Formation of Uniform and Reproducible 3D Cancer Spheroids in High-Throughput Plates

Mimic life in three dimensions

(See a list of the products featured in this article.)

The physiological characteristics of a traditional monolayer culture of cells growing on a flat 2D tissue culture substrate can differ considerably from those of cells in a 3D environment. Increasingly, spheroids grown in 3D cell culture are being used in cancer research to more closely mimic the environment associated with tumors. These 3D cancer spheroids have several physiological traits in common with tumors, including overall morphology, formation of cell–cell contacts, decreased proliferation rates, increased survival rates, and a hypoxic core. However, variability in cancer spheroid formation has been a persistent problem for researchers in this field. The reproducibility of spheroid formation appears to be linked to medium composition and volume, cell density, duration in cell culture, and most importantly, the interactions of cells with the culture dish itself. More consistent results can reportedly be achieved using high-quality cultureware with low cell-binding characteristics.

Here we describe the formation and characterization of uniform and reproducible 3D cancer spheroids in vitro using Thermo Scientific Nunclon Sphera plates. These 96-well U-bottom plates have a polymer-coated culture surface that inhibits the binding of extracellular matrix (ECM) proteins, which typically mediate cell adhesion. Furthermore, cancer spheroids form quickly in the Nunclon Sphera wells without the formation of “satellite colonies”, demonstrating superior quality when compared with spheroids formed in methylcellulose- containing media on nontreated (also called non–tissue culture treated or non–TC treated) polystyrene plates. Cell viability and cell function can be conveniently evaluated in situ using fluorescence- based and colorimetric assays; drug treatment can also be administered directly to spheroids growing in Nunclon Sphera plates. The consistent formation of cancer spheroids in the Nunclon Sphera plate makes it an ideal platform for modeling 3D tumor growth for cell-based drug discovery procedures, co-culture studies, and high-throughput screening.

The Nunclon Sphera plate surface exhibits extremely low adsorption of ECM proteins

For anchorage-dependent or adherent cells to form spheroids in suspension, the culture vessel must promote the aggregation of cells through cell–cell binding while preventing the ECM from binding to the culture vessel surface. Figure 1 demonstrates that both collagen I and fibronectin adsorption are minimal on the Nunclon Sphera plate surface when compared with adsorption on the Thermo Scientific Nunclon Delta plate surface (the standard cell or tissue culture treated, also called TC treated, polystyrene surface). These ECM protein–binding assays suggest that, unlike the TC treated surface of cell culture vessels, the Nunclon Sphera surface has minimal binding interactions with the ECM, which discourages cells from attaching to the cultureware.

Figure 1. Adsorption of ECM proteins collagen and fibronectin to the Nunclon Sphera surface is extremely low compared with adsorption to the standard cell or tissue culture treated surface.Thermo Scientific Nunclon Sphera 96-well U-bottom plates (Cat. No. 174925) and Thermo Scientific Nunclon Delta 96-well plates (also called TC treated plates, Cat. No. 143761) were each coated with 100 μL/well of a FITC bovine collagen type I conjugate (24 μg/mL in D-PBS) and incubated for 24 hr at 2–8°C. Another set of plates were coated with a TAMRA fibronectin conjugate (20 μg/mL in D-PBS) and incubated for 16 hr at room temperature. After the plates were washed three times with 200 μL/well PBS-T (PBS with 0.05% Tween 20), the fluorescence intensity was detected using excitation/ emission of 495/525 nm for FITC collagen or 543/570 nm for TAMRA fibronectin. * p < 0.01 for Student’s t-test.

The Nunclon Sphera plate surface is superior for culturing cancer spheroids

To demonstrate the formation of 3D cancer spheroids in Nunclon Sphera plates, HCT 116 (human colon carcinoma) cells were seeded into Nunclon Sphera 96-well U-bottom plates in complete DMEM. Cells were similarly seeded into nontreated 96-well U-bottom plates in complete DMEM containing 3% methylcellulose. Figure 2A shows that, after 112 hr incubation, HCT 116 cancer spheroid formation in the Nunclon Sphera plates exhibited more uniform shape, better-defined edges, cleaner backgrounds, and fewer “satellite colonies” at all seeding densities when compared with spheroid formation in the nontreated plate, leading to higher-quality spheroids. Moreover, at the lowest seeding density of 100 cells/well, the HCT 116 cancer spheroids formed after only 18 hours of incubation in the Nunclon Sphera plate (Figure 2B). These findings show that uniform and reproducible spheroids are formed after less than a day of incubation when using Nunclon Sphera 96-well U-bottom plates and complete DMEM.

