Inkjet-printed morphogenesis of tumor-stroma interface using bi-cellular bioinks of collagen-poly(N-isopropyl acrylamide-co-methyl methacrylate) mixture

Abstract

Recent advances in biomaterials and 3D printing/culture methods enable various tissue-engineered tumor models. However, it is still challenging to achieve native tumor-like characteristics due to lower cell density than native tissues and prolonged culture duration for maturation. Here, we report a new method to create tumoroids with a mechanically active tumor-stroma interface at extremely high cell density. This method, named "inkjet-printed morphogenesis" (iPM) of the tumor-stroma interface, is based on a hypothesis that cellular contractile force can significantly remodel the cell-laden polymer matrix to form densely-packed tissue-like constructs. Thus, differential cell-derived compaction of tumor cells and cancer-associated fibroblasts (CAFs) can be used to build a mechanically active tumor-stroma interface. In this methods, two kinds of bioinks are prepared, in which tumor cells and CAFs are suspended respectively in the mixture of collagen and poly (N-isopropyl acrylamide-co-methyl methacrylate) solution. These two cellular inks are inkjet-printed in multi-line or multi-layer patterns. As a result of cell-derived compaction, the resulting structure forms tumoroids with mechanically active tumor-stroma interface at extremely high cell density. We further test our working hypothesis that the morphogenesis can be controlled by manipulating the force balance between cellular contractile force and matrix stiffness. Furthermore, this new concept of "morphogenetic printing" is demonstrated to create more complex structures beyond current 3D bioprinting techniques.

Keywords: 3D printing; Cancer-associated fibroblasts; Cell-derived contraction; Interpenetrating polymer network; Tumoroids.

Fig. 1.

The hypothesis of inkjet-printed morphogenesis of tumoroids. (A) Illustration of contractile forces generated by a CAF and a cancer cell on the polymer matrix; (B) Schematic of inkjet printing of cell-laden interpenetrating-polymer inks along with the hypothesized mechanism of tissue compaction; (C) Scanning Electron Microscopy (SEM) images showing the microstructures and the elastic modulus showing the mechanical stiffness of the interpenetrating-polymer inks (IPIs) (the ratio indicates the P(NIPAM-co-MMA) to collagen ratio) (scale bar: 1 μm); (D) contraction assay and quantification of contraction index of human pancreatic cancer cells (Panc10.05) and pancreatic cancer-associated fibroblasts (CAF19) (scale bar: 500 μm). (n ≥ 3, *p < 0.05, **p < 0.01, ***p < 0.001).

Fig. 2.

Rapid creation of tumoroids by inkjet-printed morphogenesis. (A) 0–24 hr time-lapse images of active shrinking (red channel: Panc10.05, green channel: CAF19); (B) quantification of volumetric strain; (C) quantification of cell density and collagen concentration; (D) bright-field and fluorescent images of the different days of culture. (Scale bar: 500 μm).

Fig. 3.

Confocal microscopy analysis and elastic modulus measurement of tumoroids. (A) 3D reconstruction of Panc10.05, CAF19, overall structure, and nuclei by confocal microscopy; (B) schematic of image stacks at multiple z-depths, and the corresponding measured cell density; (C) photo of the indentation setup; (D) fluorescent images of the tumoroid at the initial state before loading, the maximum load, and the final state after unloading; and the displacement-load curves during loading & unloading of the tumoroid tested; (E) elastic modulus of fixed tumoroids and fixed tumors. (n ≥ 3, *p < 0.05, **p < 0.01, ***p < 0.001).

Fig. 4.

Morphological and histological examination of the tumoroids. (A) Representative images of stained tumoroid with H&E, Ki67, vimentin, and E-cadherin & DAPI (corresponding zoom-in windows highlighted with the red square are shown on the right of each image); (B) Representative image of the stained tumoroid and the mouse tumor xenograft with H&E; (C) Overlapped expression of E-cadherin, vimentin, and Ki67 (corresponding zoom-in windows highlighted with the white square is shown on the right of the image). (Scale bar: 100 μm).

Fig. 5.

Effects of the balance between cellular contractile force and matrix stiffness. (A) Schematic showing the effect of polymer stiffness, cell density, and cancer cell to CAF ratio. (B)Effect of polymer stiffness: 0–24hr time-lapse images of active shrinking at low and high polymer stiffness, and quantification of the volumetric strain at low and high polymer stiffness. (C) Effect of cell density: 0–24hr time-lapse images of active shrinking at low and high cell density, and quantification of volumetric strain at low and high cell density. (D) Effect of cancer cell to CAF ratio: 0–24hr time-lapse images of active shrinking, and the corresponding volumetric strains at 1:1, 1:2, and 2:1 cancer cell to CAF ratio (Scale bar: 500 μm).

Fig. 6.

Tumoroids from other types of cancer cells and CAFs. (A) Time-lapse images of tissue compaction of mouse pancreatic cancer cells (KPC2), mouse cancer-associated fibroblasts (mCAF), human prostate cancer cells (PC3), and human prostate cancer-associated fibroblasts (CAF5286); the corresponding contraction indices of KPC2, mCAF, PC3, and CAF5286 (n = 3, *p < 0.05, **p < 0.01, ***p < 0.001); (B) time-lapse images of active shrinking of KPC2 with mCAF, and PC3 with CAF5286; (C) quantification of volumetric strain. (Scale bar: 500 μm).

Fig. 7.

Creation of tumoroids with 3D cavity shape. (A) Schematic of layer-by-layer printing of cell-laden IPIs. (B) Time-lapse images of active shrinking of 3D cavity shape after printed at a 25 °C substrate. (Scale bar: 500 μm).

Fig. 8.

Creation of tumoroids with inner cancer cells-outer CAF. (A) Time-lapse images of active shrinking of inner cancer cells-outer CAF configuration by layer-by-layer printing at a 35 °C substrate. (B) Bright-field and fluorescent images of fusion of tumoroids on Day 0, Day 1, Day 2, and week 1. (C) 3D reconstruction of Panc10.05, CAF19, and whole tumoroid by confocal microscopy. (Scale bar: 500 μm).