

Quantitative Polymerase Chain Reaction

After and throughout the culture period of the non-frozen neurospheres, the organoids were lysed using ExtractRNA buffer (Eurogen, Russia), and RNA was isolated according to the manufacturer’s protocol. The quality of the isolated RNA was assessed by electrophoresis in 0.7% agarose gel. The RNA concentration was measured using Equalbit BR RNA Analysis Kit (Vazyme, China) and Qubit RNA Fluorometer (Invitrogen, USA). The RNA was treated with DNase I (NEB, UK) according to the manufacturer’s protocol. The complementary DNA from the RNA matrix was synthesized using MMLV RT Kit (Eurogen, Russia) in accordance with the manufacturer’s protocol. The quantitative polymerase chain reaction (qPCR) was performed by adding 5 µL of the prepared 5x qPCRmix-HS SYBR (Eurogen, Russia), 1 µL of 10 µM primer (see Table 2), 18.2 µL of water, and 1 µL of the complementary DNA matrix per well of the 96-well plate (SSI Bio, USA). The control wells were prepared by adding 1 µL of PCR water (Eurogen, Russia) instead of the complementary DNA matrix. The reaction was performed using C1000 Touch PCR Thermal Cycler (Bio-Rad, USA) for nucleic acid amplification and CFX Manager software. The total number of cycles was 39. The results were analyzed in MS Excel 2108 (Microsoft, USA) using the ΔΔCt method [41]. Non-frozen neurospheres from differentiation day 9 were used as a control.

 

Table 2. Primer sequences used in the study

Primer	Primer sequence	Definition
GAPDH for	GAAGGTGAAGGTCGGAGTCA	Glyceraldehyde 3-phosphate dehydrogenase, forward primer
GAPDH rev	GTTGAGGTCAATGAAGGGGTC	Glyceraldehyde 3-phosphate dehydrogenase, reverse primer
MAP2 for	CGAAGCGCCAATGGATTCC	Microtubule-associated protein 2, forward primer
MAP2 rev	TGAACTATCCTTGCAGACACCT	Microtubule-associated protein 2, reverse primer
NESTIN for	CAACAGCGACGGAGGTCTC	Neuroepithelial stem cell protein, forward primer
NESTIN rev	GCCTCTACGCTCTCTTCTTTGA	Neuroepithelial stem cell protein, reverse primer

 

Immunohistochemistry

The organoids were fixed in a 4% paraformaldehyde solution for 10 minutes, then transferred to a 10% sucrose solution for 1 hour, and finally to a 30% sucrose solution for 24 hours. The organoids were washed in PSB, immersed in Tissue-Tek O.C.T.-Compound (Sakura Finetek, USA), cooled to −20 °C, and then cut using a cryotome (Thermo Fisher Scientific, USA). The fixed organoids were sliced into 10-µm layers and placed on poly-L-lysine-coated slides. SOX2 (rabbit anti-human SOX2, 1:100; ABclonal, USA), MAP2 (mouse anti-human MAP2, 1:200; ELK Biotechnology, USA), and beta-3-tubulin (rabbit anti-human TUBB3, 1:250; Affinity Biosciences, USA) were used as neural markers for staining. The proliferating cell nuclear antigen (PCNA) was identified as a suitable marker for the assessment of cell proliferation (mouse anti-human PCNA, 1:50; ELK Biotechnology, USA). The chondrocyte-specific markers were stained with anti-collagen type II primary antibody (rabbit anti-human, 1:100, ab34712; Abcam, UK), SOX9 (rabbit anti-human, 1:400, ES3481; ELK, USA), and aggrecan (mouse anti-human, 1:500, AHP0022; Invitrogen, USA). The secondary antibodies used in this study included goat anti-rabbit IgG Fc (Alexa Fluor 488) (1:800, ab150089; Abcam, UK) and goat anti-mouse IgG (H+L) cross-adsorbed secondary antibody (Alexa Fluor 555) (1:500, A-21422; Invitrogen, USA). The DAPI dye (D9542-5MG; Sigma-Aldrich, USA) was used for the visualization of cell nuclei. The obtained specimens were examined using Olympus IX53 Microscope with U-LS30-3 Camera.

Statistical Analysis

The significance of variations in size and number of intact organoids was evaluated using the Fisher test and the Student’s unpaired t test. Differences were considered statistically significant at p < 0.05. The statistical calculations were performed using MS Excel 2108 (Microsoft, USA). RStudio-2023.12.1 (R Core Team, Austria) was used for constructing the plots.

