If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Cell-based assays performed in multiwell plates are utilized in basic and translational research in a variety of cell models. The assembly of these multiwell platforms and their use is often laboratory specific, preventing the standardization of methods and the comparison of outputs across different analytical sites. Moreover, when cell models are based on primary cells with specialized culture requirements, including three-dimensional (3D) cell culture, their complexity and the need for manipulation by experienced operators can add significant cost and introduce long lead times to analysis, both of which are undesirable in any preclinical situation. To address this issue, we explored adaptations of cryopreservation technology that allow cells to be cryopreserved in-plate, ready for use in analysis, and have developed a method applicable to cells from different origins and different culture formats. Here we describe the application of this technology to conventional two-dimensional (2D) monolayers of human mesenchymal stem cells (MSCs) and human macrophages derived from primary monocytes, and to 3D cultures of hepatic organoids, colon organoids, and colon tumor organoids, each presented for cryopreservation in their obligate extracellular matrix. We demonstrated that cell viability, cell physiology, and cytotoxic sensitivity were maintained after cryopreservation, such that the models offer the means to uncouple model assembly from analytical use and to standardize cell models in product form for distribution to end users.
In vitro cell culture models are valuable tools for basic and translational research. They are commonly used in drug discovery and safety testing, to investigate efficacy, cytotoxicity, or collateral effect, for example, immunogenicity. If model components are selected and assembled carefully, they have potential to reproduce physiological function, thereby providing a reliable link with in vivo responses.
However, with recognition of the importance of microenvironmental cues and cellular crosstalk, there is now potential to realize complex disease processes hitherto capable of study only in expensive and ethically unattractive in vivo models.
Historically, in vitro testing has made use mainly of immortalized and cancer cell lines, which represent the healthy cell only to a limited extent.
However, cell culture techniques have advanced substantially over the past decades, for instance, with the development of molecular strategies to reprogram induced pluripotent stem cells (iPSCs).
Those cells possess the ability to differentiate into various multipotent cell lineages with high self-renewal capacity. One example is mesenchymal stem cells (MSCs), which can be further differentiated into lineage-specific cell types, such as osteoblasts or adipocytes.
Another possibility to circumvent the phenotypic shortcomings and other drawbacks associated with ever-expanding cell lines is the use of primary cells derived directly from human tissue. This strategy is frequently employed for cells of the hematopoietic lineage, that is, myeloid and lymphoid cells. For instance, macrophages, differentiated in vitro from peripheral blood monocytes, play a key role in immune safety and drug/nano-safety and can be harnessed to perform such tests in vitro.
Another drawback deriving from the use of monocultures and/or two-dimensional (2D) settings is the lack of the microenvironmental cues that reproduce in vivo function. More advanced co-culture systems allow study of the communication between two or more cell populations and present a step forward from monoculture systems. Furthermore, the addition of scaffolds representing the extracellular matrix of the tissue of origin, which allow the generation of three-dimensional (3D) models with several cell populations, represents sophisticated systems that reinstate tissue-like architecture and function. Within this framework, an innovative cell culture method has entered center stage in recent years: the ability to culture tissue-derived adult stem cells in a 3D environment in order to obtain “mini-organs,” so-called organoids.
These multicellular structures represent in vitro the characteristics of their healthy or diseased tissue of origin, have the capacity of self-renewal, and have proven to be a valuable donor- and disease-specific model in drug discovery.
Independently of the type of culture model used, cell-based assays are mostly conducted in multiwell plates. The use of these types of formats allows the standardization of experimental outputs, the study of multiple samples in parallel, and/or a panel of readout parameters. The application of multiwell formats has itself acquired unprecedented speed and reproducibility, enabling high-throughput analysis, by the application of automation technology such as high-precision liquid dispensing and robotic devices for plate handling and manipulation. Nevertheless, automated analysis must in some cases be preceded by less easily scalable steps involving cell isolation, characterization, propagation, and differentiation. These preanalytical stages demand cell handling by experienced operators and can add significant cost. Furthermore, when assay-ready cells demand nonconventional culture format, for example, 3D culture, advanced dispensing techniques such as bioprinting may be required to achieve acceptable consistency of cell seeding within a multiwell plate format. The challenge of consistent model assembly has been a principal reason for the slow uptake of advanced cell culture models, despite their acknowledged advantage in terms of preclinical predictive value. One way to overcome these operational pitfalls, from an end user perspective, would be to source preassembled cell models in multiwell format and in cryopreserved form, thereby uncoupling model assembly from analytical use. This strategy would ideally include rapid sourcing from consistent stock, a storage option, and a short lead time to analysis, providing advanced cell-based analysis in a “plug and play” format.
The basic strategy of cryopreservation has been used for many years to preserve cells as a stock and protect them from aging.
This technique is used commonly in research, to cryopreserve cells in suspension in cryovials, and more recently, attempts have been made to translate it to monolayer and/or in-plate cell cultures.
We have previously demonstrated that cryopreservation technology can be adapted to 3D bovine colonoid cultures, by demonstrating a similar growth rate of control versus cryopreserved cultures and analogous cytotoxic response facing a cytotoxic agent.
