A Novel Multiparametric Drug-Scoring Method for High-Throughput Screening of 3D Multicellular Tumor Spheroids Using the Celigo Image Cytometer

Abstract

Three-dimensional (3D) tumor models have been increasingly used to investigate and characterize cancer drug compounds. The ability to perform high-throughput screening of 3D multicellular tumor spheroids (MCTS) can highly improve the efficiency and cost-effectiveness of discovering potential cancer drug candidates. Previously, the Celigo Image Cytometer has demonstrated a novel method for high-throughput screening of 3D multicellular tumor spheroids. In this work, we employed the Celigo Image Cytometer to examine the effects of 14 cancer drug compounds on 3D MCTS of the glioblastoma cell line U87MG in 384-well plates. Using parameters such as MCTS diameter and invasion area, growth and invasion were monitored for 9 and 3 d, respectively. Furthermore, fluorescent staining with calcein AM, propidium iodide, Hoechst 33342, and caspase 3/7 was performed at day 9 posttreatment to measure viability and apoptosis. Using the kinetic and endpoint data generated, we created a novel multiparametric drug-scoring system for 3D MCTS that can be used to identify and classify potential drug candidates earlier in the drug discovery process. Furthermore, the combination of quantitative and qualitative image data can be used to delineate differences between drugs that induce cytotoxic and cytostatic effects. The 3D MCTS-based multiparametric scoring method described here can provide an alternative screening method to better qualify tested drug compounds.

Keywords

3D multicellular tumor spheroid (MCTS)high-throughput growth inhibition invasion assay spheroid viability spheroid apoptosis Celigo image cytometry

Introduction

Two-dimensional (2D) cancer models have been playing an important role in testing the effects of anticancer drug compounds for in vitro assays.^1,2 Although 2D drug-screening assays can be easily performed, they may not accurately represent the complex in vivo tumor microenvironments. Therefore, there has been a significant increase of interest in using three-dimensional (3D) tumor spheroid models for high-throughput (HT) cancer drug-screening assays.³ The 3D tumor spheroid is a highly complex model for HT drug-screening assays. Unlike the 2D models, in which cells can be enumerated and analyzed individually, the 3D model requires a more advanced analysis method to characterize and quantify the effect of drugs on the entire spheroid consisting of clusters of cells. Therefore, multiple orthogonal assays are required to screen the drug compounds to create a multiparametric analysis platform.

Due to the rapidly increasing availability of image cytometry technologies that can be used for analyzing 3D tumor models, numerous studies have been published that investigated the effects of anticancer drugs on tumor spheroids.^3–8 However, a characterization method for ranking the effects of tested drug compounds on 3D multicellular tumor spheroids has yet to be demonstrated. Previously, we demonstrated a novel high-throughput screening method for 3D multicellular tumor spheroids (MCTS) using the Celigo Image Cytometer.² The method described the use of 384-well ultra-low attachment (ULA) U-bottom microplates for spheroid formation, growth inhibition, and viability assays. In this work, we developed a novel multiparametric scoring method for HT screening of 3D MCTS to characterize and identify potential cancer drug candidates. Four experiments were conducted to investigate the effects of 14 drug compounds on U87MG MCTS for growth inhibition,² invasion inhibition,⁹ viability, and apoptosis. First, U87MG MCTS were allowed to form for approximately 4 days and subsequently treated with the 14 drug compounds at different concentrations. The spheroid growth inhibition was determined by measuring the diameters of drug-treated MCTS over time. At the end of the growth inhibition assay, the treated MCTS were stained with calcein AM and propidium iodide (PI) to determine MCTS viability by measuring the fluorescent intensities for both fluorescent probes. In addition, the treated MCTS were stained with caspase 3/7 to determine the spheroid apoptosis level by measuring the fluorescent intensity. Finally, the invasion inhibition was determined by measuring the invasion area of drug-treated MCTS into the extracellular matrix over time.

As previously stated, a method to analyze, characterize, and score the effects of anticancer drugs on MCTS has not been published. The results from spheroid growth inhibition, invasion inhibition, viability, and apoptosis assays were used to perform the multiparametric score calculation to identify highly effective drug compounds. Each tested drug compound displayed a different level of anticancer effect on the MCTS. By calculating 3, 6, and 9 times the standard deviation of the control, each drug was characterized and ranked by assigning a score ranging from 0 to 3, where 3 was the highest effect and 0 was no effect. The proposed drug effectiveness scoring method can improve the efficiency and speed of 3D model-based HT screening assays for identifying more highly qualified compounds for cancer drug discovery.