Results

Structural Integrity of Neural Organoids at Two Weeks Post-Thaw

In the initial experiment, iPSCs obtained from healthy donor’s fibroblasts were folded into spheroids and induced into a neural lineage. Starting from differentiation day 3, the organoids were collected at 3–5-day intervals for the slow cryopreservation. The FBS medium (with 10% DMSO and 8% DMSO) and the DMEM medium (with 10% DMSO) were utilized as cryopreservation media (Table 1). After thawing, the organoids were cultured for 10 days and the number of intact organoids was determined.

There was no significant difference in the number of intact organoids between the cryomedium groups (see Fig. 2, a, b). However, freezing in medium 1 was associated with the lowest percentage of intact organoids on culture day 10. At certain time points, the post-thaw number of intact organoids was less than 40%. A decrease in the DMSO concentration in the FBS cryomedium was found to contribute to the greater post-thaw structural integrity of organoids, with an average percentage of intact organoids of 60%, but not less than 40%. No loss of the organoid integrity was observed for medium 3 after 10-day culture periods at three of the seven scheduled time points (Table 1). As shown in Fig. 1, the initial post-thaw week was critical for maintaining the integrity of organoids after cryopreservation, as the greatest loss of intact organoids was observed during this period. As for the optimal differentiation period for the slow freezing of organoids, the largest proportion of intact organoids (85%) was reported for thawing occurred on differentiation days 20, 22, and 27. The initial experiment revealed that Medium 1 should be excluded from further consideration. For medium comprising 82% DMEM + 10% FBS + 10% DMSO + 10 µM Ri, it was determined that the DMSO concentration required a reduction from 10% to 8%.

 

Fig. 2. Integrity of neural organoids post-thaw: a, the post-thaw percentage of intact neural organoids vs. the initial number, by differentiation day and cryomedium used in the initial experiment; b, the percentage of intact neural organoids on day 10 post-thaw from the initial number, by differentiation day and cryomedium used in the initial experiment; c, the percentage of intact neural organoids vs. the initial number, by differentiation day and cryomedium used in the final experiment; whiskers represent the standard deviation; d, the percentage of intact neural organoids on day 14 post-thaw from the initial number, by differentiation day and cryomedium used in the final experiment.

 

The second experiment included two cryopreservation media: medium 4 (82% DMEM + 10% FBS + 8% DMSO + 10 µM Ri) and medium 2 (FBS + 8% DMSO + 10 µM Ri). In further experiments, these media were referred to as cDMEM and cFBS, respectively. In this experiment, similar findings were obtained for mature neural organoids (aged more than three weeks after the initiation of differentiation). However, in the replicated experiments for young organoids, the proportion of intact neurospheres was found to be lower than that observed in the initial experiment. This discrepancy can be attributed to the increased post-thaw culture period (10 and 14 days) and the increased number of replicates, which increased the statistical significance of the repeated experiment. The proportion of intact organoids was higher for more mature neurospheres than for immature ones (Fig. 2, c, d). Meanwhile, the cDMEM cryomedium induced the fragmentation and formation of new spheres in some of the organoids on the following day post-thaw, thereby increasing the total number of intact organoids compared to the initial number. Furthermore, Fig. 2, d demonstrates that, the cDMEM medium promoted the improved integrity of mature organoids, although this finding did not attain statistical significance (p = 0.06, Student’s unpaired t test).

The obtained findings suggest that the media composition had a negligible effect on the structural integrity of organoids. The stage of differentiation emerged as the predominant factor in this context. However, the following trend was observed: a decrease in the DMSO concentration and the use of DMEM as the main component of the cryomedium positively affected the survival of organoids.