Here, we describe how this technology can be further adapted not only to validate cell viability and cytotoxicity but also to support other cell-based analytical responses, such as cell activation, differentiation, and gene expression. We perform this by demonstrating in-plate cryopreservation of biologically relevant 2D cell monolayers (MSCs and macrophages) and 3D human cell cultures supported in an extracellular environment (hepatic organoids, colon organoids, and colon tumor organoids) with application of cryopreservation technology (AvantiCell Science, Ayr, UK). We provide evidence for efficient recovery of these cells after in-plate cryopreservation and illustrate the retention of physiological response and gene expression pattern of in situ frozen cells by comparison with nonfrozen cells.
The basic principles of in-plate cryopreservation and its benefits for cell culture applications are outlined in Figure 1A. The technique presented here can be adapted for application to basic and translational research protocols and offers significant operational advantages for many cell-based assays, especially when advanced cell models based on primary cells or stem cell-derived cultures are an obligate analytical requirement.
Figure 1In-plate cryopreservation technology can be applied to various cell types and cell culture formats. (A) Schematic overview of in-plate cryopreservation technology. Left: Cells in 2D or 3D formats can be cryopreserved in different multiwell plate formats. Middle: Cryopreserved plates allow a better logistical transportation and long-term storage. Right: After thawing, multiwell plates allow the realization of different assays or drug screenings. (B) Phase-contrast/bright-field images of iPSC-MSCs, M1- and M2-like monocyte-derived macrophages, hepatic organoids, colon organoids, and colon tumor organoids in control and cryopreserved cultures. Cells were cryopreserved 3 h postseeding (organoids), 24 h postseeding (iPSC-MSCs), and at day 7 of differentiation (macrophages). Images were taken 24 h postthaw (iPSC-MSCs and macrophages) and 4 days postthaw (organoids).
Human tissue was obtained under ethical approval from the West of Scotland Research Ethics Committee, and in compliance with UK Human Tissue Authority guidelines. Tissues were procured as anonymized material indirectly through specialist intermediary suppliers from sources adhering fully to the legal and ethical requirements of the country of collection, which included prior written donor consent.
Human blood samples were sourced from the UK Blood and Transfusion Unit under recipient agreement N/029 as whole blood or as leukocyte cone retentate.
iPSC-derived MSCs were sourced under commercial agreement from Phenocell SAS (Grasse, France).
Cell Preparation and Culture
iPSC-MSCs
Cells obtained from Phenocell SAS were expanded in tissue culture flasks precoated with 0.1% (w/v) gelatin (Sigma-Aldrich, Irvine, UK) in iPSC-MSC proliferation medium (Suppl. Table S1). iPSC-MSCs were cultured at 37 °C in a humid atmosphere with 5% CO2, and medium was replaced every 2 days. At confluence, cells were washed with phosphate-buffered saline (PBS; Thermo Fisher Scientific, Waltham, MA), harvested with 0.05% (w/v) trypsin in 1 mM EDTA (Sigma-Aldrich), and plated at 5 × 103 cells/cm2. Cells were cultured for no more than 12 passages. For differentiation into adipocytes and osteoblasts, iPSC-MSCs were plated at 2.5 × 104 cells/well in precoated 96-well plates (Greiner Bio-One, Stonehouse, UK) and differentiated for 21 days. To induce adipogenic differentiation, iPSC-MSCs were cultured with 100 µL/well of StemMACS Adipocyte Media (Miltenyi Biotec, Surrey, UK). Alternatively, an osteogenic fate was induced by culture in 100 µL/well of StemPro Osteogenesis Differentiation Kit (Miltenyi Biotec). Differentiation media were replaced every 3 days.
Macrophages
Peripheral blood mononuclear cells (PBMCs) were isolated from samples of whole peripheral blood by density gradient centrifugation methodology. Monocytes were extracted from the PBMC isolate by positive immunomagnetic selection using magnetic CD14 MicroBeads (Miltenyi Biotec), and extracted CD14+ cells were cultured in RPMI Medium 1640–GlutaMAX (Thermo Fisher Scientific) supplemented with 10% (v/v) heat-inactivated human AB serum (Sigma-Aldrich), 50 µg/mL gentamicin (Sigma-Aldrich), and 50 µM of 2-mercaptoethanol (Sigma-Aldrich). The purity of isolated cells was analyzed using flow cytometry with fluorescein isothiocyanate (FITC)-conjugated CD14 monoclonal antibody (Miltenyi Biotec). Isolated monocytes were cultured at a density of 5 × 104 cells/well in 96-well plates at 37 °C with 5% CO2 and differentiated into M1- or M2-like macrophages after 7 days of culture in differentiation medium. To differentiate monocytes into M1-like macrophages, basal medium was supplemented with 50 ng/mL granulocyte–macrophage colony-stimulating factor (GM-CSF; R&D Systems, Abingdon, UK) at day 0. After 48 h, cultures were fed with the same volume of medium supplemented with 100 ng/mL GM-CSF. At day 6, half the medium was replaced by demidilution with fresh medium supplemented with 100 ng/mL GM-CSF. To differentiate monocytes into M2-like macrophages, basal medium was supplemented with 50 ng/mL macrophage colony-stimulating factor (M-CSF; R&D Systems) at day 0. After 48 h, cultures were fed with the same volume of medium supplemented with 100 ng/mL M-CSF. At day 6, half the medium was replaced by demidilution with fresh medium supplemented with 50 ng/mL interleukin (IL)-10 (R&D Systems) and transforming growth factor-beta (TGF-β) (R&D Systems).