Materials and Methods

U87MG Cell Culture and Spheroid Formation

U87MG cells (human glioblastoma adherent cell line, ATCC HTB-14) were cultured in EMEM (EBSS) basal media with 10% fetal bovine serum, 2 mM glutamine, 1% nonessential amino acids, and 1 mM sodium pyruvate (Gibco) in TC-treated flasks at 37 °C and 5% CO₂. The cultures were passaged when confluency reached ~80% to 90%. For MCTS formation, the U87MG cells were seeded at 500 cells/well (80 µL/well) in EMEM (EBSS) basal media, 2 mM glutamine, 1% nonessential amino acids, and 1 mM sodium pyruvate in 384-well ULA round-bottom microplates (ULA-384U, Nexcelom Bioscience, Lawrence, MA). Next, the plates were centrifuged at 300 g for 10 min to cluster cells in the bottom of the wells and then incubated for 4 d at 37 °C and 5% CO₂ to allow the MCTS to form.

Drug Compound Preparation

Fourteen drug compounds (Suppl. Table S1) were screened and characterized on U87MG MCTS. The drug compounds were generously donated by Dr. Steven Titus at the High Content Imaging and Discovery at the National Center for Advancing Translational Sciences. The drug compounds selected were sunitinib malate (Cmpd 1),^10,11 paclitaxel (Cmpd 2),^12,13 romidepsin (Cmpd 3),^14,15 panobinostat (Cmpd 4),^16,17 GSK-2126458 (Cmpd 5),¹⁸ cisplatin A (Cmpd 6),¹⁹ bortezomib (Cmpd 7),²⁰ carfilzomib (Cmpd 8),²¹ cisplatin B (Cmpd 9), sorafenib (Cmpd 10),²² trametinib (Cmpd 11),²³ BEZ-235 (Cmpd 12),²⁴ staurosporine (Cmpd 13),²⁵ and 17-AAG (Cmpd 14).² The stock concentration of each drug was 10 mM in 100% DMSO, which was then diluted to 2× working concentrations of 20, 10, 5, 2, 1, 0.2, and 0.1 µM (10, 5, 2.5, 1, 0.5, 0.1, and 0.05 µM final concentrations) for the MCTS growth inhibition, viability, and apoptosis assays. Each drug was also diluted to 2× working concentrations of 10, 5, 2.5, 1.25, and 0.625 µM (5, 2.5, 1.25, 0.625, and 0.3125 µM final concentrations) for the MCTS invasion assays. The drug compound titrations were added to the wells in the 384-well plates on day 4 after the formation of the MCTS using the plate maps for growth inhibition (Suppl. Fig. S1a) and invasion inhibition (Suppl. Fig. S1b).

Celigo Image Cytometer Instrumentation and Software

The Celigo Image Cytometer has been used for numerous cell-based assays in previous publications.^1,2,26 The Celigo used a transmission and epifluorescence optical setup for one bright-field (BF) and four fluorescent (FL) imaging channels (blue, green, red, and far red) to perform plate-based image cytometric analysis. Both BF and FL imaging channels used a high-power light-emitting diode for illumination and excitation. Each FL imaging channel used a specific fluorescent filter set for the corresponding colors: blue (EX: 377/50 nm, EM: 470/22 nm), green (EX: 483/32 nm, EM: 536/40 nm), red (EX: 531/40 nm, EM: 629/53 nm), and far red (EX: 628/40 nm, EM: 688/31 nm). The combined optics and digital imaging allowed variable imaging resolutions from 1 to 8 µm²/pixel. The proprietary optics setup captured highly uniform images of the entire well on a standard microplate. The F-Theta lens and galvanometric mirrors’ imaging optics can perform a rapid image-capturing process, in which 96 whole-well images can be captured in less than 4 min in bright field. The Celigo can perform autofocusing; however, for tumor spheroids, the focal plane was manually registered by the user, mainly because of the large thickness of the spheroids.

The Celigo software allowed the selection of different applications for analysis of 3D MCTS. The “Tumorsphere 1,” “Tumorsphere Migration,” and “Tumorsphere 1 + 2 + Mask” applications were used to measure spheroid diameter, invasion area, fluorescent intensities of calcein AM, PI, and caspase 3/7, respectively. The results were exported to Excel for further calculation for the ranking of anticancer effects of each tested drug compound.

MCTS Growth Inhibition Assay

The MCTS growth inhibitory effects of each drug compound were investigated by measuring the changes in MCTS diameters over 9 d after the drug induction. U87MG cells were seeded at 500 cells/well (80 µL/well) in two 384-well ULA round-bottom microplates and allowed to form MCTS for 4 d. On day 4, half of the medium (40 µL) was removed from each well, and 40 µL of the 2× drug compound titrations (final 10, 5, 2.5, 1, 0.5, 0.1, and 0.05 µM) and a 0.02% DMSO control (final 0.01%) were added in triplicate wells (Suppl. Fig. S1a). The two plates were then incubated for 9 d (~210 h). Each microplate was scanned and analyzed using the image cytometer on day 0, 3, 5, 7, and 9 posttreatment. On each day of image acquisition, the MCTS diameters were measured for the control and treated samples ( Figure 1a ). The measured MCTS diameter results were plotted with respect to time as well as drug compound concentrations.