Variations in Diameter of Neural Organoids at Two Weeks Post-Thaw

The organoid images were captured throughout the entire culture period and during two weeks post-thaw. This was done to measure the diameter and visually assess the morphology (see Fig. 3). On the day of thawing, the shape and size of the organoids remained consistent across cryomedia, exhibiting no significant differences compared to non-frozen organoids at the respective days of differentiation. The following day, a substantial amount of cellular debris was observed in the bioreactors. This phenomenon likely originated from the outer layer of organoids, which demonstrated an irregular distribution (Fig. 3, b). This effect was observed at all stages of neural organoid differentiation within a variety of cryogenic media. One week later, the surface of the organoids became even and smooth (Fig. 3, c). However, the thawed organoids exhibited a smaller size than the non-frozen organoids on the day of differentiation. The mean organoid volume loss was 14.5%, with a mean error of 7.21%. Noteworthy, the smallest decrease in post-thaw volume occurred in organoids frozen on differentiation day 29 (3.7% of the initial volume), while the largest decrease was observed in organoids frozen on day 43 (25.4% of the initial volume). The loss of volume caused by the death of the surface layer of cells was not counterbalanced by the growth of the organoid itself, indicating a lag phase associated with stress-induced slow-down in cell division. As shown in Fig. 3, е, the post-thaw lag phase extended for approximately 6–7 days, which included the loss of the cell surface layer across various cryomedia. After an additional week, there was a substantial increase in the size of the organoids, which suggested the resumption of cell division and ongoing differentiation. However, the size of the organoids remained below the values recorded on the respective differentiation day. After the lag phase, the growth rate was on average equivalent to that of non-frozen organoids. The only exceptions were organoids frozen on differentiation days 9 and 29: after the lag phase, the growth rate of the thawed organoids exceeded that of the non-frozen organoids. After two weeks, the sizes of the organoids from the experimental and control groups were not statistically significantly different for both time points.

 

Fig. 3. Morphological changes and size dynamics of organoids during pre- and post-cryopreservation culture."Non-frozen" refers to organoids that were not subjected to freezing; cFBS refers to organoids that were cryopreserved in the 92% FBS + 8% DMSO + 10 µM Ri medium; cDMEM refers to organoids that were cryopreserved in the 82% DMEM + 10% FBS + 8% DMSO + 10 µM Ri medium. "Day x + y" represents the actual differentiation day, where x is the day of cryopreservation and y is the number of days post-thaw; а, c, d, micrographs of organoids before and after cryopreservation; b, micrographs of organoids with a "halo" effect on days 1–7 post-thaw, with the most representative shown; scale bar = 200 µm; e, variations in organoid diameter throughout the culture period for non-frozen organoids and during two weeks post-thaw (cDMEM and cFBS). For cDMEM and CFBs, the curves were presented starting from the actual differentiation day. The mean value for 12 organoids at each point is demonstrated, with the standard deviation indicated by the whiskers. To assess the significance of size differences between frozen and non-frozen organoids, the Student’s unpaired t-test was used, with p-values shown on the curve.

 

The thawed neural organoids exhibited the loss of volume after the 6–7-day lag phase, which was attributed to the death of the surface layer. Thereafter, the organoids demonstrated a resumption in growth, exhibiting a rate that was comparable to that of the non-frozen organoids.

Expression of Neural Markers at Two Weeks Post-Thaw

At two weeks post-thaw, the organoids were divided into two portions. The first portion was lysed for further analysis of the neural marker expression by qPCR (see Fig. 4), and the second portion was fixed and stained for fluorescence microscopy (see Fig. 5, 6).

 

Fig. 4. Results of quantitative polymerase chain reaction. “Non-frozen” (blue) refers to organoids that were not subjected to cryopreservation; cFBS (yellow) refers to organoids that were cryopreserved in the 92% FBS + 8% DMSO + 10 µM Ri medium; cDMEM (pink) refers to organoids that were cryopreserved in the 82% DMEM + 10% FBS + 8% DMSO + 10 µM Ri medium. Variations in the MAP2 and NES expression are demonstrated for non-frozen organoids (blue) throughout the culture process and thawed organoids (pink and yellow) at two weeks post-thaw. In this case, the expression values for the thawed specimens are shifted two weeks to the right from the freezing day. This adjustment was made to account for the post-thaw culture time. The dots represent the mean expression value of 2–ΔΔCt, with the whiskers showing the standard deviation.

 

Fig. 5. Immunohistochemical staining of intact and cryopreserved neural organoids at different stages of differentiation using various cryopreservation media. “Non-frozen” refers to organoids that were not subjected to cryopreservation; cFBS refers to organoids that were cryopreserved in the 92% FBS + 8% DMSO + 10 µM Ri medium; cDMEM refers to organoids that were cryopreserved in the 82% DMEM + 10% FBS + 8% DMSO + 10 µM Ri medium. "Day x + y" represents the actual differentiation day, where x is the day of cryopreservation and y is the number of days post-thaw. Scale bar = 50 µm; a, TUBB3 (green), PCNA (red), DAPI (blue); b, SOX2 (green), MAP2 (red), DAPI (blue). An enlarged fragment is highlighted in the yellow box to demonstrate the specificity of staining.

