Background

The extracellular matrix (ECM) is the non-cellular compartment of all tissues; it is fundamental for cellular processes, providing biomechanical and biochemical cues, which can be sensed by cells activating downstream signaling pathways that can affect cell behavior (Vennin et al., 2017; Cox, 2021; Romani et al., 2021). The ECM is a highly regulated scaffold whose function is dependent on its precise molecular composition and assembly of fibrillar tissue. The highly regulated ECM is an adaptive network of fibers that respond to external forces and tensional strains, whilst also providing cells with biomechanical feedback regulating homeostasis and normal tissue function (Cox, 2021). However, in cancer, the co-option of this tightly organized infrastructure, including ECM deposition and remodeling, can trigger oncogenic signaling, leading to increased cell motility, proliferation, survival, and metastasis (Nicolle et al., 2017; Neesse et al., 2019;Cox, 2021).

Although the cancer stroma was initially thought to predominantly stimulate tumor progression, recently it was shown to also restrain cancer cells (Rhim et al., 2014; Özdemir et al., 2014). Therefore, a fine-tuned approach to target the pro-tumorigenic functions of the stroma while maintaining its anti-tumorigenic roles may provide potential for anti-cancer therapies. Cancer associated fibroblasts (CAFs) are a prominent stromal cell type, which can be activated and educated by cancer cells (Pereira et al., 2019; Sahai et al., 2020). Their ability to synthesize and remodel ECM is a key driver of tumor progression. As such, there is a clear need to study the influence of the ECM on cancer cell behavior and assess the potential of ECM-targeted therapies to alter cellular locomotion and chemosensitivity.

Initial methodologies and protocols to generate cell-derived matrices (CDMs) were designed to better understand the complexity of cell-matrix adhesion and cell surface structures in a 2.5D to 3D context (Cukierman et al., 2001; Cukierman, 2002; Erami et al., 2016; Franco-Barraza et al., 2016). To decipher the multifaceted influence of the ECM on cancer progression and therapeutic response, CDMs can be used to characterize the effect that genetic or pharmacological manipulation have on ECM ultrastructure. Here, we provide a step-by-step protocol for the generation of CDMs and subsequent quantification of matrix abundance, architecture, and organization (Figure 1). This protocol was utilized in our recent publication to assess fine-tuned manipulation of the matrix following focal-adhesion kinase inhibition, to reduce cancer cell locomotion and improve the efficiency of chemotherapy (Murphy et al., 2021). The protocols presented below contain steps that can be modified in a cell type–specific manner, for normal fibroblasts as well as CAFs. We also highlight the utility of CDMs to assess the potential of anti-fibrotic agents to alter ECM production and remodeling, which has recently shown promise to reduce metastasis and improve chemotherapy performance in a range of cancer types (Miller et al., 2015; Rath et al., 2018; Vennin et al., 2017, 2019; Boyle et al., 2020; Wu et al., 2020; Murphy et al., 2021). Furthermore, we provide protocols for assessing cancer cell behavior when seeded on CDMs, which can provide insight into cancer cell proliferation, survival, and single vs. collective cell migration in an ECM-rich environment (Figure 2). Here, CDMs can provide a rapid and pliable readout of stromal and cancer response to genetic and pharmacological intervention in vitro, which can be used to inform subsequent in vivo studies.

Figure 1. Schematic representation of cell-derived matrix (CDM) production and cancer cell seeding on top of CDMs following fibroblast decellularization. Here, fibroblasts are seeded and allowed to grow to confluence (a.) prior to supplementation with ascorbic acid to enhance matrix production (b.). Following matrix production, fibroblasts are decellularized from the matrix (c.) and cancer cells may be seeded on top (d.). As an example timeline (e.), telomerase-immortalized fibroblasts (TIFs) are provided with ascorbic acid 24 h after cell seeding to allow matrix production for six days, prior to decellularization on day 7, followed by matrix analysis or cancer cell seeding and tracking [here shown as the example of pancreatic cancer cells isolated from the KPC ( Kras ^G12D/+ ; p53 ^R172H/+ ; Pdx-1Cre ) mouse model of pancreatic cancer (Morton et al., 2010; Murphy et al., 2021)]. Parts a-d are adapted from Murphy et al. (2021), CC BY 4.0 ( https://creativecommons.org/licenses/by/4.0/ ).

