Advancements in the Metabolic Profiling of Three-Dimensional Brain Tumor Spheroids for Drug Screening

Summary

Here, we establish a metabolic profiling platform integrating three-dimensional (3D) glioma spheroids and imaging-based normalization via extracellular flux analysis. This approach enables more accurate assessment of tumor metabolic responses to drug treatments and provides insights into potential mechanisms of resistance.

Abstract

Brain tumors, especially gliomas, are challenging to treat because of their aggressive nature, complex tumor microenvironment, and resistance to conventional therapies. Traditional two-dimensional (2D) cell cultures often fail to replicate the true tumor environment, leading to inaccurate predictions of drug efficacy. Extracellular flux analysis technology, typically used for real-time metabolic analysis in 2D cultures, measures key metabolic parameters, such as the extracellular acidification rate (ECAR) and oxygen consumption rate (OCR), providing insights into cellular metabolism. The use of 3D models represents a significant advancement, as they more accurately mimic the in vivo tumor environment. The extracellular flux analyzer was adapted to three-dimensional (3D) glioma cell models, enabling the analysis of critical metabolic pathways, including glycolysis and oxidative phosphorylation, in a more physiologically relevant context.

U87 cells were seeded at appropriate densities in a 96-well low-attachment plate and cultured for 5 days. On day 5, 3D spheroid formation was observed via high-content imaging. The successfully formed spheroids were then transferred to a metabolic assay plate coated with poly-L-lysine for metabolic analysis. To improve the accuracy of these measurements, high-content imaging systems assess 3D cell size, allowing for precise normalization of extracellular flux data and minimizing metabolic variations due to differences in cell size. This integrated approach provides a more reliable analysis of glioma cell metabolic responses to drug treatments, revealing potential mechanisms of drug resistance. Ultimately, this methodology offers valuable insights into the metabolic dynamics of gliomas and supports the development of novel, clinically relevant therapeutic strategies.

Introduction

Gliomas, a prevalent and aggressive form of brain tumor, pose significant clinical challenges because of their rapid progression, highly heterogeneous tumor microenvironment, and marked resistance to conventional therapies¹. These complexities necessitate more physiologically relevant experimental models to improve the accuracy of therapeutic evaluations. Traditional two-dimensional (2D) cell culture systems fail to recapitulate the in vivo tumor architecture and cell-cell interactions², often leading to misleading conclusions regarding drug efficacy and tumor biology³.

To address these limitations, three-dimensional (3D) tumor models have emerged as a superior alternative, offering enhanced biological relevance by mimicking the structural and functional characteristics of tumors in vivo^4,5. Concurrently, metabolic reprogramming is increasingly recognized as a hallmark of glioma progression and treatment resistance^6,7,8. Extracellular flux analysis has become a widely adopted technique for real-time measurement of key metabolic parameters, such as the extracellular acidification rate (ECAR, measuring proton efflux primarily from glycolytic lactate production) and the oxygen consumption rate (OCR, quantifying mitochondrial oxygen utilization in oxidative phosphorylation), providing critical insights into glycolytic and oxidative phosphorylation pathways^9,10,11.

In this study, extracellular flux analysis was adapted to analyze 3D glioma spheroids, enabling the investigation of metabolic pathways in a context that closely resembles in vivo tumor conditions. U87 glioma cells were seeded at a density of 1,500 cells per well in transparent round-bottom ultra-low attachment 96-well plates and cultured for 5 days at 37 °C with 5% CO₂ to facilitate spheroid formation. High-content imaging was employed to confirm the spheroid morphology and assess the cell size. Prior to metabolic analysis, dense and circular 3D cell spheroids with a diameter of approximately 220 µm were transferred to assay plates pre-coated with poly-L-lysine to ensure adhesion and measurement stability. Importantly, normalization of metabolic data on the basis of spheroid size was conducted via integrated high-content imaging, reducing variability and enhancing data reliability.

This integrated platform, combining 3D culture, extracellular flux analysis technology, and imaging-based normalization, enables a more accurate assessment of glioma metabolic responses to therapeutic interventions. This study provides a robust framework for studying the metabolic underpinnings of drug resistance and contributes to the development of more effective, clinically translatable treatment strategies for gliomas.

Protocol

NOTE: The complete specifications of all reagents and consumables are detailed in the Table of Materials.

1. 2D cell culture

1. Maintain U87 cells in minimum essential medium (MEM) supplemented with 10% (v/v) fetal bovine serum (FBS) under sterile conditions. Perform standard passaging procedures once the cells reach 80% confluence.
   NOTE: Throughout the culture process, closely monitor the cell status and exclude cells exhibiting abnormal growth or morphological changes from subsequent experiments.
2. Seed U87 cells into 10 cm cell culture dishes and incubate in a sterile cell culture incubator at 37 °C with 5% (v/v) CO₂ and 90-95% relative humidity.
   NOTE: The medium should be replaced regularly to maintain optimal growth conditions.