Human Hepatic Organoids
Human hepatic tissue specimens were processed within 24 h after surgical excision. The epithelial stem cells were isolated, cultured, and differentiated in accordance with the proprietary methods of Broutier et al.
with minor modifications. Hepatic organoids were passaged at a ratio of 1:2 to 1:4 every 4–5 days and seeded in domes of 35 µL of Reduced-Growth Factor Basement Membrane Matrix type 2 (RGF BME2; Amsbio, Abingdon, UK) in 24-well culture plates (Greiner Bio-One). During cell expansion, domes were overlaid with hepatic expansion medium (Suppl. Table S2). Medium formulation was adapted from Broutier et al.
Following a maximum period of 48 h after surgical excision, human colonic tissue was washed, and the upper mucosal layer was dissected from the tissue by cutting it into small mucosal strips of approximately 1 × 3 mm. Methods for the isolation of colon crypts were adapted from the protocols developed by Sato et al.
The mucosal strips were washed with cold wash tissue solution (Suppl. Table S3) at least three times, followed by 30 min of incubation with chelation buffer (Suppl. Table S4) at 4 °C to release the colon crypts. After removal of chelation buffer, colon crypts were washed with cold tissue wash solution and gently vortexed (1–2 s). A small sample was taken to observe the presence of crypts, and the washing + vortex procedure was repeated until no more crypts were observed by microscopic examination. Crypts were seeded at a density of 500 crypts per 35 μL of ice-cold BME2 (Amsbio) per well of a 24-well culture plate (Greiner Bio-One). BME2 domes were allowed to polymerize at 37 °C in a humid atmosphere with 5% CO2 for 15 min before they were overlaid with 500 μL of human colon organoid complete medium (Suppl. Table S5). Medium formulation was adapted from previous publications by Sato et al.
Embedded crypts were cultured at 37 °C in a humid atmosphere with 5% CO2, and culture medium was replaced every 2 days. Approximately 10 days after seeding, colon organoids were passaged using TrypLE Express (Thermo Fisher Scientific) with 10 µM rock inhibitor (Y-27632, Cambridge Bioscience, Cambridge, UK) for 10 min, reembedded in 35 μL of BME2 (Amsbio) in a 24-well plate format (Greiner Bio-One), and cultured in 500 μL of human colon organoid complete medium.
Human Colon Tumor Organoids
Following a maximum period of 48 h after surgical excision, human colorectal cancer tissue was washed, dissected, and digested in a digestive solution (Suppl. Table S6) until approximately 80% of the cell suspension was composed of single cells. Methods for the isolation of cells from tumor tissue were adapted from previous protocols.
Cells were seeded at a density of 5 × 104 cells per 20 μL of cold BME2 (Amsbio) in each well of a 48-well culture plate (Greiner Bio-One). BME2 domes were allowed to polymerize at 37 °C in a humid atmosphere with 5% CO2 for 15 min before they were overlaid with 200 μL of human colon tumor organoid complete medium (Suppl. Table S7). Medium formulation was adapted from previous publications.
Embedded single cells were cultured at 37 °C in a humid atmosphere with 5% CO2, and culture medium was replaced every 2 days. Approximately 8 days after seeding, colon tumor organoids were passaged using TrypLE Express (Thermo Fisher Scientific) with 10 µM rock inhibitor (Y-27632, Cambridge Bioscience) for 15 min, reembedded in 35 μL of BME2 (Amsbio) in a 24-well plate format (Greiner Bio-One), and cultured in 500 μL of human colon tumor organoid complete medium. For cytotoxicity assays, colon tumor organoids were treated with sodium butyrate (NaBt; 150 mM; cat. B5887, Sigma-Aldrich) for 24 h.
In-Plate Cryopreservation
Cell cultures were cryopreserved in situ in their culture plate and revived according to their specific conditions (Suppl. Table S8). After equilibration with cryopreservation solution (Cryotix, AvantiCell Science Ltd, Ayr, UK) at 37 °C, 5% CO2, temperature was gradually decreased to a final temperature of –80 °C using an EF600 controlled-rate freezer (Asymptote Ltd, Cambridge, UK). Cryopreserved plates were transferred to a cryogenic ultra-low-temperature (ULT) freezer (model MDF-C2156VAN-PE, Panasonic, Kadoma, Japan) at –150 °C for long-term storage. During the thawing process, cell cultures were preincubated at –20 °C for 30 min (Suppl. Table S5). Prewarmed culture medium was added to each well and the plate was transferred to a heated static platform at 37 °C until completely thawed. Once the plate was thawed, media were immediately replaced with fresh cell culture media and cultures were incubated at 37 °C, 5% CO2.