Figure 1.

Examples of bright-field and fluorescent images acquired for growth inhibition, viability, apoptosis, and invasion assays. (a) Bright-field and fluorescent images used to measure multicellular tumor spheroid (MCTS) diameter and fluorescent intensities of calcein AM and propidium iodide for control and treated MCTS. (b) Bright-field, caspase 3/7, and Hoechst images used to measure fluorescent intensities of caspase 3/7 for control and treated MCTS. (c) Bright-field images used to measure invasion area for control and treated MCTS.

Fluorescent Staining and Image Analysis Protocol for MCTS Viability

One of the prepared 384-well ULA microplates for growth inhibition was obtained after the last growth inhibition assay scan on day 9. The plate served as the endpoint measurement for MCTS viability. Two fluorescent stains, calcein AM and PI from Nexcelom Bioscience, were used to stain the U87MG MCTS to determine viability. A stock staining solution was prepared by mixing 10 µL of 1 mM calcein AM and 40 µL of 1 mg/mL PI in 20 mL of phosphate-buffered saline (PBS). The U87MG MCTS were stained by gently removing 40 µL of media and replaced with 40 µL of the stock staining solution. The plate was then incubated at 37 °C and 5% CO₂ for 60 min prior to performing the Celigo image cytometric analysis.

Immediately after the incubation, the plate was scanned using the “Tumorsphere 1 (Green) + 2 (Red) + Mask (BF)” application in the software. The exposure times for the green, red, and BF channels were 5000, 60,000, and 9000 µs, respectively ( Fig. 1a ). For the analysis, the MCTS were first identified in the bright-field images, and then the fluorescent intensities of calcein AM and PI were measured and exported directly into Excel. Next, the exported average calcein AM and PI fluorescent intensities were used to calculate $\frac{A V E c a l c e i n A M F L I n t e n s i t y}{A V E P I F L I n t e n s i t y}$ , which was used to determine the cytotoxicity effect of the drug compounds on the U87MG MCTS. High calcein AM and low PI FL intensity (high ratio value) indicate higher viability in comparison with low calcein AM and high PI FL intensity (low ratio value). The calculated ratios for all drugs were plotted for comparison to the control.

Fluorescent Staining and Image Analysis Protocol for MCTS Apoptosis

The second prepared 384-well ULA microplate for growth inhibition was also obtained on day 9 after the drug treatment. The plate served as the endpoint measurement for MCTS apoptosis. Two fluorescent stains, caspase 3/7 and Hoechst 33342 (Hoechst) from Nexcelom Bioscience, were used to stain the U87MG MCTS to determine the level of caspase 3/7 activation. The caspase 3/7 substrates are bound to high-affinity DNA fluorescent dyes. When the substrates are cleaved by caspase 3/7 during apoptosis, the fluorescent dyes migrate to the nucleus to bind to the DNA and then emit green fluorescence. A stock staining solution was prepared by mixing 80 µL of 1 mM caspase 3/7 and 8 µL of 10 mg/mL Hoechst in 20 mL of PBS. The U87MG MCTS were stained by gently removing 40 µL of media and replaced with 40 µL of the stock staining solution. The plate was then incubated at 37 °C and 5% CO₂ for 60 min prior to being scanned and analyzed using the Celigo analysis software.

Immediately after the incubation, the plate was retrieved and scanned using the “Tumorsphere 1 (Green) + 2 (Blue) + Mask (BF)” application in the software. The exposure times for the green, blue, and BF channels were 30,000, 30,000, and 9000 µs, respectively ( Fig. 1b ). For the analysis, the MCTS were first identified in the bright-field images, and then the fluorescent intensities of caspase 3/7 were measured and exported directly into Excel. Next, the exported average caspase 3/7 intensities were used to determine the apoptosis effect of the drug compounds on the U87MG MCTS. The average caspase 3/7 intensities for all drugs were plotted for comparison to the control.