2. 3D cell spheroid formation

1. Experimental preparation
   1. Review the entire workflow in Figure 1 before commencing experimental procedures.
      CAUTION: Strictly adhere to aseptic techniques throughout the experiment to ensure cell health and prevent contamination.
2. Cell dissociation and counting
   1. Remove the spent medium from the cell culture dish and wash the U87 cells three times with 9 mL of phosphate-buffered saline (PBS) (3 mL per wash). Dispose of used reagents in a 20-25% (v/v) bleach solution to inactivate all bioactive materials.
   2. Add 1 mL of trypsin solution to the dish and incubate at 37 °C for 2-3 min. Neutralize trypsin with 2 mL of DMEM containing 10% FBS preheated to 37 °C once all the cells are detached under microscopic observation.
   3. Transfer the cell suspension to a 15 mL centrifuge tube and centrifuge at 200 × g for 3 min. Discard the supernatant, and resuspend the cell pellet in 1 mL of DMEM supplemented with 10% FBS. Pipette the cells gently to ensure a uniform cell suspension. Then count via a hemocytometer or automated cell counter, and record the initial cell count.
      NOTE: Handle the cells gently during washing and media addition to avoid mechanical damage. The pipetting force can be controlled to prevent bubble formation, which can affect cell viability and counting accuracy.
3. Cell suspension dilution
   1. Preheat the unused medium to 37 °C. Maintain the cell suspensions at room temperature (RT; 25 ± 2 °C) for 30 min.
   2. On the basis of the initial cell count, dilute the cell suspension with DMEM containing 10% FBS to a final volume of 20 mL, ensuring approximately 200,000 cells.
      NOTE: Prior to final dilution, gently pipette the cell suspension to prevent sedimentation. The unused medium was maintained at 37 °C, and the cell suspensions were incubated at RT. If 96 wells with 1,500 cells per well were seeded, a total volume of 22 mL was prepared to account for excess. The diluted cell suspension was transferred to a 50 mL sterile centrifuge tube and mixed thoroughly.
4. Cell seeding in ultra-low attachment 96-well plates
   1. Mix the final cell suspension thoroughly via a 200 µL pipette. Add 200 µL of the suspension (1,500 cells/well) slowly to each well of a transparent round-bottom ultra-low attachment 96-well plate along the edge.
      NOTE: For efficiency, consider the use of a multichannel pipette.
   2. Discard any remaining suspension after all the wells are filled to prevent contamination.
5. Centrifugation
   1. Seal the plate tightly with parafilm to prevent leakage during centrifugation. Centrifuge the mixture at 200 × g for 2 min to pellet the cells at the bottom of each well.
   2. Remove the plate from the centrifuge, carefully remove the parafilm, and place the plate in a 37 °C, 5% CO₂ incubator.

3. Poly-D-Lysine coating of the metabolic assay plates

NOTE: Metabolic assay microplate is a specialized microplate designed for cell spheroid adhesion during extracellular flux analysis.

1. Prepare a coating solution by dissolving 5 mg of poly-D-lysine in 50 mL of sterile water. Mix thoroughly.
2. Add 30 µL of the coating solution to each well of the metabolic assay plate. Ensure the absence of bubbles by gently shaking or aspirating the excess solution.
3. Incubate the plate and cover it for 20 min at RT. Aspirate the coating solution, and wash the wells twice with 200 µL of sterile water.
4. Air dry the plate for at least 30 min and then heat at 37 °C without CO₂ for 30 min to stabilize the coating.
5. Add 175 µL of preheated medium to each well, and keep the plate at 37 °C in a CO₂-free incubator until it is ready for cell spheroid transfer.

4. Monitoring spheroid growth

1. Transfer the spheroid plates to a high-content imaging system after 5 days of culture.
2. Configuration imaging parameters: Set the temperature to 37 °C and the CO₂ to 5%, and adjust the plate type, channels, and focus range.
3. Acquire spheroid images via a confocal microscope with 10x objectives and an imaging software.
4. Identify wells containing intact spheroids (e.g., Figure 2A-C) and record their positions for transfer. Exclude wells with irregular spheroids (e.g., Figure 2D)."

5. Spheroid treatment

NOTE: Drug treatment was performed on mature 3D spheroids to assess metabolic responses.

