Abstract
Three dimensional models of cell culture enables researchers to recreate aspects of tumour biology not replicated by traditional two dimensional techniques. Here we describe a protocol to enable automated high throughput phenotypic profiling across panels of patient derived glioma stem cell spheroid models. We demonstrate the use of both live/dead cell end-points and monitor the dynamic changes in the cell cycle using cell lines expressing the FUCCI cell cycle reporter. Together, these assays provide additional insight into the mechanism of action of compound treatments over traditional cell viability assay endpoints.
Keywords
1. Introduction
Glioblastoma multiforme (GBM) is the most common type of primary malignant brain tumour in adults. Molecular genetic profiling of patient samples has revealed considerable cellular heterogeneity within and between GBM tumours [
[1]
]. Studies based on bulk gene expression profiles suggest that at least three GBM cell subtypes exist: proneural, classical and mesenchymal [[2]
,[3]
]. Multi-region tumour sampling has shown that these distinct subtypes can also co-exist in different regions of the same tumour and adapt over time and in response to treatment [3
, 4
, 5
]. In addition, GBM hijacks mechanisms of neural development to produce sub-compartments of glioblastoma stem cells (GSCs) that are thought to exhibit resistance to radiotherapy and chemotherapies contributing to tumour-propagating potential and recurrence following treatment [6
, 7
, 8
]. The proportion of GSCs in tumours and also their expression of the cell proliferation biomarker, Ki67 can often determine the aggressiveness of GBM and inversely correlates with patient outcomes [[9]
]. Many of the genetic drivers of GBM are involved in the regulation of the cell cycle either directly (for example loss of CDKN2A/CDKN2B or amplification of CDK4 or CDK6) or indirectly (for example amplification or mutation of EGFR or PI(3)K genes) [[10]
] and therefore targeting of the dysregulated GBM cell cycle is an attractive therapeutic strategy. A further layer of heterogeneity within GBM exists at the level of the tumour microenvironment including regions of necrosis and tissue hypoxia which influences GBM cell behaviour, including maintenance of the GSCs [[11]
].Two-dimensional (2D) monolayer cell culture techniques have been used routinely to model cancer cell behaviour in the laboratory and support drug discovery and pharmacogenomic studies. Recent advances in defining the culture conditions that maintain and expand cancer stem cells in vitro has supported small molecule and CRISPR screening on patient-derived GSCs [
[12]
,[13]
]. However, despite these advances, such 2D assays fail to recapitulate the full complexity of the GBM microenvironment that exists in patients including specific segments of GBM pathophysiology associated with three-dimensional (3D) tissue architecture including cell-cell communication, hypoxia and invasion.The use of 3D cell culture techniques to model aspects of tumour biology not captured in 2D monolayer models has been growing in interest over the last decade and has benefited from the convergence of advanced imaging technologies [
[14]
], 3D bioprinting [[15]
], organoid tissue culture [[16]
] and microfluidic devices [[17]
]. Examples range from simple 3D spheroid models of free-floating cell aggregates [[18]
] to more complex multicellular organoid and organ-on-a-chip devices. Here we provide a detailed protocol of a previously described method used in our laboratory to kinetically profile and quantify transition through the cell cycle in 3D spheroid tumour models using the Fluorescent Ubiquitination-based Cell Cycle Indicator (FUCCI) reporter [[18]
].Methods for isolating, cell culture maintenance and genetic manipulation of GSCs in vitro have been established [
[13]
,[19]
]. In the current protocol we demonstrate application of the 3D spheroid FUCCI method across a panel of patient derived GSCs to support drug discovery and personalized medicine strategies for the treatment of GBM.2. Materials
2.1 Cell culture
Culture of glioblastoma stem cells is outlined below and is a modified protocol originally published by the Pollard laboratory [
[20]
].Patient derived glioblastoma stem cell lines (E13, classical; E57, mesenchymal) used in this study were obtained from the Glioma Cellular Genetics Resource (gcgr.org.uk).
2.2 Reagents
- •D-(+)-Glucose solution (100 g L−1) (Sigma-Aldrich, G8644).
- •MEM nonessential amino acids (100x) (Gibco, 11140-035).