MCTS Invasion Assay

The effect of the drug compounds on the level of MCTS invasion was investigated by measuring the changes in MCTS invasion area in a basement membrane-like matrix (BMM; e.g., Matrigel) gel over 3 d after the drug induction.⁹ The U87MG cells were seeded at 500 cells/well in a 384-well ULA round-bottom microplate and allowed to form spheroids for 4 d. On day 4, the stock drug compound solutions were diluted directly in soluble BMM (on ice) to produce the 2× working concentrations. Next, half of the medium (40 µL) was removed from each well, and 40 µL of the drug compound/BMM titrations (final 5, 2.5, 1.25, 0.625, and 0.3125 µM) and a 0.02% DMSO control (final 0.01%) were gently added in triplicate wells (Suppl. Fig. S1b). Finally, the BMM gel was solidified at 37 °C for 30 min. The MCTS in BMM gel were incubated with the drug compounds for 3 d (~68 h), scanned, and analyzed using the Celigo Image Cytometer on days 0, 1, 2, and 3 posttreatment. On each day of image acquisition, the MCTS invasion area was measured for the control and treated samples ( Fig. 1c ). The measured MCTS invasion area results were plotted with respect to time as well as drug compound concentrations compared with the control.

Drug Compound Effect Scoring Method for MCTS

A novel multiparametric scoring method was developed and used to determine and rank the anticancer effect effects of each drug compound screened in this work. First, the averages and standard deviations of the control samples in growth, invasion, viability, and apoptosis assays were calculated. Next, the standard deviations of the control samples were multiplied by 3, 6, and 9 to provide thresholds for ranking the effects of each drug. If the measured results of MCTS diameter, invasion area, $\frac{A V E c a l c e i n A M F L I n t e n s i t y}{A V E P I F L I n t e n s i t y}$ ratio, and average caspase 3/7 fluorescent intensity were within 3 times the standard deviation of the control sample, the drug was assigned a score of 0. Similarly, if the results fell between 3 and 6 times the standard deviation, a score of 1 was assigned. If the results fell between 6 and 9 times the standard deviation, a score of 2 was assigned. Finally, if the results were greater than 9 times the standard deviation of the control sample, a score of 3 was assigned. After assigning the scores of each assay, the scores were summed to determine the overall drug effects on U87MG MCTS. The drug effects were directly correlated with the total scores, where higher scores indicated combined greater drug effects and a score of 0 represented no effect.

Results and Discussion

3D Multicellular Tumor Spheroid Growth Inhibition Results

Fourteen drug compounds were screened to determine their growth inhibitory effects on the U87MG MCTS. Figure 2a contained example bright-field images acquired over the duration of the assay for the control and compound 8 treatment. The increase in MCTS size for the control and growth inhibition of compound 8 can be clearly observed over the 9-day assay. The MCTS diameters for each drug treatment were compared with the control MCTS diameter of 603 ± 7 µm, and each drug treatment at 5 µM was plotted with respect to time and shown in Figure 2b . The results demonstrated that compounds 1, 6, 9, and 10 had no inhibitory effects on growth and produced MCTS of approximately 600 µm in size. In contrast, compounds 3, 4, 5, 12, 13, and 14 had immediate growth inhibitory effects on the MCTS, thus limiting the size to approximately 350 µm. Interestingly, compounds 2, 7, and 8 displayed a delayed inhibitory effect, in which the MCTS grew until 69 h posttreatment, and were followed by a reduction in MCTS diameter. Similarly, MCTS treated with compound 11 also displayed a delayed inhibitory effect, in which the MCTS appeared to grow until 114 h posttreatment, and then the diameters reduced. Time-course and endpoint plots for all the drug concentrations are displayed in Supplementary Figures S2 and S3, respectively. The endpoint plot for 5 µM treatment clearly demonstrated the growth inhibition of the tested compounds in comparison with the control ( Fig. 3a ).

Figure 2.

Time-course bright-field images of U87MG multicellular tumor spheroid (MCTS) growth inhibition for the control and compound 8 treatment. (a) The images show increasing size and growth inhibition for the control and compound 8 treatment over 210 h, respectively. (b) A graph of changes in spheroid diameter with respect to time showing growth inhibition induced by compounds 2, 3, 4, 5, 7, 8, 11, 12, 13, and 14. (c) The growth inhibition dose-dependent bright-field images of U87MG MCTS treated with compound 8, in which the images show decreasing size as the drug concentration increases. (d) A graph of changes in spheroid diameter with respect to drug concentrations showing dose-dependent growth inhibitory effects.

Figure 3.

Endpoint graphs of U87MG multicellular tumor spheroid (MCTS) treated with the 14 compounds at 5 µM for (a) growth inhibition, (b) invasive inhibition, (c) viability, and (d) apoptosis. The solid black line represents the average of the control sample. The red, blue, and green dotted lines in the graphs represent 3, 6, and 9 times the standard deviations of the control sample, allowing the assignment of scores 0 to 3 for each compound.