Figure 2. Potential application of cell-derived matrices for analysis of matrix ultrastructure. Examples with intact fibroblasts, including picrosirius red staining (a.) imaged using bright field microscopy (top left), binary overlay (bottom left), and by polarized light (birefringence, top right) to assess collagen I and III abundance and orientation (bottom right). Second-harmonic generation (SHG) (b.) imaging (top) and analysis of fiber orientation (bottom). Applications of CDMs with fibroblast removed and cancer cells seeded on top to assess single cancer cell migration (left) and collective migration (right).

Materials and Reagents

For CDMs

1. 12-well plate

2. Cultured fibroblasts [e.g., telomerase-immortalized fibroblasts (TIFs) or CAFs]

3. Trypsin/EDTA solution (Life Technologies, catalog number: 15400)

4. Fibroblast growth medium [specific to fibroblast line; e.g., for TIFs: DMEM (Thermo Fisher Scientific, catalog number: 11995065) supplemented with 10% FBS]

5. Phosphate-buffered saline (PBS) (Life Technologies, catalog number: 14190)

6. Gelatin (Sigma-Aldrich, catalog number: G1393), store at 4 °C

7. 10% neutral buffered formalin (Australian Biostain P/L, catalog number: ANBFC.10L)

8. Glycine (Sigma-Aldrich, catalog number: G7126)

9. (+)-Sodium L-ascorbate (Sigma-Aldrich, catalog number: A7631), make fresh

10. DNase I (Roche, catalog number: 05025), aliquot and store at -20 °C (do not freeze-thaw)

11. Fungizone (Amphotericin B, Life Technologies, catalog number: 15290)

12. Penicillin/streptomycin (pen/strep) (Thermo Fisher Scientific, catalog number: 15070-063)

13. 70% ethanol

14. 0.2% sterile gelatin (see Recipes)

15. 1 M sterile glycine (see Recipes)

16. Ascorbic acid (1,000×), prepare fresh (see Recipes)

17. Extraction buffer (see Recipes)

    1. Triton X-100 (Sigma-Aldrich, catalog number: 9284)

    2. Ammonium hydroxide

    3. Sodium deoxycholate (Sigma-Aldrich, catalog number: D6750)

18. Supplemented PBS (see Recipes)

    1. Calcium chloride (Chem Supply, catalog number: CA033)

    2. Magnesium sulfate (Chem Supply, catalog number: MA048)

19. Supplemented PBS with antibiotics and antifungals (see Recipes)

    
    

For picrosirius red staining

20. 0.1% picrosirius red stain (Abcam, catalog number: ab150681)

21. Phosphomolybdic acid hydrate (Sigma-Aldrich, catalog number: 221856)

22. Glacial acetic acid (Chem Supply, catalog number: AA009-2.5L-P)

23. Reverse Osmosis (RO) water

24. 0.02% phosphomolybdic acid (see Recipes)

25. 10% phosphomolybdic acid stock (see Recipes)

26. Acidified water (see Recipes)

Equipment

1. 12-well glass bottom dishes (Celvis, catalog number: D35-20/4.5N)

2. Sterile glass bottles (for storage of gelatin, extraction buffer, and glycine)

3. Picrosirius red staining kit (Australian Biostain, APSRA, 500 mL)

Software

1. TWOMBLI (Wershof et al., 2021): https://github.com/wershofe/TWOMBLI

2. FibrilTool (Boudaoud et al., 2014): https://www.quantitative-plant.org/software/fibriltool

3. Automated ECM fiber orientation and alignment (Mayorca-Guiliani et al., 2017): https://github.com/TCox-Lab/Collagen_Orientation

4. Automated picrosirius red and collagen birefringence analysis using polarized light (Vennin et al., 2019): https://github.com/TCox-Lab/PicRed_Biref

5. Cell Tracker (not only) for Dummies (Piccinini et al., 2016): http://celltracker.website/about-celltracker.htmL

Procedure