Cell Viability Assay
Cell viability assay was performed on cells cultured in 96-well plates, incubated for 3 h in the dark, at 37 °C in a humidified atmosphere with 5% CO2, with a buffered solution that contains highly purified resazurin (Sigma-Aldrich). Cell viability was measured as turnover of resazurin to fluorescent resorufin by metabolically active cells. Fluorescence (excitation 530 nm/emission 590 nm) was detected with a Synergy H4 Hybrid Multi-Mode Microplate Reader (BioTek Instruments Inc., VT).
Alternatively, cell viability was determined by quantification of ATP present in metabolically active cells, using the CellTiter-Glo 3D cell viability assay (Promega, WI). The assay was performed according to the manufacturer’s instructions in 96-well culture plates (Greiner Bio-One). For quantitative analysis, an ATP standard curve was generated immediately prior to adding the CellTiter-Glo reagent. Luminescence signal was detected using a Synergy H4 Hybrid Multi-Mode Microplate Reader (BioTek Instruments).
ELISA
Secreted cytokines were measured by enzyme-linked immunosorbent assay (ELISA) in supernatants collected at selected culture time points. Samples were diluted 1:10 to 1:100 and analyzed using IL-8, IL-6, and tumor necrosis factor-alpha (TNF-α) Duo Set ELISA kits (R&D Systems, Abington, UK), according to the manufacturer’s instructions. Readouts were obtained with a Synergy H4 Hybrid Multi-Mode Microplate Reader (BioTek Instruments).
RNA Isolation and Real-Time Reverse Transcription PCR
Total RNA was isolated from human colon tumor organoids using the RNeasy mini kit (Qiagen, Manchester, UK), including a genomic DNA digestion step using RNase free DNase (Qiagen). Total RNA was quantified with the Quant-iT RiboGreen RNA assay kit (Invitrogen, MA). Total RNA (0.2 μg/sample) was reverse transcribed to cDNA using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, CA). Quantitative real-time (qRT) PCR was performed with the QuantiTest SYBR Green PCR Kit (Qiagen) and detected with the Light Cycler 480 System (Roche, Basel, Switzerland). Predesigned primers were obtained from Sigma-Aldrich (Irvine, UK). Colon tumor organoid samples were tested for the presence of LGR5, MYC, VIL, MUC2, CA2, BAMBI, GLUT1, and PKM2. Data were analyzed using the advanced relative quantification of target and control samples. The expression values of target samples were normalized to the expression levels of the housekeeping gene GAPDH and presented as fold change compared with control samples.
Fluorescence Imaging
All images were acquired using a Cytation 5 Multi-Mode Cell Imaging reader (BioTek Instruments).
Live/Dead Staining
Cells were exposed to a freshly prepared solution of 8 μg/mL fluorescein diacetate (FDA; Sigma-Aldrich) and 20 μg/mL propidium iodide (PI; Sigma-Aldrich) in PBS (Thermo Fisher Scientific) for 4 min at room temperature. After incubation, cells were washed once with PBS and immediately imaged.
DAPI Staining
4′,6-Diamidino-2-phenylindole (DAPI) nuclear staining was performed using Fluoroshield (Sigma-Aldrich) according to the manufacturer’s instructions. Briefly, drops of Fluoroshield (Sigma-Aldrich) were added directly to the cells and, after 25–30 min of incubation in the dark, washed twice with PBS. After imaging, the number of DAPI-positive cells was expressed relative to the total number of cells, to normalize recovery using cell viability assays in 96-well plates.
Staining for Markers of Cell Differentiation
Differentiation of iPSC-MSC to an osteogenic lineage was demonstrated by detection of alkaline phosphatase activity and calcium deposition. Alkaline phosphatase activity was identified using SigmaFast 5-bromo-4-chloro-3-indolyl phosphate (BCIP)/nitro blue tetrazolium (NBT) (Sigma-Aldrich). Briefly, cells were washed with PBS (Thermo Fisher Scientific), fixed with 4.2% (v/v) formaldehyde solution (BD Bioscience, NJ) for 1 min, and washed with washing buffer consisting of 0.05% (v/v) Tween 20 (Sigma-Aldrich) and PBS (Thermo Fisher Scientific). Cells were incubated with BCIP/NBT substrate solution for 5–10 min, washed with washing buffer, and immediately imaged. Calcium mineralization was identified using Alizarin Red S (Millipore Sigma, MA) staining. After cell fixation with 4.2% (v/v) formaldehyde solution (BD Bioscience) for 30 min, cells were rinsed twice with deionized water (Thermo Fisher Scientific) and stained with Alizarin Red S solution (Millipore Sigma) for 45 min. After incubation, the cell monolayer was washed four times with deionized water (Thermo Fisher Scientific), and images were recorded without delay. Adipocyte differentiation was confirmed by using the lipophilic red dye Oil Red O (Sigma-Aldrich). Briefly, 0.05% (w/v) Oil Red O-isopropanol stock solution (Alfa Aesar, MA) was diluted into a working solution (3 parts dye to 2 parts distilled water) before being applied to the monolayer for 30 min. Before dye incubation, cells were fixed with methanol (Sigma-Aldrich) for 5 min. Once the staining was completed, cells were washed with deionized water and images were recorded.