Bright-field images of the dose-dependent growth inhibition of compound 8 are displayed in Figure 2c . Interestingly, the morphology of the MCTS changed from a tight and dark feature to having a loose aggregate morphology at 0.5 µM. Routine MCTS morphologies were proposed by Vinci et al. in a modified model of the Ivascu classification of MCTS.^1,27 The MCTS appeared to have a layer of lighter-intensity cells surrounding the darker MCTS core, which could be due to the treatment inducing shedding of the outer layer of cells from the spheroid. Further investigation into the time-course bright-field images of compound 8 at different concentrations showed that the shedding or the formation of the layer of cells surrounding the spheroid began to occur at day 3 only for 0.5 µM (Suppl. Fig. S4). The growth inhibition dose-response plot is displayed in Figure 2d . The results demonstrated that compounds 3, 4, 5, 7, 12, and 13 induced consistent growth inhibition for all the tested concentrations (10, 5, 2.5, 1, 0.5, 0.1, and 0.05 µM). In contrast, compounds 6 and 9 had no inhibitory effects for all the tested concentrations. Compounds 1, 2, 8, 10, 11, and 14 displayed varying degrees of MCTS growth inhibitory properties. Indeed, compounds 1 and 10 required a much higher concentration to induce growth inhibition than compounds 2, 11, and 14. Furthermore, compound 8 produced a clear dose-response curve within the range of tested concentrations.

The ability to capture and analyze changes in MCTS diameters over time is highly important for characterizing the effects of potential anticancer drugs. Kinetic measurements can provide insights into the rate of drug inhibitory effects. In addition, the high-throughput method can screen numerous test conditions simultaneously and easily generate dose-response results at different time points.

3D MCTS Invasion Results

The 14 drug compounds were screened to determine their invasion inhibitory effects on the U87MG MCTS. In Figure 4a , bright-field images of control and MCTS treated with compound 8 in BMM gel are displayed with and without a pseudo green fill color. The pseudo green fill color in the Celigo software improves the visualization of the invasion area by filling the measured area with a transparent green color. The control MCTS invasion area was 757,984 ± 57,701 µm², and each drug treatment at 5 µM was plotted with respect to time shown in Figure 4b . The results demonstrated that compounds 1, 6, 9, 10, and 12 have limited invasion inhibitory properties and produced an MCTS invasion area of approximately 600,000 µm². In contrast, compounds 2, 3, 4, 5, 7, 8, 11, 13, and 14 showed immediate invasion inhibitory properties and limited the MCTS invasion area to approximately 180,000 µm². Time-course and endpoint plots for all the drug concentrations are reported in Supplementary Figures S5 and S6, respectively. Furthermore, the endpoint invasion data shown in Figure 3b clearly demonstrated the inhibition of the invasion area of the tested compounds at 5 µM in comparison with the control.

Figure 4.

Time-course bright-field images of U87MG multicellular tumor spheroid (MCTS) invasion assay for the control and compound 8 treatment. (a) The images show increasing invasion area and invasive inhibition for the control and compound 8 treatment over 68 h, respectively. (b) A graph of changes in invasion area with respect to time showing invasive inhibition induced by compounds 2, 3, 4, 5, 7, 8, 11, 13, and 14. (c) The invasive inhibition dose-dependent bright-field images of U87MG MCTS treated with compound 5, in which the images show decreasing invasion area as the drug concentration increase. (d) A graph of changes in invasion area with respect to drug concentrations showing dose-dependent invasion inhibitory effects.

Bright-field images of compound 5 dose-dependent inhibition of MCTS invasion area are displayed in Figure 4c . The previously described pseudo green fill visualization tool showed the dose-dependent invasion area inhibition induced by compound 5. The invasion dose-response plot is displayed in Figure 4d . The results demonstrated that compounds 2, 3, 7, 8, 11, 13, and 14 produced a consistent reduction in invasion area for all the tested concentrations (5, 2.5, 1.25, 0.625, and 0.3125 µM), whereas compounds 6 and 12 displayed no invasion inhibitory properties. Compounds 1, 4, 5, 9, and 10 displayed varying degrees of invasive inhibition, where compounds 1, 9, and 10 required a much higher concentration compared with compounds 4 and 5 to induce invasion area inhibition.

Although growth retardation and reduction of MCTS size would be an obvious and clear goal within this screen, a further important activity of any potential drug candidate is its ability to prevent tumor cell invasion into healthy tissue. In this experiment, it was clearly demonstrated that a number of the drug compounds have dose- and time-dependent invasion inhibitory properties. Understanding the ability of drug compounds to not only inhibit cell growth but also tumor cell invasion into healthy tissue can allow a more appropriate selection of potential anticancer drug candidates.