 

Fig. 6. Micrographs of chondrospheres before and after cryopreservation during dynamic culture (days 0, 1, 7, and 14). Scale bar = 200 µm

 

Organoids that were cryopreserved at varying stages of differentiation demonstrated different levels of the gene expression for microtubule-associated protein 2 (MAP2) and neuroepithelial progenitor intermediate filament protein (NES) (Fig. 4). The MAP2 expression levels in organoids frozen in both cFBS and cDMEM differed by culture stage at the time of cryopreservation. The non-frozen organoids were characterized by complex patterns of the MAP2 expression. Starting from differentiation day 9, there was an increase in the expression, with a peak observed on differentiation day 29. Thereafter, the expression decreased and plateaued at a relatively constant level until two months post-differentiation. In the thawed organoids, the MAP2 expression patterns generally mirrored those in the non-frozen organoids, peaking on differentiation day 29, despite the lag phase. A noteworthy observation was the finding that, between differentiation days 29 and 43, the MAP2 expression in the thawed organoids was on average lower compared to the non-frozen organoids on the respective differentiation days. However, an opposite trend was observed between days 43 and 57, with a significant increase in the MAP2 expression in the thawed organoids. The cryomedium composition did not induce a significant difference in the MAP2 gene expression levels. Furthermore, the organoids frozen in the DMEM and FBS media exhibited a comparable level of protein expression.

NES expression patterns were similar among experimental organoids, but significantly different from those observed in non-frozen organoids. The non-frozen organoids were characterized by two peaks of the NES expression, i.e., on differentiation days 16 and 29. However, the peaks of the NES expression were less accentuated in the thawed organoids. Noteworthy, the experimental groups of organoids had generally lower levels of the NES expression compared to the control group, except the experimental points on days 22 and 43 post-cryopreservation. Similar to the MAP2 gene, the NES expression in the thawed organoids was independent of the cryogenic medium composition. Evidently, the stage at which the neural organoids were frozen played a critical role.

All control and experimental organoids demonstrated the immunoreactivity to the neural microtubule protein TUBB3 and PCNA, as detected by immunohistochemistry (Fig. 5, b). Although TUBB3 levels remained consistent across all groups, the PCNA expression varied significantly among organoids cryopreserved in the FBS medium and those frozen after 16 culture days (days 22, 29, and 43) in the DMEM medium.

Furthermore, the protein MAP2 and the transcription factor SOX2 were observed in all the groups (Fig. 5, a). However, the cFBS-frozen organoids demonstrated a proportion of SOX2-positive cells that increased with the longer culture period in which cryopreservation was performed. In contrast, the cDMEM-frozen organoids showed a lower fluorescence intensity on days 22 and 43 compared to days 16 and 29. Both experimental and control groups exhibited comparable levels of the MAP2 synthesis.

Structural Integrity of Chondrospheres and Expression of Chondrocyte-Specific Markers at Two Weeks Post-Thaw

To assess the suitability of the neural organoid freezing protocol for chondrospheres, chondrocytes were isolated from the patient’s biopsy specimen, which was followed by their expansion and spheroid formation. After the transfer to the bioreactor on culture day 29, the chondrospheres were cryopreserved in the DMEM cryomedium containing 8% DMSO. After thawing, the aggregation (clumping) of the chondrospheres was observed over 14 days. On day 14, the spheroids spontaneously self-organized to form a conglomerate (Fig. 6). Despite the aggregation of individual spheroids into a single organoid, the spheroids did not disintegrate, indicating the post-thaw survival of the spheroids. Noteworthy, the non-frozen chondrospheres also clumped together, thereby forming conglomerates. In contrast to the observations obtained with the neurospheres, no decrease in the diameter of individual chondrospheres or loss of the surface layer of cells was detected post-thaw. To confirm the chondrocytic phenotype after cryopreservation, immunohistochemistry was performed to assess the expression of chondrocyte-specific markers before freezing and at two weeks post-thaw.