Statistical Analysis
Statistical tests were run as indicated in the individual figure legends using GraphPad Prism 6.02 software (GraphPad, La Jolla, CA). Results were considered significant at p < 0.05. Results are represented as mean ± SEM. The p values are given as *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, and ****p ≤ 0.0001.
Results
In-Plate Cryopreservation Can Be Applied to a Variety of Cell Culture Conditions
In order to demonstrate in-plate cryopreservation as a robust, novel cell culture methodology, which can be applied to multiple cell culture conditions, a variety of cell types in both 24- and 96-multi-well plate formats were tested (Fig. 1A). Cells were cryopreserved in-plate applying the gradient-freezing technique described in Materials and Methods and stored at ultralow temperature (–150 °C) before returning them to standard cell culture at 37 °C. The proof of concept for in situ cryopreservation is provided in Figure 1B, depicting bright-field/phase-contrast images of control cell cultures that had not undergone cryopreservation, and their cryopreserved counterparts 24 h postthaw (iPSC-MSCs, macrophages) and 4 days postthaw (hepatic and colon organoids and colon tumor organoids). Control cultures were expanded for the same number of days as cryopreserved cultures. Both the state of confluence and cellular morphology did not alter upon cryopreservation, irrespective of cell type.
Cell Viability and Multipotency Are Maintained across Multiwell Plates upon In Situ Cryopreservation of iPSC-MSCs
iPSC-MSCs were seeded at a density of 5 × 103 cells/cm² in 96-well plates. After 24 h, cells were cryopreserved by applying the cryopreservation technique outlined in Materials and Methods. Noncryopreserved iPSC-MSC cultures were seeded and cultured in parallel to serve as comparable controls. Cell viability was analyzed by live/dead staining (relative FDA-to-PI fluorescence intensity) 48 h postthawing or after 48 h of normal culture in the control group. Cell viability was assessed per well and averages were calculated in at least 32 replicate wells per treatment, in a total of four independent experiments (Fig. 2A). Variability of in-plate viability was expectedly small in control cultures and was only dependent on seeding density. However, analysis of initial cryopreserved iPSC-MSC culture plates revealed that the freezing–thawing procedure required several optimization steps for variability to be minimalized. While the first nonoptimized cryopreservation attempts were performed with only the inner 60 cell-containing wells filled with cryoprotectant (CPA) during cryopreservation, the optimized procedure contained CPA in all 96 wells of the multiwell plate, irrespective of cell content. Furthermore, the thawing procedure from ultralow temperatures of cryopreserved plates was modified by introducing a 30 min prethawing step at –20 °C prior to complete thawing. Through the introduction of those optimization steps, the variability of in-plate viability was lowered from a standard deviation (SD) of 33.8% (highest value in the four independent experiments of the nonoptimized method) to an SD of 8.8% (highest value in the four independent experiments of the optimized method), being close to that of control plates with an SD of 7.6% (highest value in the four independent experiments of control plates) (Fig. 2A). Visual evaluation of FDA staining 48 h postthaw, in a plate subjected to the optimized treatment, demonstrated an even distribution of living cells within the plate and within each individual well (wells C7–C11, D7–D11, E7–E11, F11, and G7–G11), and the absence of living cells in wells corresponding to the lysis control (wells F7–F10), or where cells had been cryopreserved using standard medium instead of appropriate CPA (negative control, wells B7–B11) (Fig. 2B).
Figure 2Cryopreservation of iPSC-MSCs in multiwell plates. (A) Cell viability within control and cryopreserved 96-well plates 48 h postthaw. Cells were cryopreserved using an optimized freeze–thaw procedure or a nonoptimized freeze–thaw procedure (see main text), and individual plates were evaluated for variability of viability. The graph represents the relative FI of whole-well montage FDA stains with at least 32 replicate wells positioned within the inner 60 wells of a 96-well plate. (B) Whole-well montage FDA stain at 48 h postthaw following the optimized freeze–thaw procedure within a 96-well plate. The plate contains positive controls wells where cells were cryopreserved using the appropriate cryopreservation Cryotix iPSC-MSC medium (wells C7–C11, D7–D11, E7–E11, F11, and G7–G11), negative control wells cryopreserved using culture medium instead of appropriate CPA (wells B7–B11), and lysis control wells in which cells were lysed using Triton X-100 prior to FDA staining (wells F7–F10). (C) Cryopreserved iPSC-MSCs were differentiated into osteoblasts or adipocytes, after cryopreservation. Differentiation was confirmed by staining calcium deposits with Alizarin Red stain and lipids with Oil Red O stain to ensure osteocyte and adipocyte differentiation, respectively.
The differentiation potential of cryopreserved iPSC-MSCs was investigated to assess the cells’ multipotency following recovery into culture. iPSC-MSCs were successfully differentiated to an osteocyte phenotype and to adipocytes by applying the respective culture stimuli documented in Materials and Methods. Osteocyte phenotype was indicated by staining for calcium deposits with Alizarin Red stain, whereas adipocyte-associated lipid accumulation was demonstrated with Oil Red O staining, in each case after a stimulation of similar duration to that of noncryopreserved iPSC-MSC (Fig. 2C).