- •Bovine Albumin Serum Fraction V (7.5% solution, w/v) (Gibco, 15260-037).
- •2-Mercaptoethanol (50 mmol L−1) (Gibco, 31350010).
- •B27 supplement (50x) (with vitamin A and insulin) (Gibco, 17504-044).
- •N2 supplement (100x) (Gibco, 17502-048).
- •Mouse EGF (PeproTech, 315-09), 100 μg mL−1 stock.
- •Human FGF (PeproTech, 100-18b) 50 μg mL−1 stock.
- •Laminin 1 (from murine Engelbreth-Holm-Swarm (EHS) tumor) (Cultrex, 3446-005-01), 1 mg mL−1 stock.
- •Accutase Solution (1x) (Sigma-Aldrich, A6964).
- •Puromycin (Invivogen, ant-pr-1), 10 mg mL−1 stock.
- •Phosphate buffered saline (Thermo Scientific, BR0014G) supplemented with 0.1 g L−1 calcium chloride (Sigma-Aldrich, 383147) and 0.2 g L−1 magnesium chloride hexahydrate (Sigma-Aldrich, M9272).
- •IncuCyte Cell Cycle lentivirus (0.2 mL; >1 × 106 TU/mL) (Sartorius, 4779).
- •Corning tissue culture plastic ware.
- •General tissue culture materials.
- •Ultra low attachment ‘U’ shaped 96 well plates (Corning, 7007) or S-BIO PrimeSurface 3D culture spheroid plates (96 well, MS-9096UZ; 384 well, MS-9384UZ).
2.2 Equipment
- •Plate centrifuge.
- •Incucyte S3 (Sartorius) with spheroid analysis module.
- •General tissue culture equipment.
2.3 Media preparation
- a)Neural Stem Cell media:
Supplement a bottle of DMEM/Ham's-F12 (Sigma-Aldrich, D8437; 500 mL) with the following;
- 7.25 mL D-(+)-Glucose solution
- 5 mL MEM nonessential amino acids
- 0.8 mL bovine serum albumin solution (7.5%)
- 1 mL 2-Mercaptoethanol
- 5 mL B27 supplement
- 2.5 mL N2 supplement
This media can be kept at 4°C of 4 weeks or frozen at -20°C for longer term storage.
- b)Complete growth media:
Supplement Neural Stem Cell media with the following;
- 10 ng mL−1 EGF (PeproTech, 315-09)
- 10 ng mL−1 FGF (PeproTech, 100-18b)
- 2 μg mL−1 laminin 1 (Cultrex, 3446-005-01)
Note: keep laminin stocks cold (using ice) whilst preparing stocks and dilutions and use pre-chilled buffers to prevent gelation.
Note: supplementing culture media with laminin is sufficient to support growth of GBM stem cell lines and reduces the time and cost of pre-coating tissue culture plastic; some cell lines may require optimisation of the laminin concentration for monolayer cell attachment and growth.
This media can be kept at 4°C for up to 4 weeks.
- c)Wash media: DMEM/Ham's-F12 (Sigma-Aldrich, D8437; 500 mL) supplemented with 1 mL bovine serum albumin solution (7.5%) (Gibco, 15260-037)
3. Methods
3.1 Generation of a stable population of FUCCI expressing cells
- 1Pre-coat a 6-well cell culture plate with a 1 mL solution per well of laminin (10 μg mL−1) diluted in cold PBS (with CaCl2 and MgCl2).
- 2For a T25 cell culture flask, aspirate media from adherent monolayer cell culture and replace with 1 mL of pre-warmed Accutase solution (1 ×). Incubate flask at 37 °C for 1-5 minutes.
- 3Gently tap the flask by hand to observe cell monolayer detachment and visually confirm detachment using a tissue culture microscope.
- 4Add 4 mL of pre-warmed wash media to rinse cells from the bottom of the flask and transfer the resultant 5 mL cell suspension to a centrifuge tube.
- 5Centrifuge the cell suspension at 300 × g for 3 minutes to obtain a cell pellet.
- 6Resuspend the cell pellet in 5 mL of complete growth medium.
- Note - If desired take an appropriate split for further culture in a fresh T25 flask.