Figure 2. Comparison of cancer spheroid formation using Nunclon Sphera plates vs. nontreated plates containing methylcellulose. HCT 116 cells, maintained in Thermo Scientific Nunclon Delta cell culture flasks (Cat. No. 136196), were seeded in Thermo Scientific Nunclon Sphera 96-well U-bottom plates (Cat. No. 174925) at densities of 100–3,000 cells/well in 200 μL/well Gibco DMEM (with high glucose, GlutaMAX supplement, and pyruvate) containing 10% FBS, 1X MEM Non-Essential Amino Acids, 100 U/mL penicillin–streptomycin, and 25 mM HEPES. Traditional plates not treated for cell or tissue culture (non–TC treated plates) were similarly seeded in complete DMEM medium that also contained 3% methylcellulose. Plates were briefly centrifuged at 250 x g for 5 min and then incubated at 37°C and 5% CO2; cells were re-fed every 72 hr by carefully removing 100 μL of medium from each well and replenishing with 100 μL of fresh growth medium. Formation and growth of spheroids were imaged. (A) After 112 hr incubation, cancer spheroids grown in Nunclon Sphera plates show more uniform shape, better-defined edges, and cleaner backgrounds than those grown in nontreated plates at all seeding densities. (B) At a seeding density of 100 cells/well, the early time-course of spheroid formation in the Nunclon Sphera plate reveals spheroids after only 18 hr and many fewer satellite colonies than in the nontreated plate. Images used with permission from Helmut Dolznig, Institute of Medical Genetics, Medical University of Vienna.

Assaying the cell health of cancer spheroids

To monitor spheroid growth over time, the A549 (human lung carcinoma) and HCT 116 cancer cell lines were seeded at different densities in Nunclon Sphera plates and then cultured for 2 weeks. Both cell types showed adequate spheroid growth, as demonstrated by size measurements (Figure 3A) throughout the incubation period. Additionally, the cell health of A549 and HCT 116 spheroids was assessed using the PrestoBlue cell viability assay, which detects the reducing power of live cells (Figure 3B), and the LIVE/DEAD viability/cytotoxicity assay, which detects plasma membrane integrity and intracellular esterase activity (Figure 3C). Each of these assays showed that spheroids from both the A549 and HCT 116 cell lines were viable and healthy, indicating that Nunclon Sphera 96-well U-bottom plates are reliable and convenient tools for both routine and high-throughput cancer spheroid applications.

To further demonstrate the health and vitality of the cancer spheroids grown in the Nunclon Sphera plates, we used a third fluorescence- based viability assay, as well as a probe for oxidative stress. First we subjected the spheroids to treatment either with niclosamide, a drug that inhibits oxidative phosphorylation, or with menadione, which induces oxidative stress. The niclosamidetreated spheroids were assayed using the LIVE/DEAD Cell Imaging Kit, which provides a sensitive two-color fluorescence cell viability assay that is optimized for FITC (green) and Texas Red (red) optical filters. Increasing the niclosamide concentration resulted in a higher proportion of dead cells in the spheroids, as expected (Figure 4A–C).

Next, the menadione-treated spheroids were assayed with CellROX Deep Red Reagent, a cell-permeant dye that exhibits bright red fluorescence upon oxidation by reactive oxygen species. With increasing menadione concentration, we observed an increasing number of cells in the spheroid undergoing oxidative stress (Figure 4D–F). These results indicate that both the LIVE/DEAD Cell Imaging Kit and CellROX Deep Red Reagent provide effective and convenient fluorescence assays of cell function in multicellular spheroid structures.

Figure 3. Assessments of spheroid growth, cell health, and cell viability on Nunclon Sphera plates.(A) Growth kinetics of A549 and HCT 116 cancer spheroids grown in Thermo Scientific Nunclon Sphera 96-well U-bottom plates (Cat. No. 174925) at increasing seeding densities were evaluated over a period of 13 days. Data represent the mean ± SD of 3 replicates for each cell number. (B) Spheroid cell health was assessed in situ using PrestoBlue Cell Viability Reagent. After 12–13 days of spheroid culture, 20 μL of 10X PrestoBlue Cell Viability Reagent was added to each well of the Nunclon Sphera plates, which were then incubated at 37ºC and 5% CO2 for an additional 2–5 hr before reading on a fluorescence-based microplate reader (Ex/Em ~560/590 nm). The fluorescence signals were normalized by spheroid size; a higher ratio indicates healthier spheroids. (C) Spheroid cell viability was evaluated using the LIVE/DEAD Viability/Cytotoxicity Kit. After 12–13 days of spheroid culture, the LIVE/DEAD kit reagents were added to the Nunclon Sphera plates, which were then incubated for 30–45 min, rinsed at least 3 times with a half-volume change of D-PBS, and imaged using a fluorescence microscope; live cells fluoresce green, dead cells fluoresce orange. Data were analyzed using ImageJ image analysis software; scale bar = 1,000 μm.