Primary Human Immune Cells Maintain Their Activation Profiles after In Situ Cryopreservation
In-plate cryopreservation of human immune cells was demonstrated using primary monocytes cultured at a density of 3.5 × 104 cells/cm2 in 96-well plates, and then differentiated into M1- and M2-like macrophages using phenotype-specific differentiation factors as elucidated in Materials and Methods. After 7 days of differentiation, mature M1- and M2-like macrophages were cryopreserved in-plate, subsequently recovered from frozen storage, and cultured again for a period of 2 days, during which time cells were challenged for 24 h with an endotoxin stimulus (25 ng/mL lipopolysaccharides [LPSs]). Simultaneously, noncryopreserved control cells were seeded, differentiated, and assayed following the same procedure as cryopreserved cells (Fig. 3A). Due to the diversity of donors of primary monocyte-derived macrophages, donor variability needs to be considered in the data processing of primary human innate immune cells.
Figure 3Cryopreservation of M1- and M2-like monocyte-derived macrophages in 96-well plates. (A) Schematic representation of the cryopreservation process. (B) Live/dead staining of in-plate cryopreserved M1- and M2-like macrophages 48 h postthawing. (C) Percentage of cell recovery of cryopreserved M1- and M2-like macrophages in comparison with control macrophages, calculated quantifying live/dead and DAPI staining analysis. Results represent the live FDA staining relative to the DAPI cell count of the whole-well montage with at least 12 replicate wells. (D) Cell viability of cryopreserved M1- and M2-like macrophages. Cells in control and frozen plates, cryopreserved using an optimized procedure in Cryotix macrophage medium, were tested with a viability assay 48 h postthawing. The graph represents the relative luminous intensity signal of the whole-well montage in a 96-well plate. (E) Percentage of activation of cryopreserved M1- (left) and M2- (right) like macrophages in comparison with control cells. Macrophages were exposed to medium alone or medium containing bacterial LPS (25 ng/mL) for 24 h, 24 h postthawing. The secretion of TNF-α, IL-8, and IL-6 after stimulation was measured by ELISA in conditioned medium. Controls and cryopreserved cells not stimulated by LPS were not activated by the CPA. Both M1- and M2-like cryopreserved macrophages showed activation after LPS stimulation. Cytokine secretion (pg/mL) was normalized for all the samples with cell recovery values. The activation of cryopreserved cells is shown as a percentage of the activation of control cells. Samples were analyzed using one-way analysis of variance (p < 0.0001).
The viability of cryopreserved macrophages was initially examined by live/dead staining 48 h after thawing of cryopreserved cells, and at the equivalent culture time in control (Fig. 3B). Based on live/dead staining analysis, using DAPI staining to normalize the cell number, the cell recoveries of cryopreserved M1- and M2-like macrophages were measured as 79% and 82%, respectively, compared with control macrophages (Fig. 3C). To confirm the obtained data, the viability of M1- and M2-like macrophages obtained from a panel of donors was measured in control and cryopreserved cells 48 h postthawing. The percentage of viable cryopreserved cells in comparison with control cells was an average of 77% for M1-like macrophages and 83% for M2-like macrophages (Fig. 3D). The robustness of cell viability after cryopreservation is defined by SD values of 10.8 and 8.2, respectively, in M1- and M2-like macrophages. These data supported the consistency of in situ cryopreservation across cell cultures differentiated from eight different donors (four donors were used to assess cell viability and four were used to measure cell recovery).
The use of human immune cells in cell-based analysis is generally to measure the immunogenicity or immunotoxicity of test agents, and it is therefore critical that a cryopreserved cell platform retains sensitivity and responsiveness to immune stimuli. The impact of cryopreservation technology on macrophage immunosensitivity was tested by challenging cryopreserved macrophages with LPS 24 h postthawing and quantifying the response in terms of cytokine secretion (i.e., TNF-α, IL-8, and IL-6). Cytokine induction in cryopreserved and nonfrozen cells from the same donor was compared after normalization with respect to cell number. Neither M1 nor M2-like macrophages were activated by the cryopreservation procedure in the absence of endotoxin stimulation. Conversely, both M1 and M2 cryopreserved macrophages were activated by LPS, and cytokine secretion showed activation responses similar to those of nonfrozen cells in five independent experiments (Fig. 3E). Statistical analysis showed no significant differences between cryopreserved and control cell cultures of either M1- or M2-like macrophages with respect to TNF-α and IL-6 secretion. A higher secretion of IL-8 from cryopreserved macrophages was detected (fold change, 1.29) in comparison with control macrophages after LPS stimulation. Taking donor variability into consideration, the data show comparable cell activation across several experiments.