- 1Seed approximately 5 × 104 cells in 2mL of complete growth media in a 6-well plate pre-coated with laminin (from step 1) to give an approximate confluence of 40% and incubate for 24 hours.
- 2Add 20 μL (1.5 × 105 TU mL−1) IncuCyte Cell Cycle lentivirus (Sartorius, 4779) to 1 mL of complete growth media containing 0.8 μg mL−1 polybrene and add this mixture on to the cells. Incubate for 24 hours.
Note – handling of lentivirus particles should be conducted under the appropriate local Biosafety rules.
- 1After incubation, carefully change the culture media for fresh complete growth media and incubate for a further 24-48 hours.
- 2Carefully change culture media for fresh complete growth media supplemented with 1 μg mL−1 puromycin.
- Note – a puromycin kill curve should be performed to optimise puromycin concentration for new cell lines.
- 1Replace culture media every 48-72 hours to remove dead cells.
- 2Collect culture supernatant and test for the absence of residual virus by checking there is no reinfection of fresh cells.
- 3Confirm expression of FUCCI in cells.
- For example, image plates using an Incucyte S3 using the ‘phase’, ‘green’ and ‘red’ channels. See Figure 1 for example images and quantification.
![Fig 1]()
Fig. 1FUCCI cell cycle reporter expression in glioma stem cell lines. A, Sequential images obtained across time points of two distinct patient derived glioma stem cells (E13 and E57) stably expressing the FUCCI cell cycle reporter. Scale bar is 300 μm. B, FUCCI cell cycle reporter schematic. C, IncuCyte quantification of G1 (Red FUCCI) and G2/M (green FUCCI) cell cycle phases. Mean counts of nuclei expressing either Green or Red FUCCI is shown ± SEM.
- 1Culture, expand and maintain labelled cell population.
- 2Freeze down aliquots of cells for long term storage in liquid nitrogen.
- 3Compare growth rates of FUCCI expressing cells to parental (Supplementary Figure 1).
3.2 Generation of spheroids
- 1Prepare a suspension of your cells (see section 3.1).
- 2Centrifuge cells at 300 × g for 3 minutes and resuspend the cell pellet in an appropriate volume of complete growth media, for example, 5 mL.
- 3Count cells and dilute in media to appropriate concentration for spheroid formation (e.g. 1,000 cells per well – this should be determined for each cell type).
- 4Add 200 μL (96 well) or 50 μL (384 well) of cell suspension to a U-bottom ultra-low attachment plate.
- Note – if possible, don't use outer wells as they tend to grow slower due to plate edge effects over long experiments.
- 1Fill the outer wells with sterile PBS to reduce evaporation.
- 2Centrifuge plate for 10 min at 150 × g with the brake off.
- 3Carefully check a few wells on a microscope. Cells should have gathered in the middle of the well at the bottom (Figure 2).
![Fig 2]()
Fig. 2Glioma Stem Cell (E57) Spheroid formation over 3 days. Brightfield images (IncuCyte S3) obtained at sequential timepoints following spheroid formation. - 4Allow spheroids to form for 2-3 days depending on cell line until the cells have aggregated into a tight ball.
- 5Spheroids are typically 300-500 μm in diameter by day 3 depending on the cell line.
3.3 Imaging of spheroids using the IncuCyte S3
- 1Image spheroids using the Incucyte S3 Spheroid software module using ‘Phase + Brightfield’ imaging channels with either 4 × or 10 × objectives and desired time intervals over the duration of the assay (typically 7-11 days) (Fig. 3, Fig. 4).
![Fig 3]()
Fig. 3E57 FUCCI spheroids. Images shown are day 0, 1, 2, and 3 after cell seeding (left). Quantification of FUCCI reporter expression from six spheroids (right). The mean integrated intensity for Green and Red channels is shown ± SEM.![Fig 4]()
Fig. 4FUCCI spheroid cell cycle analysis. A. E57 (glioma stem cells representative of the mesenchymal subtype) treated with paclitaxel 1 μmol L−1 and vehicle control (DMSO, 0.1% v/v). B. Quantification of spheroid area and FUCCI expression. C, E13 (glioma stem cells representative of classical subtype) treated with paclitaxel 1 μmol L−1 and vehicle control (DMSO, 0.1% v/v). D, Quantification of spheroid area and FUCCI expression. For B and D, mean ± SEM is shown (n = 6 spheroids).
