Research ArticleENGINEERING

Human tumor microenvironment chip evaluates the consequences of platelet extravasation and combinatorial antitumor-antiplatelet therapy in ovarian cancer

See allHide authors and affiliations

Science Advances  21 Jul 2021:
Vol. 7, no. 30, eabg5283
DOI: 10.1126/sciadv.abg5283
  • Fig. 1 Microengineering of the OTME-Chip.

    (A) A conceptual infographic of the human tumor microenvironment showing that the cancer cells interact with the neighboring blood vessels, making them permeable, and recruiting platelets into their vicinity. These extravasated platelets actively bind their ligands to the tumor surface receptors and result in tumor proliferation and chemoresistance, which can potentially be arrested by antiplatelet drugs. (B) Engineering drawing of the microdevice containing two PDMS compartments separated by a thin porous membrane that reproduces the microarchitecture of the tumor-vascular interface (left). On the right, cross-sectional side view of the OTME-Chip describes tissue organization inside the chip. (C) Illustration showing the timeline and steps of OTME-Chip formation, platelet extravasation into the tumors, and following consequences. Cancer cell dynamics and molecular readouts are analyzed every 24 hours after platelet extravasation. (D) Schematic diagram of OTME-Chip that shows tumor invasion dynamics can be systematically visualized and characterized after platelet extravasation from the bottom vascular chamber into the top tumor chamber. On the right, cross-sectional side view of 3D confocal scan of OTME-Chip showing cancer cells (yellow), endothelial cells (red), and platelets (cyan) at 0 hours (left) and 72 hours (right) after platelet extravasation. Scale bars, 100 μm.

  • Fig. 2 Platelet GPVI expression is shear dependent and it binds to tumor galectin-3 in OTME-Chip.

    (A) Infographic describing platelet extravasation into the tumors inside the chip and platelet GPVI interaction with ovarian tumors through binding to galectin-3, which results in a prometastatic and chemoresistive tumor microenvironment. Therefore, GPVI inhibitors may be a therapeutic target to arrest ovarian cancer metastasis. Representative flow cytometry tracings at variable time points from cells recovered from the OTME-Chip, corresponding to (B) galectin-3 expression on the surface of ovarian cancer cells and (C) GPVI expression on the surface of extravasated platelets. (D) Surface plasmon resonance tracings (n = 3 independent experiments) show strong binding affinity between GPVI and galectin-3 proteins isolated from the platelets and cancer cells (red) against platelets exposed to anti-GPVI monoclonal antibody (mAb) (blue). GPVI-free platelet lysate shows no binding with galectin-3 (black). Binding affinity (KD) values are indicated in brackets.

  • Fig. 3 Platelet GPVI promotes ovarian metastasis through galectin-3 binding to cancer cells in OTME-Chip.

    (A) Western blot and corresponding densitometry analyses shows highly reduced galectin-3 (Gal3) protein in the KO cancer cells against wild-type (WT) controls. (B) Representative fluorescence microscopy images showing cancer cell (green) invasion (marked by arrows) into hydrogel ECM due to extravasated platelets (yellow). Scale bar, 200 μm. (C) Bar graph showing the quantification of ECM invasion in Control-Chip, OTME-Chip, and KO–OTME-Chip at 48 hours. Analysis of (D) cancer cell proliferation, (E) flow cytometry–based cell cycle phases, (F) excreted growth factors (top) and cytokines (bottom) over time, and (G) real-time polymerase chain reaction (RT-PCR) heatmaps showing expressional alteration of cell proliferation and metastasis regulatory genes in Control-Chip, OTME-Chip, and KO–OTME-Chip. *P < 0.05 and **P < 0.01; n = 3 individual experiments; error bars are means ± SEM. One-way analysis of variance (ANOVA) is done followed by Dunnett’s multiple comparisons test.

  • Fig. 4 Revacept (GPVI inhibitor) arrests ovarian metastasis in OTME-Chip.

    (A) Surface plasmon resonance tracings (n = 3 independent experiments) show strong binding affinity between GPVI and galectin-3 proteins isolated from the platelets and cancer cells (red) against platelets exposed to Revacept (blue). Binding affinity (KD) values indicated in brackets. (B) Representative fluorescence microscopy images showing cancer cell (green) invasion into hydrogel ECM due to extravasated platelets (yellow). Scale bar, 200 μm. (C) Bar graph showing the quantification of ECM invasion in Control-Chip, OTME-Chip, and Rx–OTME-Chip. Analysis of (D) cancer cell proliferation, (E) flow cytometry–based cell cycle phases, (F) excreted growth factors (top) and cytokines (bottom) over time, and (G) RT-PCR heatmaps showing expressional alteration of cell proliferation and metastasis regulatory genes in Control-Chip, OTME-Chip, and Rx–OTME-Chip. *P < 0.05 and **P < 0.01; n = 3 individual experiments; error bars are means ± SEM. One-way ANOVA is done followed by Dunnett’s multiple comparisons test.

  • Fig. 5 Revacept (GPVI inhibitor) decreases chemoresistance in OTME-Chip.

    Analysis of (A) ECM invasion, (B) cancer cell proliferation, (C) flow cytometry–based cell cycle phases, (D) excreted growth factors (top) and cytokines (bottom) over time, and (E) RT-PCR heatmaps showing expressional alteration of cell proliferation and metastasis regulatory genes in Control-Chip, OTME-Chip, Cx–OTME-Chip, and CxRx–OTME-Chip. *P < 0.05 and **P < 0.01; n = 3 individual experiments; error bars are means ± SEM. One-way ANOVA is done followed by Dunnett’s multiple comparisons test.

  • Fig. 6 RNA-seq and differential gene expression analysis reveals the efficacy of antiplatelet therapy to attenuate tumor-promoting pathways.

    (A) Schematic of cancer cell isolation, cell sorting, and mRNA isolation process used in the study. Samples were then assessed for quality and processed for RNA sequencing. (B) Figure shows Venn diagram of differentially expressed genes among all groups. RNA sequencing and differential gene expression analysis revealed enrichment of 12,452, 5566, 6909, and 3821 genes in the OTME-Chip, Cx–OTME-Chip, CxRx–OTME-Chip, and KO–OTME-Chip, respectively, with respect to Control-Chip. (C) Group-wise comparison of common genes for all combinations. OTME-Chip and CxRx–OTME-Chip had the highest number common genes (~5500 genes). (D) KEGG pathway clustering revealed differential presence of EMT regulatory pathways among groups relative to Control-Chip (P < 0.05). (E) Volcano plots showing differentially expressed genes for all groups relative to Control-Chip (black dots). Red dots signify genes regulating EMT and metastasis. (F) Heatmap showing row-scaled z scores of ~900 genes belonging to the pathways as depicted in (E). For each group, n = 3 number of samples were sequenced.

Supplementary Materials

  • Supplementary Materials

    Human tumor microenvironment chip evaluates the consequences of platelet extravasation and combinatorial antitumor-antiplatelet therapy in ovarian cancer

    Biswajit Saha, Tanmay Mathur, James J. Tronolone, Mithil Chokshi, Giriraj K. Lokhande, Amirali Selahi, Akhilesh K. Gaharwar, Vahid Afshar-Kharghan, Anil K. Sood, Gang Bao, Abhishek Jain

    Download Supplement

    This PDF file includes:

    • Figs. S1 and S2

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article