Cryopreservation of 3D Cultures in Multiwell Platforms and Application in Drug Testing
Following successful cryopreservation of 2D cultures in multiwell platforms, the same technology was applied to 3D cultures. Human hepatic organoids, colon organoids, and colon tumor organoids were seeded in either 96-well or 24-well plate formats, cryopreserved 2 h after seeding, stored at –150 °C, revived from frozen, and cultured for a period of 2 days (hepatic) and 4 days and assayed as controls. The high content of live (FDA—green stained) compared with dead (PI—red stained) cells in 96-well plate formats (Fig. 4B) indicates a positive outcome in survival, independently of the culture type. The same experiment was performed in 24- and 96-well plate formats, with an analogous positive outcome in survival (representative images of hepatic organoids in Fig. 4C). The high recovery of viable cells after cryopreservation was reflected in the maintenance of the 3D dome-containing organoid structure, as visualized by representative bright-field pictures taken of the 24-well plates with tumor organoids (Fig. 4D).
Figure 4Cryopreservation of 3D cultures. (A) Schematic representation of the generation of 3D cultures and cryopreservation process. (B) Colon organoids and tumor organoids were seeded in 96-well plate formats and cryopreserved with Cryotix organoid medium with a gradual freezing protocol, revived, and cultured for 5 days. Cultures were assessed for the content of live/dead cells using FDA/PI staining. (C) Hepatic organoids were seeded in 24- and 96-well plate formats and cryopreserved with Cryotix organoid medium with a gradual freezing protocol, revived, and cultured for 2 days. Cultures were assessed for the content of live/dead cells using FDA/PI staining. (D) The integrity of the 3D dome was visualized by bright -ield images of 24-well plates for both control and cryopreserved cultures. Representative pictures are showed for colon tumor organoids. (E) Gene expression of colon tumor organoids after cryopreservation was determined by qRT-PCR and compared with control noncryopreserved cultures. Results represent the fold change of gene expression of cryopreserved to control noncryopreserved cultures. Samples were obtained at day 5 of culture. (F) Control or cryopreserved human colon tumor organoids were treated for 24 h with NaBt, 5 days postthaw. Left: Viability was accessed using live/dead staining assay. Right: Graph representing the percentage of dead cells (PI staining/[PI + FDA]). The graphs represent the results from three independent experiments, three replicates per gene per screen, or three replicates per condition per experiment, mean ± SEM. Mann-Whitney rank-sum test and two-way analysis of variance.
Control and cryopreserved colon tumor organoid cultures were analyzed by qRT-PCR to determine the effect of cryopreservation on gene expression. Cryopreserved tumor organoid cultures, in comparison with their control noncryopreserved counterparts, demonstrated similar gene expression to those of stem cell marker LGR5; proliferation marker MYC; epithelial cell markers VIL, MUC2, and CA2; cancer-associated gene BAMBI; and metabolism-associated genes GLUT1 and PKM2 (Fig. 4E). The stability of gene expression after cryopreservation is a positive indicator that the cryopreservation process does not affect the physiological behavior of the organoid cultures.
Exploring the interest and value of cancer models in drug discovery and the potential of using cryopreserved 3D cancer models to this end, control and cryopreserved colon tumor organoid cultures were treated with a cytotoxic agent, NaBt. NaBt, a bacterial byproduct found in the gut, plays an important regulatory role in gut barrier function and has previously been demonstrated to possess antitumorigenic properties.
Tumor organoids were expanded for 5 days after thawing, treated with 150 mM NaBt for a period of 24 h, and assayed with live/dead staining (Fig. 4F, left). The size of tumor organoids reflects the tumorigenic nature of these cells, and tumor organoid size was variable among all experiments. Quantification of live/dead staining demonstrated an average of 49.5% ± 9.6% dead cells following NaBt treatment of control tumor organoids and 50.8% ± 12.4% in cryopreserved tumor organoids. No statistically significant differences were found between either the two treated groups or the control groups (Fig. 4F, right).
Discussion
Cell culture models are constantly challenged and improved. This evolution has moved in vitro research to a point where cell models are sufficiently advanced to allow reconstruction of normal cell biology or complex disease states to a meaningful degree. Use of these models in life science research is typically low-throughput and compatible with customized, manual assembly of cell models by experienced operators. The needs of industry, and specifically the preclinical adoption of in vitro models in drug discovery, demand high-throughput systems requiring the scalable assembly of cell models to high levels of standardization. Meeting this challenge depends upon meticulous assembly of cell models and minimizing the setup costs and lead time to analysis. To address these issues, we considered the use of cryopreservation as a method of preserving cell models in situ, for subsequent supply to analysis end users. Cryopreservation has been utilized for several years as a method of preserving cells, tissues, and organs for downstream processing.
However, its methodology is mostly applied to cell banking in cryovials, or in platforms that require further cell culture and cell model assembly after thawing. Here, we drive this technology toward a user-friendly solution by demonstrating a method of in situ cryopreservation of various cell types and models in multiwell culture formats suitable for screening purposes.
A critical step in optimization of in situ, multiwell plate cryopreservation involved addressing the variability of ice nucleation between culture wells. Uncontrolled nucleation, characterized by the extracellular formation of ice crystals during freezing, is well known to cause cellular damage, leading to variability in cell recovery, viability, and function.