- Note – if imaging FUCCI expressing cell lines, also image using the ‘Green’ and ‘Red’ channels using the default exposure settings (Green = 300ms; Red = 400ms).
- 1Perform image analysis using the IncuCyte S3 Spheroid Analysis Module.
- iSelect 3-5 representative spheroids from your experiment to train the image segmentation on.
- iiUsing the Brightfield image, segment the spheroid using the segmentation slider.
- iiiApply filters to clean up segmentation, for example filter out small objects by setting a minimum object size using the ‘Area’ filter.
- ivIf imaging spheroids expressing FUCCI, perform a ‘Top-Hat No Mask’ background subtraction (radius = 500 μm) of the green and red image.
- vPreview settings across all training images and correct any images with poor spheroid segmentation.
- viSave analysis definition and apply to all wells and all time points.
- viiExample data is shown in Figure 4.
- i
- 2Optional (for unlabelled spheroids), perform dead cell labelling at the end of the experiment by adding ethidium homodimer-1 (10 μmol L−1 final concentration; Sigma-Aldrich, 46043-1MG-F) and incubating for 30 minutes. Image plate on the IncuCyte S3 using the red filter channel (Figure 5).
![Fig 5]()
Fig. 5Live/dead spheroid assay. A, Brightfield images of E57 spheroids treated with paclitaxel (1 μmol−1) or staurosporine (1 μmol L−1) for 3 days. On day 3 spheroids were treated with 10 μmol L−1 ethidium homodimer-1 for 30 minutes to label the non-viable cells (red). B, Quantification of non-viable cells after 3 days of paclitaxel or staurosporine treatment. n = ≥3 spheroids, mean ± SEM is shown.
3.4 Troubleshooting
- •Poor viral transduction – care should be taken to optimise viral transduction of different cell lines. This is especially important when working with the patient-derived GBM stem cell lines described here. Validation of stable expressing FUCCI cell lines should be undertaken to ensure the retention of the parental cell lines characteristics such as growth and morphology (Supplementary Figure 1).
- •Cell line makes poor spheroids or does not aggregate – some optimisation of culture conditions may be required for new cell lines. Laminin or basement membrane extract (final concentration of 2%) can be added to cells when forming spheroids to increase cellular aggregation. Alternatively co-culture with other cell lines, for example non-cancerous cells such as astrocytes for brain derived cell lines or fibroblasts for other types of cancer, may improve spheroid formation.
4. Expected results
E57- and E13-FUCCI expressing spheroids treated with the microtubule destabilising agent paclitaxel, displayed a decrease in the FUCCI-red channel after 6 days of treatment, indicating a loss of cells in the G1 phase and a shift towards the G2/M stage of the cell cycle (Figure 4). An increase in the FUCCI-green signal, indicative of cell-cycle arrest at late G2/M phase, was observed in the E57 spheroids after 6 days (144 hours; Figure 4A) of paclitaxel treatment while E13-FUCCI spheroids displayed a rapid peak of FUCCI-green accumulation after 24-48 hours of paclitaxel treatment (Figure 4B and D), highlighting potential differences between the cell lines with respect to how they mechanistically respond to paclitaxel treatment. Assessment of the variability of the spheroid area and FUCCI intensity measurements using the coefficient of variation demonstrates reasonable values for both cell lines (E13 < 12%, and E57 < 25%) and suitability for high throughput screening (Supplementary Figure 2). To confirm that the observed increase in spheroid area following treatment with paclitaxel is as a result of cell death, spheroids generated with unlabelled parental cells can be stained for dead cells (for example, ethidium homodimer-1) (Figure 5). Comparison of ethidium homodimer-1 staining in paclitaxel and staurosporine (a promiscuous kinase inhibitor which induces apoptosis) treated E57 spheroids highlights the spatial effect of each drug treatment; paclitaxel induces cell death resulting in a halo of cells being ‘shed’ from the spheroid while staurosporine is less selective and induces cell death through-out the spheroid.