1. Spheroid group assignment
   1. Assign intact spheroids to two groups:
      Control group: 6 replicate wells
      Treatment group: 6 replicate wells (fenofibrate-treated)
2. Drug solution preparation
   1. Prepare the fenofibrate working solution in fresh pre-warmed medium at the optimal concentration (e.g., 100 µM).
      NOTE:Treatment groups were reconstituted in 100% DMSO to create stock solutions, which were subsequently diluted in fresh pre-warmed medium. The control groups were administered the same final concentration of DMSO (0.1% v/v) as was present in the drug-treated samples. The final concentration was determined via dose-response assays.
3. Medium replacement
   1. Aspirate 100 µL of medium from each well slowly along the wall.
      NOTE: Avoid contacting the well bottom to prevent spheroid disturbance.
   2. Add 100 µL of fresh medium to the control wells.
   3. Add 100 µL of fenofibrate working solution to the treatment wells.
4. Posttreatment incubation
   1. Return the plate gently to 37 °C in a 5% CO₂ incubator. Culture for 72 h.
5. Morphological assessment
   1. Assess the integrity of the spheroids following the imaging procedure in step 4.

6. Transfer of cell spheroids

1. Carefully remove the spheroid-containing low-attachment plate from the imaging system and place it on a sterile surface within a biosafety cabinet. Place the poly-L-lysine-coated assay plate adjacent to each other for transfer.
2. Trim the tip of a 20 µL pipette to facilitate gentle aspiration of the cell spheroids without damage.
3. Gently aspirate each intact spheroid from the bottom of the well and transfer it to the corresponding well of the cell plate, allowing it to settle for 20 s. Confirm adhesion by the absence of floating fragments under 10× microscopy.
4. Repeat for all spheroids and verify transfer efficiency via the imaging system described in step 4.

7. Metabolic assay

1. Sample degassing
   1. Place the drug-treated spheroid assay plate into a preequilibrated 37 °C hydration chamber devoid of CO₂.
      NOTE: Continuous degassing was performed for 1 h with strict environmental control to ensure metabolic measurement accuracy.
2. Reagent preparation and drug administration
   1. Parallely prepare the reagent and load the drug during the 1-h degassing period.
      NOTE: The primary objective of this experiment is to assess mitochondrial stress; all procedures must be meticulously aligned with this goal. Specific protocols must be followed for distinct experimental types to guarantee data accuracy and reliability.
   2. Drug preparation: Use the following compounds:
      Oligomycin (mitochondrial complex V inhibitor)
      Carbonyl cyanide-4 (trifluoromethoxy) phenylhydrazone (FCCP, uncoupler)
      A mixture of rotenone (a complex I inhibitor) and antimycin A (a complex III inhibitor)
      ​NOTE: Drugs were obtained from the Mitochondrial Stress Test Kit used in this study. Some compounds are biohazardous; strict biosafety protocols (personal protective equipment and emergency preparedness) must be used.
   3. Drug formulation
      1. Prepare the stock solutions of the powdered drugs as shown in Table 1.
      2. Dilute the stock solutions to the working concentrations specified in Table 2.
         NOTE: The solutions are vortexed or agitated thoroughly to ensure homogeneity across concentration gradients.
3. Loading compounds into detection probe cartridges
   NOTE: Detection probe cartridge is a disposable cartridge containing biosensor probes for real-time measurement of oxygen and pH.
   1. Probe cartridge activation (must precede compound loading):
      NOTE: Ensure sensor activation by completing these steps at least 12 h before use:
      1. Add 200 µL of the calibrant solution to each well of the hydration plate.
      2. Position the detection probe cartridge carefully onto the hydration plate, ensuring complete immersion of all the probes in the calibrant solution.
      3. Incubate the assembled unit in a 37 °C CO₂-free incubator for 12-24 h to activate the sensors.
         NOTE: Incomplete immersion or insufficient hydration causes measurement drift (±15% OCR error).
   2. Use multichannel or electronic pipettes to deliver precise volumes into wells, avoiding residual droplets at well edges.
      NOTE: As acute drug stimulation is not needed, follow standard loading sequences.
   3. Place the probe plate on a stable surface after loading to prevent disturbances during transport.
      NOTE: Loading accuracy significantly impacts the results.
4. Software configuration
   1. Launch software and select the mitochondrial stress test assay protocol.
   2. Assign groups according to the predefined experimental grouping.
   3. Set the file name and click Start Run.
5. Calibration of detection probe cartridges.
   1. Gently insert the loaded probe plate into the instrument slot per system prompt.
   2. Initiate calibration (15-20 min). Maintain stable conditions; avoid vibrations.
   3. Remove the hydration plate promptly post-calibration (following the on-screen instructions).
      NOTE: Remove the plate lid prior to calibration.
6. Metabolic profiling of 3D spheroids
   1. Transfer the degassed 3D spheroid assay plate to the instrument measurement slot. Ensure precise alignment with the probe array by confirming the following: (i) Plate corner match slot guides; (ii) Well A1 aligns with the indexed position; (iii) No visible gap between the plate and instrument surface
   2. Click Start to initiate mitochondrial stress assays (duration: ~1.5 h). Monitor instrument parameters throughout.
      NOTE: Remove 3D spheroid assay plate lids before measurement.
7. Data analysis
   1. Imaging analysis
      1. Transfer 3D spheroid assay plates to a high-content imaging system.
         NOTE: Images were acquired as described in step 4. Imaging revealed that the majority of the spheroids remained intact, whereas a minority sustained damage during the transfer process, as shown in Figure 3.
      2. Analyze the morphological data (e.g., area, length, width) of the cell spheroids via Harmony software version 5.2, following steps 7.7.1.3-7.7.1.7.
      3. Apply the Texture Region module coarsely to separate 3D spheroids from the background by dividing the image content into three clusters on the basis of the texture features of objects in the images.
      4. Apply the Find Image Region module to refine the separation of pre-isolated 3D spheroids via volume-based size characteristics to remove impurities and retain intact spheroids.
      5. Measure the morphological parameters of the 3D spheroids via the Morphology Properties module.
      6. Apply the Select Population module to ultimately filter out genuine 3D cell spheroids.
      7. Export the morphological data results, as shown in Table 3.
   2. Data processing
      NOTE: Normalize the cell data via parameters from the imaging outputs to minimize experimental artifacts.
      1. Open the raw metabolic data via the analysis software.
      2. Select the Normalization Option to access the normalization mode editing interface.
      3. Input the cross-sectional area of spheroids contained in each corresponding well of the assay plate in the Normalization Values field.
      4. Set the scale factor parameter to 1,000,000.
      5. Click the Apply button to finalize the normalization process. The final results are shown in Figure 4.