This phenomenon, initially observed as significant variability of in-plate cell viability on thawing, was measured across the 96 wells of multiwell plates seeded with iPSC-MSC (Fig. 2A). Optimization took the form of empirical adjustment of the time of contact between the CPA and the cells prior to initiation of gradient freezing, and the inclusion of CPA across all wells of the multiwell plate, irrespective of cell content. In addition, a stepwise thawing protocol was introduced, whereby plates were equilibrated at –20 °C before complete thawing. This optimization resulted in consistent iPSC-MSC viability in thawed cultures comparable to that of unfrozen control plates (Fig. 2A). This optimized protocol was subsequently applied to the other cell models, with minor adjustments in the equilibration times before gradient freezing and before complete thawing (Suppl. Table S8). With these adjustments, consistent cell viability and recovery were observed for macrophages (Fig. 3B,C), colon organoids and tumor organoids, and hepatic organoids (Fig. 4C). Based on this criterion, our protocol enabled the efficient, reproducible cryopreservation of preassembled cell models to a standard supportive of postthawing use in cell-based analysis.
Viable cell recovery is a prerequisite of successful cryopreservation in-plate and has been reported elsewhere.
However, these studies do not indicate cell functionality allowing the application of recovered cells in cell-based analysis. In fact, several studies describe functional and molecular changes of cells after cryopreservation.
We have previously demonstrated that cryopreservation of colon bovine organoids in multiwell plates was associated with retention of cytotoxic sensitivity under cell-based analysis conditions.
Here, we further tested the capacity of cryopreserved cells to respond to physiological stimulus in a similar way to control cells, by analysis of several parameters intrinsic to several cell models: differentiation of iPSC-MSC, activation of macrophages, gene expression of organoids, and cytotoxicity in tumor organoids.
Cryopreserved iPSC-MSCs qualitatively expressed differentiated functions representative of osteoblasts and adipocytes after challenge with the same differentiation stimuli as control cells (Fig. 2C). Furthermore, quantitative analysis of LPS-stimulated M1- and M2-like monocyte-derived macrophages, challenged 24 h after thawing, elicited TNF-α, IL-8, and IL-6 cytokine release comparable to that of control cells, offering the means to assemble and stockpile analytical resources, with minimal lead time, for use in immunological screening. It was confirmed that the cryopreservation itself did not induce an inflammatory response, as cryopreserved macrophages were not activated after thawing (Fig. 3E). Also, the human donor variability in inflammatory response typically observed in control cell cultures was maintained after freezing and thawing of macrophage isolates from different donors; macrophages derived from eight different donors exhibited characteristic cytokine responses, irrespective of whether cells had been cryopreserved or not (Fig. 3D).
The cryopreservation technology also proved suitable for cryopreservation 3D structures in the form of organoid cultures, whose physiological relevance and analytical potential are well documented.
Both hepatic organoid cultures and tumor organoid cultures remained physically stable after in situ cryopreservation (representative data of colon tumor organoids in Fig. 4C). Moreover, colon tumor organoid performance was unaffected by cryopreservation when assessed in terms of expression of proliferation marker MYC; epithelial cell markers VIL, MUC2, and CA2; cancer-associated gene BAMBI; and metabolism-associated genes GLUT1 and PKM2, at day 5 of culture (Fig. 4E). Furthermore, the sensitivity of the colon tumor organoid model to the antineoplastic agent NaBt
remained unaltered by the cryopreservation procedure (Fig. 4F). The stability of gene expression in cryopreserved cultures further confirms the potential to use this cryopreservation technology for preassembled cell models without detrimental impact on their subsequent analytical performance.
Although we can demonstrate that cryopreservation in situ is an innovative method with numerous applications, it is important to mention that this method is limited to adherent cultures. The cryopreservation of dendritic cells in situ was tested, but since the process involves several stages of washing steps and the addition/removal of cryoprotectant, most of the initially seeded cells were lost.
The ability to cryopreserve 2D and 3D cell culture models in-plate opens a wide range of potential applications. First, model preassembly can give access to high-performance cell-based tools without the need for demanding assembly or staff expertise. Second, cell model stock can be stored for the long term, uncoupling model production from analytical use. Third, evidence of in situ, frozen cell storage without loss of cell function offers the prospect of shortening lead time to analysis, and the possibility of rapid response or just-in-time provision of analytical resources. Together, these features may promote the further uptake of human cell-based analysis as the ethically acceptable, commercially sustainable tool for preclinical research and development.
Supplemental material is available online with this article.
Declaration of Conflicting Interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: All authors are employed by AvantiCell Science Ltd, and their research and authorship of this article were completed within the scope of their employment with AvantiCell Science Ltd.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors wish to acknowledge European Commission support for this work under EU FP7 and Horizon2020 projects HUMUNITY (FP7-PEOPLE-INT-2012 GA no. 316383), ANAPRINT (H2020 GA no. 697804), PANDORA (H2020 GA no. 671881), and TRANSMIT (H2020 GA no. 722605).
06665-5&id=gr1.jpg){kind=link}
06665-5&id=gr2.jpg){kind=link}
06665-5&id=gr3.jpg){kind=link}
06665-5&id=gr4.jpg){kind=link}