MCTS Viability and Apoptosis

The drug-treated MCTS from the first duplicate plate were stained with calcein AM and PI on day 9 (~210 h posttreatment). The calculated $\frac{A V E c a l c e i n A M F L I n t e n s i t y}{A V E P I F L I n t e n s i t y}$ ratio results ( Fig. 3c ) demonstrated that compounds 5, 6, 9, 10, and 12 induced a low level of cytotoxicity in U87MG MCTS, whereas drugs 1, 2, 3, 4, 7, 8, 11, 13, and 14 induced a higher level of cytotoxicity. The control $\frac{A V E c a l c e i n A M F L I n t e n s i t y}{A V E P I F L I n t e n s i t y}$ ratio was approximately 1.65 ± 0.11 relative units.

The drug-treated MCTS from the second duplicate plate were also stained with caspase 3/7 on day 9 (~210 h posttreatment). The measured average caspase 3/7 fluorescent intensity results ( Fig. 3d ) demonstrated drugs 1 and 3 induced a high level of apoptosis in U87MG MCTS, whereas drugs 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, and 14 did not induce or induced a low level of apoptosis within the MCTS. The control average caspase 3/7 fluorescent intensity was approximately 34.0 ± 1.3 RU.

Drug Compound Multiparametric Ranking Score Results

The effects of the screened drug compounds were ranked by the sum of assigned values depending on the calculated standard deviations. Using the calculated 3, 6, and 9× standard deviation ( Fig. 3 ), each drug compound was scored, ranked, and grouped by the level of its effects on the U87MG MCTS ( Fig. 5a ). Scores greater than or equal to 10 indicated a high effect (drugs 3, 4, 7, 8, 11, 13, and 14), scores ranging from 6 to 8 indicated a medium effect (drugs 5, 2, and 1), scores ranging from 1 to 5 indicated a low effect (drugs 12, 9, and 10), and, finally, a score of 0 indicated no effect (drug 6 and control). A table of drug-ranking results is demonstrated in Figure 5b , in which drugs 3 (romidepsin) and 4 (panobinostat) showed the highest cumulative ranking score in the U87MG MCTS model, whereas interestingly, drug 6 (cisplatin form B) produced a low ranking score and was deemed to have no effect. In addition, the calcein AM/PI ratios were plotted with respect to the spheroid diameters ( Fig. 5c ), which showed drug compounds that were highly cytotoxic (lower left), cytostatic (upper left), and had low to no cytotoxic effect (right). This graph allowed a simple overview of what effects the drug compounds have induced.

Figure 5.

(a) Drug compound effect score sheet showing the assigned scores for each assay and the total to generate the final score to rank the compounds. (b) The final scores of the compounds are ranked in the bar graph from high effect to no effect. (c) A graph of calcein AM/propidium iodide ratio with respect to spheroid diameter, which showed the cytotoxic and cytostatic effects of the drug compounds.

The bright-field and fluorescent images of each ranked compound group are displayed in Figure 6 . For the highest ranked compound group ( Fig. 6a ), the inhibition of growth and invasion area is demonstrated in the bright-field images. In the viability assay, the MCTS were mostly PI positive and displayed little or no calcein AM fluorescence. In the apoptosis assay, very few MCTS displayed caspase 3/7 fluorescence. For the medium-ranked compound group ( Fig. 6b ), both the MCTS growth and invasion were not fully inhibited. In addition, higher calcein AM and lower PI fluorescent intensities were measured. With the exception of compound 1, little or no caspase 3/7 fluorescence was detected in this group. For the lower-ranked compound group ( Fig. 6c ), both the MCTS growth and invasion were not inhibited. In addition, the MCTS were mostly calcein AM positive and displayed what appeared to be a small necrotic core, as indicated with PI staining. Finally, only a small amount of caspase 3/7 fluorescence was measured in the MCTS in this group. Only compound 6 (cisplatin form A) displayed no effect on the spheroids ( Fig. 6d ), and the results were highly comparable with the control in growth and invasion inhibition, viability, and apoptosis assays. It is important to note that previous literatures have shown that sunitinib malate was anti-invasive to human glioblastoma.¹¹ In addition, paclitaxel and 17-AAG drug compounds can induce apoptosis, cytotoxicity, and growth inhibition in glioblastoma.^2,12,13 These previously described results were comparable with the screening results in this work.

Figure 6.

Bright-field and fluorescent images of multicellular tumor spheroids grouped by high, medium, low, and no effect. (a) High-effect compounds are 3, 4, 7, 8, 11, 13, and 14. (b) Medium-effect compounds are 5, 2, and 1. (c) Low-effect compounds are 12, 9, and 10. (d) Finally, only compound 6 showed no effect. The compound numbers are shown in the order of final score values.