An analysis of the immunohistochemical staining for typical chondrocyte markers, such as SOX9, aggrecan, and type II collagen (see Fig. 7), showed positive staining for aggrecan and collagen in both the pre-cryopreservation and post-thaw specimens. However, SOX9 was only detected in the thawed specimens. The proliferating cell nuclear antigen (PCNA) was not identified in the examined spheroid specimens. This was most likely attributable to the low proliferative potential of the chondrocytes that had formed a spheroid and were subject to contact inhibition, as PCNA was found exclusively in the actively proliferating cells. This finding correlated with the absence of variations in the size of individual chondrospheres throughout the culture process under dynamic conditions, both before and after thawing.

 

Fig. 7. Immunohistochemical analysis of chondrospheres before cryopreservation and 14 days post-thaw: a, expression of SOX9 (green), aggrecan (ACAN, red); b, expression of collagen 2 (Сol2, green) and PCNA (red). DAPI (blue) nuclear staining. Scale bar = 50 µm

 

Discussion

The optimization of organoid cryopreservation protocols represents a vital biomedical task. The primary criteria for successful cryopreservation include the preservation of the post-thaw viability and functional activity of organoids over an extended period. The present study adopted slow freeze protocols for neural and chondrocytic organoids.

The neural organoids of the hindbrain demonstrated a series of post-thaw changes when cultured in various media and at different stages. A notable observation was the presence of a 7–10-day lag phase, where no increase in organoid size was evident. The neural 2D cultures were characterized by the post-thaw cell proliferation recovery occurring within 2–3 days [42]. In the context of 3D culturing, neural iPSC derivatives were likely to experience greater stress, attributable to the disparity in the availability of cells for cryomedium components. This, in turn, substantially prolonged the post-thaw recovery period.

A decrease in the proliferative activity was accompanied by the loss of the surface layer of cells in the brain organoids. A ~10% reduction in the size of these organoids was likely attributable to the slow freeze procedure [43–45]. However, the available data for neural organoids are limited and inconclusive. The slow freezing using DMSO as a cryoprotectant was also observed to result in a slight increase in the size of neural organoids [21]. The loss of the outer layer of organoids was only critical for the initial stages of differentiation, when their sizes were relatively small, yet cryopreservation did not affect the proliferative potential of cells. More mature neural organoids demonstrated a better post-thaw integrity. Therefore, the maturation stage of organoids influences their post-thaw structural integrity, and more mature organoids are thus recommended for freezing.

Compared to the generally accepted DMSO concentration of 10% [21], the use of 8% DMSO in the present experiment produced a positive effect on the preservation of organoids. The cryomedium (DMEM or FBS) used did not have a significant effect on maintaining the structural integrity. FBS stabilized osmotic pressure throughout the freezing and thawing processes, while simultaneously reducing DMSO concentrations. The freezing of hematopoietic cells was shown to be unaffected by the use of either 20% or 70% FBS in the cryosolution [46]. In the context of neural organoids, a modification in FBS concentrations with the constant DMSO concentration was likely to have negligible effects on maintaining the integrity of the organoids.

Furthermore, the expression of neural markers was consistent across all cryomedium groups. The MAP2 expression patterns in the thawed neural organoids were comparable to those in the non-frozen ones, despite the lag phase. Notably, earlier studies suggested that the normal MAP2 expression pattern was expected to show an increase after one and two months of differentiation. However, in the present experiment, where shorter time intervals between measurements were applied, more complex expression patterns were observed [47, 48]. Specifically, the MAP2 expression decreased on differentiation day 43, with two peaks observed on days 29 and 57, both in the non-frozen organoids and in the thawed specimens. These fluctuations associated with the differentiation and maturation stages were identified as hallmarks of the transition between the differentiation stages [49].

As evidenced elsewhere, the NES expression remains unchanged with the use of the serum or culture medium [20]. However, in the present experiment, the NES expression patterns were observed over a longer period of time than was previously documented. A decrease in the expression of the NES gene, a marker of early neural precursors, suggested a maturation of the neurospheres as they were being cultured. Differences in the NES expression between the thawed and non-frozen neurospheres were observed on culture day 29 (i.e., 14 days after the initiation of differentiation). These differences were found to be unrelated to the cryomedium used. This period was critical for the expansion of neural precursors [49], and the thawing-induced stress apparently contributed to abrupt maturation.