8. Waste disposal

1. Dispose of all experimental waste (media, reagents, consumables, and cell suspensions) according to biohazard and laboratory waste management protocols. Autoclave or chemically disinfect the bioactive waste before disposal to prevent environmental contamination.

Results

The three most critical steps in the energy metabolism detection process of 3D cell spheroids are the culture of 3D cell spheroids, the spheroid transfer process, and the imaging and quantification of spheroids after detection (Figure 1). Additionally, observing the state of the cell spheroids is highly important. When the experiments were conducted according to the abovementioned experimental procedures, spheroids were detected on the 5^th day (Figure 2

Discussion

The adaptation of extracellular flux analysis to 3D glioma models represents a critical methodological advancement in the study of tumor cell metabolism. While traditional 2D cultures have long served as the foundation for in vitro experimentation, they fail to capture the spatial architecture, cell-cell and cell-matrix interactions, and oxygen and nutrient gradients characteristic of in vivo tumors¹². These features are especially relevant in gliomas, where metabolic reprogrammi...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
10 cm dish	Corning	430167
15 mL centrifuge tube	Corning	430791
50 mL centrifuge tube	Corning	430828
Calcein, AM	Invitrogen	C1430
Countess 3	Thermo Fisher		Automatic Cell Counter
Fetal Bovine Serum	Gibco	A5669701
Harmony 5.2	PerkinElmer		Imaging software
Hoechst 33342	Solarbio	C0031
MEM EBSS Medium	HyClone	SH30024.01
Operetta CLS	PerkinElmer		High-content imaging analysis system
PBS	HyClone	SH30256.01
Poly-D Lysine hydrobromide	Sigma	P7280
Round-bottom ultra-low attachment 96-well plates	Corning	4515
Seahorse Wave Desktop software	Agilent
Seahorse XFe96	Agilent		Extracorporeal metabolic testing machine
Sorvall ST8R	Thermo Fisher		Centrifuge
TRYPSIN 0.25% EDTA	Gibco	25200072
U87MG cell line	National Infrastructure of Cell Resource
XF 1.0 M Glucose Solution, 50 mL	Agilent	103577-100
XF 100 mM Pyruvate Solution, 50 mL	Agilent	103578-100
XF 200 mM Glutamine Solution, 50 mL	Agilent	103579-100
XF Calibrant Solution 500 mL	Agilent	100840-000
XF Cell Mito Stress Test Kit	Agilent	103015-100
XF DMEM Medium, pH 7.4, 500 mL	Agilent	103575-100
XFe96/XF Pro Fluxpak-Detection Probe Array	Agilent	103792-100
XFe96/XF Pro Fluxpak-Metabolic Assay Plates	Agilent	103792-100