The multiparametric readouts described in this report can be used not only to triage the potential drug candidates but also to delineate particular drug functions and pathways involved in their activities. In Figure 6 (or Suppl. Fig. S7), the bright-field and fluorescent images of four potential candidate drugs from the high- (compound 3 and 4), medium- (compound 5), and low- (compound 12) ranked groups are displayed and compared with the control. All four drug candidates, compounds 3, 4, 5, and 12, inhibited MCTS growth, with the reduction in diameter being 300, 325, 340, and 360 µm, respectively, compared with the 600 µm in control MCTS. Earlier, we discussed the importance of considering drug candidates that have both growth and invasion inhibitory properties, which is shown in Supplementary Figure S7, where one can observe that compounds 3, 4, and 5 have reduced the invasion area, unlike compound 12. Therefore, by using this additional information, one may consider eliminating compound 12 from further investigation. However, it may be also plausible that compounds 3, 4, and 5 did not have any invasion inhibitory properties, and the effects observed in the invasion assay were due to growth retardation alone. Interestingly, although compound 12 inhibited MCTS growth, it appeared that the pathway involved promoted a cytostatic rather than a cytotoxic effect on the MCTS, as the cells within the MCTS were highly stained with calcein AM, indicating high viability. Further delineation of pathways or mechanisms can be elucidated when we compare MCTS treated with compounds 3 and 4. Both compounds displayed growth and invasion inhibitory properties, and both MCTS were highly stained for PI, indicating low viability, thus promoting a cytotoxic rather than cytostatic effect. Furthermore, one can delineate this cytotoxic effect further as the MCTS treated with compound 3 was positive for caspase 3/7. This suggests that compound 3 was inhibiting MCTS growth through a pathway that induced programmed cell death, as compared with compound 4, which appeared to induce growth retardation using a necrotic pathway. In addition, the invasion inhibition mechanisms can also be due to cytotoxic or cytostatic effects.

These data provide valuable information that can be used to decide what drugs should continue through the drug discovery pathway. In addition, the information may provide evidence that can be used to derive potential clinical treatments of cancers associated with this particular MCTS model. For example, compounds 3 and 4 may appear as the ideal drug candidates initially; however, with the increasing use of combination therapies and immuno-oncology approaches, one may consider compounds 5 and even 12 as potential candidates to continue further down the drug discovery pathway.

Footnotes

Acknowledgements

The authors would like to thank Dr. Steven Titus at the High Content Imaging and Discovery at the National Center for Advancing Translational Sciences for kindly donating 14 cancer drug compounds for this work.

Supplementary 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 of the authors declare competing financial interests, and the work performed in this article is for reporting on a novel high-throughput screening and scoring method for 3D MCTS on the image cytometry technology of Nexcelom Bioscience, LLC. The proposed drug effectiveness scoring method can improve the efficiency and speed of 3D model-based HT screening assays for identifying more highly qualified cancer drug candidates.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Vinci

Gowan

Boxall

. Advances in Establishment and Analysis of Three-Dimensional Tumor Spheroid-Based Functional Assays for Target Validation and Drug Evaluation. BMC Biol. 2012, 10, 1–20.

Kessel

Cribbes

Déry

. High-Throughput 3D Tumor Spheroid Screening Method for Cancer Drug Discovery Using Celigo Image Cytometry. J. Lab. Autom. In press.

Hirschhaeuser

Menne

Dittfeld

. Multicellular Tumor Spheroids: An Underestimated Tool Is Catching up Again. J. Biotechnol. 2010, 148, 3–15.

W. Y.

Yeap

S. K.

C. L.

. Development of Multicellular Tumor Spheroid (MCTS) Culture from Breast Cancer Cell and a High Throughput Screening Method Using the MTT Assay. PLOS One 2012, 7, e44640.

Kelm

J. M.

Timmins

N. E.

Brown

C. J.

. Method for Generation of Homogeneous Multicellular Tumor Spheroids Applicable to A Wide Variety of Cell Types. Biotechnol. Bioengin. 2003, 83, 173–180.

LaBarbera

D. V.

Reid

B. G.

Yoo

B. H.

The Multicellular Tumor Spheroid Model for High-Throughput Cancer Drug Discovery. Exp. Opin. Drug Discov. 2012, 7, 819–830.

Raghavan

Mehta

Horst

E. N.

. Comparative Analysis of Tumor Spheroid Generation Techniques for Differential In Vitro Drug Toxicity. Oncotarget 2016, 7, 16948–16961.

Zhang

Wang

. Development of an In Vitro Multicellular Tumor Spheroid Model Using Microencapsulation and Its Application in Anticancer Drug Screening and Testing. Biotechnol. Prog. 2005, 21, 1289–1296.