The distribution of the expression of the neuronal markers MAP2 and TUBB3, and the proliferative marker PCNA, in the thawed organoids was generally consistent with that observed in the non-frozen organoids at almost all stages of differentiation. This finding suggests that the stress triggered by the loss of the outer layer of organoids did not disrupt the cytoskeleton organization within the neural organoid cells. Noteworthy, the cryopreservation of organoids at late stages of differentiation, characterized by a >1 mm diameter, was not associated with cortical necrosis, as evidenced by the DAPI staining of cortical organoids and the maintenance of intact nuclei. Therefore, the slow-freezing technique provides a uniform cooling process for the organoid cells, thereby preventing the formation of water crystals. The distribution of the SOX2 expression in the thawed organoids was similar to that in the non-frozen organoids at almost all stages of differentiation. However, the organoids frozen in the cFBS medium on differentiation day 43 exhibited an increase in the SOX2 expression. The SOX2+ cells appeared to be randomly arranged, rather than organized into “sockets” that would typically line the cavity of secondary structures. The association of SOX2 with proliferation and repair suggests a potential mechanism by which the thawing-induced stress might have led to the increased SOX2 expression [50, 51]. The neural organoids frozen in the cDMEM cryomedium on the same day exhibited a different pattern. The SOX2 expression exhibited a distribution pattern similar to that observed in the non-frozen controls. Similar effects have not been described in the relevant publications [20, 21]. Furthermore, the increased SOX2 expression and its irregular distribution in the neural organoids were associated with impaired intercellular interactions [52–54]. In consideration of the above, it appears that a spontaneous post-thaw disruption in the development of organoids occurred.

In summary, this study suggests that the most optimal stage for cryopreservation of neural organoids is 3–4 weeks from the initiation of differentiation. This scenario was associated with minimal damage induced by cryopreservation, the preservation of the intact organoids, and the rapid recovery of their morphofunctional characteristics. The influence of the cryopreservation medium on the process was not found to be significant. However, the use of DMEM as a cryomedium has been shown to reduce the overall cost of freezing compared to pure FBS. Therefore, with all other factors being equal, Medium 4 (82% DMEM + 10% FBS + 8% DMSO + 10 µM Ri) was identified to be the optimal medium for the cryopreservation of neurospheres in this study.

This medium was used for the cryopreservation of chondrospheres. After thawing, all the spheroids survived. No cell loss was observed from the surface layers, suggesting a high number of intercellular contacts and extracellular matrix in the 3D structure. Furthermore, the enhanced adhesive potential of spheroids can be ascertained through the observation of a single conglomerate on culture day 14 after thawing. Importantly, this effect was not attributable to the cryopreservation; rather, it was associated with the properties of the extracellular matrix, which was found to be more actively synthesized in 3D cultures. This finding aligns with the hypothesis that the process is mechanism-dependent [55, 56].

Aggrecan and collagens, hallmark markers of mature hyaline cartilage, were found to be intact after thawing [57]. The observation of the post-thaw SOX9 expression, a marker of chondrogenesis that is abundant in chondrocyte progenitor cells, may be attributable to the osmotic and temperature stresses experienced by the cells [58, 59]. This observed effect appears to be associated with the activation of reparative processes in the chondrosphere cells.

PCNA plays a significant role in DNA replication processes [60]. The absence of this protein marker for proliferating cells was attributed to the fact that dynamic 3D cultures caused cells to switch from the active proliferation to the increased synthesis of extracellular matrix components. This hypothesis is consistent with the increased adhesive potential of cells inside the chondrospheres and spheroids in relation to each other.

Attempts have been made to freeze whole pieces of cartilage biopsy specimens. The approaches described in the relevant studies yielded inconsistent results, which were found to be contradictory [61, 62]. Vitrification has been demonstrated to exert no adverse effect on the mechanical properties of cartilage and can be used as an alternative to fresh allografts with limited storage time [63]. Chondrospheres are complex cell structures, and there is no evidence-based freezing protocol for such biological objects. The solution to this challenge will facilitate the transportation of chondrospheres from the laboratory to clinical centers, as the shelf life of the finished cell-derived product is limited to two days [33].

Therefore, this study has enhanced the current protocol of the slow cryopreservation of neural organoids and chondrospheres by reducing the DMSO concentration and selecting the optimal stage of differentiation for neurospheres. The post-thaw number of intact organoids has increased compared to other protocols. Additionally, new findings have emerged for variations in the diameter of organoids throughout the long-term culture period after thawing. Future investigations will focus on the impact of freezing and thawing processes and the use of non-penetrating cryoprotectants on the morphofunctional characteristics of cryopreserved neural organoids. Vitrification of chondrospheres and neurospheres is a promising avenue for further research.