5. Discussion
One important road block to making progress in GBM translational research has been overcome in recent years, i.e. the generation of more physiologically-relevant GSC models [
[13]
,[19]
]. The use of 3D tumour spheroid assays has also gained momentum recently in an effort to increase the physiological relevance of cell culture assays and improve success rates of translating preclinical discoveries into positive clinical outcomes [[21]
]. However, protocols for 3D spheroid assays vary significantly between different laboratories and are often not adequately described in research publications or provide sufficient mechanistic information of compound/drug activity. An international consortium (MiSpheroID) has defined and recommended a minimum set of experimental parameters (cell line, spheroid formation method; media; and spheroid size) required to report spheroid research to improve inter-laboratory reproducibility and data mining across multiple studies [[22]
]. Many 3D spheroid studies only report size as the primary assay endpoint, however spheroid size provide little information on compound mechanism-of-action and can sometimes be misleading. For example, a compound treatment which induces cell death may disrupt the integrity of the spheroid resulting in an increase in spheroid size which could be misinterpreted as an increase in cell proliferation. In addition, compounds which induce cell cycle arrest and not immediate cell death may result in a temporary accumulation of cycling cells in the outer core of the spheroid in the G2/M phase of the cell cycle which could result in increased spheroid size and again be misinterpreted as an induction of cell proliferation. Here we present a detailed method, adhering to MiSpheroID guidelines, for monitoring the kinetic changes in the cell cycle following compound treatment in patient derived glioblastoma stem cell lines. Stable expression of the FUCCI cell cycle reporter in these glioblastoma stem cell lines allows kinetic evaluation of the cells cultured in a 3D environment in situ, retaining spatial resolution (e.g. location of non-viable cells within the spheroid) that would otherwise be lost following spheroid dissociation for conventional cell cycle analysis by fluorescent activated cell sorting. The method provides both temporal and spatial resolution of cell-cycle progression within 3D spheroids to more precisely inform on compound mechanism-of-action relative to spheroid size measurements alone. The method can be applied to panels of genetically distinct patient derived glioblastoma cell lines in multiwell plate formats to support high throughput phenotypic screening and pharmacogenomics and personalized medicine studies. Future developments and application of the method to incorporate novel microfluidic devices [[23]
] and new high speed light-sheet microscopes can further enhance throughput and analysis at single cell resolution [[24]
].6. Notes
FUCCI reporter: The Fluorescent Ubiquitination-based Cell Cycle Indicator allows visualisation of real time cell cycle progression by utilising temporally expressed proteins specific for each phase. The concept was originally developed by Sakaue-Sawano et al., [
[25]
] and variations have evolved since, however, most still use Cdt1 which accumulates in G1 and Geminin which accumulates in S/G2/M. Here we have used a commercially available construct Incucyte Cell Cycle Lentivirus EF1, Puro, (Sartorius, 4779) which uses mKate2 tagged Cdt1 (λabs/λem = 483/506nm) and TagGFP2 tagged Geminin (λabs/λem = 588/633mn). The result is cells with nuclei appearing red in G1, and green in S/G2/M.Glioma Stem Cells: Molecularly annotated human patient derived glioma stem cells are available from the Glioma Cellular Genetics Resource: http://gcgr.org.uk/#Home
Declaration of interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:
Neil Carragher reports financial support was provided by Cancer Research UK. Neil Carragher reports financial support was provided by Brain Tumour Charity. Neil Carragher reports a relationship with Cancer Research UK that includes: funding grants and travel reimbursement.
Acknowledgement
This work was supported by a Cancer Research UK Accelerator award (A29368) and a joint Cancer Research UK (A28596) and The Brain Tumour Charity award (GN-000676) to N.O.C. We also acknowledge Professor Steven Pollard and Dr Gillian Morrison and the CRUK Glioma Cellular Genetic resource for the provision of patient derived GSCs.