Vinci

Box

Eccles

S. A.

Three-Dimensional (3D) Tumor Spheroid Invasion Assay. J. Vis. Exp. 2015, 99, e52686.

10.

Adams

V. R.

Leggas

Sunitinib Malate for the Treatment of Metastatic Renal Cell Carcinoma and Gastrointestinal Stromal Tumors. Clin. Ther. 2007, 29, 1338–1353.

11.

Boüard

S. D.

Herlin

Christensen

J. G.

. Antiangiogenic and Anti-Invasive Effects of Sunitinib on Experimental Human Glioblastoma. Neuoro-Oncol. 2007, 9, 412–423.

12.

Karmakar

Banik

N. L.

Ray

S. K.

Combination of All-Trans Retinoic Acid and Paclitaxel-Induced Differentiation and Apoptosis in Human Glioblastoma U87MG Xenografts in Nude Mice. Cancer 2007, 112, 596–607.

13.

Merighi

Benini

Mirandola

. Hypoxia Inhibits Paclitaxel-Induced Apoptosis through Adenosine-Mediated Phosphorylation of Bad in Glioblastoma Cells. Mol. Pharmacol. 2007, 72, 162–172.

14.

Grant

Rahman

Piekarz

. Romidepsin: A New Therapy for Cutaneous T-Cell Lymphoma and a Potential Therapy for Solid Tumors. Exp. Rev. Anticancer Ther. 2010, 10, 997–1008.

15.

Piekarz

R. L.

Frye

Prince

H. M.

. Phase 2 Trial of Romidepsin in Patients with Peripheral T-Cell Lymphoma. Blood 2011, 117, 5827–5834.

16.

Dickinson

Ritchie

DeAngelo

D. J.

. Preliminary Evidence of Disease Response to the Pan Deacetylase Inhibitor Panobinostat (LBH589) in Refractory Hodgkin Lymphoma. Br. J. Haematol. 2009, 147, 97–101.

17.

Younes

Sureda

Ben-Yehuda

. Panobinostat in Patients with Relapsed/Refractory Hodgkin’s Lymphoma after Autologous Stem-Cell Transplantation: Results of a Phase II Study. J. Clin. Oncol. 2012, 30, 2197–2203.

18.

Mott

B. T.

Eastman

R. T.

Guha

. High-Throughput Matrix Screening Identifies Synergistic and Antagonistic Antimalarial Drug Combinations. Sci. Rep. 2015, 5, 1–14.

19.

Barry

M. A.

Behnke

C. A.

Eastman

Activation of Programmed Cell Death (Apoptosis) by Cisplatin, Other Anticancer Drugs, Toxins and Hyperthermia. Biochem. Pharmacol. 1990, 40, 2353–2362.

20.

Chauhan

Catley

. A Novel Orally Active Proteasome Inhibitor Induces Apoptosis in Multiple Myeloma Cells with Mechanisms Distinct from Bortezomib. Cancer Cell 2005, 8, 407–419.

21.

Kuhn

D. J.

Chen

Voorhees

P. M.

. Potent Activity of Carfilzomib, a Novel, Irreversible Inhibitor of the Ubiquitin-Proteasome Pathway, against Preclinical Models of Multiple Myeloma. Blood 2007, 110, 3281–3290.

22.

Wilhelm

Carter

Lynch

. Discovery and Development of Sorafenib: A Multikinase Inhibitor for Treating Cancer. Nat. Rev. Drug Discov. 2006, 5, 835–844.

23.

King

A. J.

Arnone

M. R.

Bleam

M. R.

. Dabrafenib; Preclinical Characterization, Increased Efficacy when Combined with Trametinib, while BRAF/MEK Tool Combination Reduced Skin Lesions. PLOS One 2013, 8, e67583.

24.

Liu

Greger

Shi

. Novel Mechanism of Lapatinib Resistance in HER2-Positive Breast Tumor Cells: Activation of AXL. Cancer Res. 2009, 69, 6871–6878.

25.

Xue

L.-Y.

Chiu

S.-M.

Oleinick

N. L.

Staurosporine-Induced Death of MCF-7 Human Breast Cancer Cells: A Distinction between Caspase-3-Dependent Steps of Apoptosis and the Critical Lethal Lesions. Exp. Cell Res. 2003, 283, 135–145.

26.

Chan

L. L.-Y.

Smith

Kumph

K. A.

. A High-Throughput AO/PI-Based Cell Concentration and Viability Detection Method Using the Celigo Image Cytometry. Cytotechnol. 2016, 68, 2015–2025.

27.

Ivascu

Kubbies

Diversity of Cell-Mediated Adhesions in Breast Cancer Spheroids. Int. J. Oncol. 2007, 31, 1403–1413.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

2.54 MB