Conclusion

The potential applications of organoids are broad, encompassing in vitro testing of new therapeutic agents and clinical use. The practical value and widespread use of this technology demand its expansion and give rise to biobanking. In this context, the standardization of cryopreservation protocols for the complex structures, such as neural organoids and chondrospheres, emerges as a pressing priority in both fundamental and translational medicine. The present study proposes a simple, reproducible, and effective cryopreservation protocol for brain and cartilage organoids. It is reasonable to hypothesize that this protocol will continue to improve in the future.

Additional information

Author contributions: A.A. Barinova: investigation, writing—original draft; P.A. Golubinskaya: investigation, resources, writing—review & editing; A.S. Pikina: writing—original draft; E.S. Ruchko: resources, investigation; A.V. Eremeev: writing—review & editing. All the authors approved the version of the manuscript to be published and agreed to be accountable for all aspects of the work, ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Acknowledgments: The authors express their gratitude to Irina Kopylova and Liliya Belikova, staff members of the Laboratory of Cell Biology, Lopukhin Federal Research and Clinical Center of Physical-Chemical Medicine of Federal Medical Biological Agency for their assistance with qPCR and IHC, as well as Olga S. Lebedeva and Alina Davidenko for editorial support.

Ethics approval: The study was approved by the Local Ethics Committee of the Lopukhin Federal Research and Clinical Center of Physical-Chemical Medicine of Federal Medical Biological Agency (protocol No. 2019/02, April 9, 2019).

Informed consent: The authors obtained written informed consent from the patients, using a form approved by the Local Ethics Committee of the Lopukhin Federal Research and Clinical Center of Physical-Chemical Medicine of Federal Medical Biological Agency, for publication of their data in the scientific journal Genes & Cells, including the online version. The scope of the published data was approved by the patient.

Funding sources: This publication (the work on chondrospheres) was supported by the Russian Science Foundation (grant No. 22-15-00250).

Disclosure of interests: The authors have no relationships, activities, or interests for the last three years related to for-profit or not-for-profit third parties whose interests may be affected by the content of the article.

Statement of originality: No previously published material (text, images, or data) was used in this work.

Data availability statement: All data generated during this study are available in the article.

Generative AI: No generative artificial intelligence technologies were used to prepare this article.

Provenance and peer review: This paper was submitted unsolicited and reviewed following the standard procedure. The review process involved an external reviewer, two members of the editorial board, and an in-house scientific editor.

About the authors

Anna A. Barinova

Lopukhin Federal Research and Clinical Center of Physical-Chemical Medicine of Federal Medical Biological Agency

Author for correspondence.
Email: barinova.anna.al@mail.ru
ORCID iD: 0009-0001-1212-8154
SPIN-code: 1955-4313
Russian Federation, Moscow

Polina A. Golubinskaya

Lopukhin Federal Research and Clinical Center of Physical-Chemical Medicine of Federal Medical Biological Agency

Email: polinapigeon@gmail.com
ORCID iD: 0000-0002-1765-9042
SPIN-code: 5299-9693

MD, Dr. Sci. (Medicine)

Russian Federation, Moscow

Arina S. Pikina

Lopukhin Federal Research and Clinical Center of Physical-Chemical Medicine of Federal Medical Biological Agency

Email: arina.pikina@yandex.ru
ORCID iD: 0000-0002-8967-2318
SPIN-code: 8654-7318
Russian Federation, Moscow

Evgenii S. Ruchko

Lopukhin Federal Research and Clinical Center of Physical-Chemical Medicine of Federal Medical Biological Agency; Koltzov Institute of Developmental Biology Russian Academy of Science

Email: ruchkoevgeny@yandex.ru
ORCID iD: 0000-0002-1361-666X
SPIN-code: 7220-6031
Russian Federation, Moscow; Moscow

Artem V. Eremeev

Lopukhin Federal Research and Clinical Center of Physical-Chemical Medicine of Federal Medical Biological Agency; Koltzov Institute of Developmental Biology Russian Academy of Science

Email: art-eremeev@yandex.ru
ORCID iD: 0000-0002-3428-7586
SPIN-code: 4825-5440

Cand. Sci. (Biology)

Russian Federation, Moscow; Moscow

References

Supplementary files