Appendix. Supplementary materials
References
- Comparison of glioblastoma (GBM) molecular classification methods.Semin Cancer Biol. 2018; 53: 201-211
- Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1.Cancer Cell. 2010; 17: 98-110
- Tumor evolution of glioma-intrinsic gene expression subtypes associates with immunological changes in the microenvironment.Cancer Cell. 2017; 32 (42-56 e6)
- Single-cell RNA-seq highlights intratumoral heterogeneity in primary glioblastoma.Science. 2014; 344: 1396-1401
- Intratumor heterogeneity in human glioblastoma reflects cancer evolutionary dynamics.Proc Natl Acad Sci U S A. 2013; 110: 4009-4014
- Glioma stem cells promote radioresistance by preferential activation of the DNA damage response.Nature. 2006; 444: 756-760
- A restricted cell population propagates glioblastoma growth after chemotherapy.Nature. 2012; 488: 522-526
- Cancer stem cells in glioblastoma.Genes Dev. 2015; 29: 1203-1217
- Cancer stem cell analysis and clinical outcome in patients with glioblastoma multiforme.Clin Cancer Res. 2008; 14: 8205-8212
- The somatic genomic landscape of glioblastoma.Cell. 2013; 155: 462-477
- Hypoxia in the glioblastoma microenvironment: shaping the phenotype of cancer stem-like cells.Neuro Oncol. 2017; 19: 887-896
- Genome-wide CRISPR-Cas9 screens expose genetic vulnerabilities and mechanisms of temozolomide sensitivity in glioblastoma stem cells.Cell Rep. 2019; 27 (971-986 e9)
- Glioma stem cell lines expanded in adherent culture have tumor-specific phenotypes and are suitable for chemical and genetic screens.Cell Stem Cell. 2009; 4: 568-580
- Time-lapse 3-D measurements of a glucose biosensor in multicellular spheroids by light sheet fluorescence microscopy in commercial 96-well plates.Sci Rep. 2016; 6: 37777
- 3D bioprinting for high-throughput screening: drug screening, disease modeling, and precision medicine applications.Appl Phys Rev. 2019; : 6
- A fully automated high-throughput workflow for 3D-based chemical screening in human midbrain organoids.Elife. 2020; : 9
- A directional 3D neurite outgrowth model for studying motor axon biology and disease.Sci Rep. 2021; 11: 2080
- A synergistic anticancer FAK and HDAC inhibitor combination discovered by a novel chemical-genetic high-content phenotypic screen.Mol Cancer Ther. 2020; 19: 637-649
- Digital transcriptome profiling of normal and glioblastoma-derived neural stem cells identifies genes associated with patient survival.Genome Med. 2012; 4: 76
- In vitro expansion of fetal neural progenitors as adherent cell lines.in: Reynolds B.A. Deleyrolle L.P. Neural Progenitor Cells: Methods and Protocols. Humana Press, Totowa, NJ2013: 13-24
- 3D tumor spheroid models for in vitro therapeutic screening: a systematic approach to enhance the biological relevance of data obtained.Sci Rep. 2016; 6: 19103
- MISpheroID: a knowledgebase and transparency tool for minimum information in spheroid identity.Nat Methods. 2021; 18: 1294-1303
- Drug screening of biopsy-derived spheroids using a self-generated microfluidic concentration gradient.Sci Rep. 2018; 8: 14672
- Dual-view oblique plane microscopy (dOPM).Biomed Opt Express. 2020; 11: 7204-7220
- Visualizing spatiotemporal dynamics of multicellular cell-cycle progression.Cell. 2008; 132: 487-498
Article info
Publication history
Published online: September 15, 2022
Accepted:
September 13,
2022
Received in revised form:
September 3,
2022
Received:
August 6,
2022
Identification
Copyright
© 2022 The Authors. Published by Elsevier Inc. on behalf of Society for Laboratory Automation and Screening.
User license
Creative Commons Attribution (CC BY 4.0) | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution (CC BY 4.0)
Permitted
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article
- Reuse portions or extracts from the article in other works
- Sell or re-use for commercial purposes
Elsevier's open access license policy
13694-4&id=gr1.jpg){kind=link}
13694-4&id=gr2.jpg){kind=link}
13694-4&id=gr3.jpg){kind=link}
13694-4&id=gr4.jpg){kind=link}
13694-4&id=gr5.jpg){kind=link}