Figure 4. Assaying cell viability and oxidative stress in drug-treated HeLa spheroids. HeLa cells were grown in Gibco MEM , seeded at 600 cells/well in a Nunclon Sphera 96-well U-bottom plate, centrifuged at 200 x g for 5 min, and cultured for 3 days to allow spheroid formation. One set of HeLa spheroids was (A) left untreated, or treated with (B) 100 nM niclosamide or (C) 10 μM niclosamide for 24 hr, and then stained using the LIVE/DEAD Cell Imaging Kit. A second set of HeLa spheroids was (D) left untreated, or treated with (E) 100 nM menadione or (F) 10 μM menadione for 1 hr at 37°C to induce oxidative stress, and then stained with NucBlue Live ReadyProbes Reagent and CellROX Deep Red Reagent. After staining, spheroids were transferred to Nunc glass-bottom dishes with 200 μL pipette tips (with tip ends cut off) and imaged on a Zeiss™ LSM 710 confocal microscope with EC Plan-Neofluar™ 10x/0.3 objective and 488 nm and 561 nm lasers, in addition to the 488/561 nm main beam splitter; stacks were projected using 3D shadow rendering. (A–C) In the spheroids assayed with the LIVE/DEAD Cell Imaging Kit, live cells fluoresce green, and dead cells with permeable membranes fluoresce red. (D–F) In the spheroids stained with CellROX Deep Red Reagent and NucBlue Live ReadyProbes Reagent, cells showing oxidative stress fluoresce red, and live-cell nuclei fluoresce blue.

Detecting hypoxic cores in cancer spheroids

One feature of 3D cancer spheroids that distinguishes them from cells grown in monolayer culture is a low-oxygen (hypoxic) core. Importantly, this hypoxic core is also present in solid tumors in vivo, where cells rapidly outgrow the blood supply, leaving cells at the center of the tumor in an environment with an extremely low oxygen concentration.

To detect low-oxygen conditions in the cancer spheroids, we used Image-iT Hypoxia Reagent, a cell-permeant compound that is nonfluorescent in an environment with normal oxygen concentrations and becomes increasingly fluorescent as oxygen levels are decreased. Because it responds quickly to a changing environment, Image-iT Hypoxia Reagent can serve as a real-time oxygen detector, with a fluorescent signal that increases as atmospheric oxygen levels drop below 5% and decreases if oxygen concentrations increase up to 5%. Figure 5 shows HeLa cells cultured in a Nunclon Sphera 96-well U-bottom plate for 2 days in complete medium, stained in situ with Image-iT Hypoxia Reagent, and counterstained with NucBlue Live ReadyProbes Reagent. The red-fluorescent staining beneath the spheroid surface provides evidence of a hypoxic core, mimicking the physiological conditions inside a tumor.

Figure 5. Assessment of the hypoxic core in a single HeLa spheroid. HeLa cells (250 cells/well) were cultured on Nunclon Sphera 96-well U-bottom plates (Cat. No. 174925) for 2 days in complete medium. The spheroids were then incubated in situ with 5 μM Image-iT Hypoxia Reagent (red) for 3 hr, and nuclei were counterstained with NucBlue Live ReadyProbes Reagent (blue). The stained spheroids were transferred by pipetting using wide-bore pipette tips to a Nunc Glass Bottom Dish (12 mm, Cat. No. 150680), and images were taken on a confocal microscope.

Bring 3D spheroid culture to your lab

We have demonstrated that the surface of the Nunclon Sphera 96-well U-bottom plates exhibits extremely low ECM-binding properties, effectively discouraging cell attachment and promoting spheroid formation. These plates have been shown to support uniform and reproducible formation and growth of cancer spheroids across commonly used cancer cell lines, and are compatible with in situ fluorescence assays of cell health. The presence of hypoxic cores in cancer spheroids indicates that 3D cancer spheroid culture in Nunclon Sphera plates presents an effective in vitro system for modeling tumor growth.