Exploiting Cancer Dormancy Signaling Mechanisms in Epithelial Ovarian Cancer Through Spheroid and Organoid Analysis

Abstract

Epithelial ovarian cancer (EOC) exhibits a unique mode of metastasis, involving spheroid formation in the peritoneum. Our research on EOC spheroid cell biology has provided valuable insights into the signaling plasticity associated with metastasis. We speculate that EOC cells modify their biology between tumour and spheroid states during cancer dormancy, although the specific mechanisms underlying this transition remain unknown. Here, we present novel findings from direct comparisons between cultured EOC spheroids and organoids. Our results indicated that AMP-activated protein kinase (AMPK) activity was significantly upregulated and protein kinase B (Akt) was downregulated in EOC spheroids compared to organoids, suggesting a clear differential phenotype. Through RNA sequencing analysis, we further supported these phenotypic differences and highlighted the significance of cell cycle regulation in organoids. By inhibiting the G2/M checkpoint via kinase inhibitors, we confirmed that this pathway is essential for organoids. Interestingly, our results suggest that specifically targeting aurora kinase A (AURKA) may represent a promising therapeutic strategy since our cells were equally sensitive to Alisertib treatment as both spheroids and organoids. Our findings emphasize the importance of studying cellular adaptations of EOC cells, as there may be different therapeutic targets depending on the step of EOC disease progression.

Keywords: G2/M checkpoint pathway; cancer dormancy; epithelial ovarian cancer; high-grade serous; organoids; spheroids; transcriptome.

Figure 1

Morphological and histological comparisons of HGSOC cell lines cultured using 3D model systems. (A,B) Brightfield images of established cells lines and patient ascites-derived immortalized cell lines cultured as spheroids and organoids. The structures were categorized based on appearance as compact, grape-like, or sparse spheroids, and dense or complex organoids, as indicated on the right side of each cell line-associated image. Spheroids and organoids were also collected for formalin-fixed paraffin-embedding (FFPE), sectioning and staining for H&E and IHC with the Ki67 proliferation marker. The scale bar is 200 µm or 250 µm for brightfield, and 200 µm for H&E and IHC images.

Figure 2

Growth analysis in relation to proliferation status of HGSOC spheroids and organoids. (A) Average organoid number (μm²/image) over time was calculated using organoid number and total organoid area measurements. (B) Organoid darkness was measured to observe the density of cells or debris within a structure over time. (C) Organoid eccentricity was measured to observe the difference in shape over time. (D) Representative brightfield images of iOvCa195, iOvCa256 and iOvCa398 organoids at day 21, with white arrows for organoid darkness and yellow arrows for organoid eccentricity. The scale bar is 600 µm. (E) The IHC sections were used to measure the percentage of Ki67 positive (Ki67+) nuclei, and they were binned based on pixel intensity within ImageJ for each indicated cell line (n = 2–3). (F) High magnification images of representative organoids with arrows indicating high positive (green), positive (blue), low positive (black), Ki67+ nuclei and Ki67- nuclei (red). (G) Immunoblots of PCNA were completed to confirm the proliferative index of spheroids versus organoids. (H) Relative quantification of PCNA in spheroids (SPH) and organoids (ORG). The graphs were generated using GraphPad Prism 10 demonstrating mean ± SD for (E) and mean ± SEM for (A–C,H). Statistical analysis for (A) was completed on the simple linear regression curve of each cell line over all timepoints; statistical analysis for (B,C) was completed on only the Day 21 data, both using an ordinary one-way ANOVA with Tukey’s multiple comparisons test and displayed with compact lettering for each cell line, indicated as bold letters; statistical analysis for (E,H) was a Student’s t test for each cell line (n ≥ 3, ns = not significant, * p < 0.05, ** p < 0.01).

Figure 3

Altered phosphorylation status of AMPK and Akt between culture model systems. (A,B) Relative protein expression and quantified fold change in phosphorylated AMPK at T172 (P-AMPK) to total AMPK in adherent (ADH), spheroid (SPH) and organoid (ORG) cultures. (C,D) Relative protein expression and quantified fold change in phosphorylated Akt at S473 (P-Akt) to total Akt in adherent (ADH), spheroid (SPH) and organoid (ORG) cultures. The graphs were generated using GraphPad Prism 10 demonstrating mean ± SEM and a one-way ANOVA with Tukey’s multiple comparisons test for statistical analysis of each individual cell line (n = 3, ns = not significant * p < 0.05, ** p < 0.01, **** p < 0.0001).

Figure 4

Discovery-based analysis of HGSOC spheroids and organoids based on bulk RNA-sequencing results. (A) Principal component analysis plot of all samples, where each colour indicates a cell line and each hue change differentiates spheroids (SPH) from organoids (ORG). (B) Differential gene expression between spheroids and organoids with the coloured dots (blue—elevated in organoids, red—elevated in spheroids) indicating significant genes. (C) Normalized enrichment scores (NES) of cancer hallmark gene sets in spheroids (NES = positive) and organoids (NES = negative) with the red bars showing two pathways of interest, G2/M checkpoint and E2F Targets (FDR < 0.05). (D,E) Enrichment plots for G2M checkpoint and E2F target grouped gene sets within the cancer hallmark pathways. (F) RT-qPCR validation of selected genes within the G2M checkpoint and E2F target cancer hallmark and curated sets, where any bar above the dotted line (fold change = 1) shows an increase. The graphs were generated using GraphPad Prism 10, demonstrating mean ± SEM with a Student’s t test for statistical analysis grouping all cell lines to show the difference between spheroids and organoids (n = 3, ** p < 0.01, *** p < 0.001).

Figure 5

Expression and regulation of G2M checkpoint proteins in HGSOC organoids versus spheroids. (A) Relative protein expression of phosphorylated CDK1 at Y15 (P-CDK1), total CDK1, phosphorylated cdc25C at S216 (P-cdc25C) and total cdc25C in spheroid (SPH) and organoid (ORG) cultures. (B,C) Quantified fold change in P-CDK1 and P-cdc25C to actin housekeeping protein. The graphs were generated using GraphPad Prism 10 demonstrating mean ± SEM a Student’s t test for statistical analysis (n ≥ 3, ns = not significant, * p < 0.05, ** p < 0.01, *** p < 0.001).

Figure 6

Varying sensitivities to G2M checkpoint small molecule inhibitors on HGSOC spheroid and organoid viability. (A–C) Dose response curves of a Adavosertib, Alisertib and Volasertib normalized to either DMSO control (Adavosertib and Volasertib) or the lowest drug concentration (Alisertib) based on the alamarBlue readings in spheroid (SPH) and organoid (ORG) cultures. (D) Bar graphs of area under the curve (AUC) for each of the above dose–response curves to demonstrate sensitivity between spheroids and organoids. The graphs were generated using GraphPad Prism 10, demonstrating mean ± SEM with two-way ANOVA for statistical analysis (n = 3–4, ns = not significant, * p < 0.05, ** p < 0.01, *** p < 0.001).