Adjustable extracellular matrix rigidity tumor model for studying stiffness dependent pancreatic ductal adenocarcinomas progression and tumor immunosuppression

Abstract Pancreatic ductal adenocarcinomas (PDAC) is one of the stiffest malignancies with strong solid stresses. Increasing stiffness could alter cellular behavior and trigger internal signaling pathways and is strongly associated with a poor prognosis in PDAC. So far, there has been no report on of an experimental model that can rapidly construct and stably maintain a stiffness gradient dimension in both vitro and in vivo. In this study, a gelatin methacryloyl (GelMA)‐based hydrogel was designed for in vitro and in vivo PDAC experiments. The GelMA‐based hydrogel has porous, adjustable mechanical properties and excellent in vitro and in vivo biocompatibility. The GelMA‐based in vitro 3D culture method can effectively form a gradient and stable extracellular matrix stiffness, affecting cell morphology, cytoskeleton remodeling, and malignant biological behaviors such as proliferation and metastasis. This model is suitable for in vivo studies with long‐term maintenance of matrix stiffness and no significant toxicity. High matrix stiffness can significantly promote PDAC progression and tumor immunosuppression. This novel adaptive extracellular matrix rigidity tumor model is an excellent candidate for further development as an in vitro and in vivo biomechanical study model of PDAC or other tumors with strong solid stresses.

promote both pancreatic cancer proliferation, promote its metastasis, and block the uptake of chemotherapy drugs. 3,4 Extracellular matrix (ECM) is a three-dimensional noncellular network structure that plays a vital role in maintaining the function and structure of an organization. Cell microenvironment disorders are caused by ECM synthesis, deposition, remodeling, cross-linking, and enzyme modification. Deposition, remodeling, and cross-linking of ECM composition can cause fibrosis to stiffen the stroma, thus exerting mechanical forces on the cells, known as stiffness or rigidity. 5 PDAC has a large stromal component, making it one of the most stiff malignancies with solid stresses that exceed 10 kPa. 3 While traditional views focus primarily on chemical properties, recent studies have demonstrated that physical properties contribute significantly to physiological and pathological outcomes of the ECM. The ECM has been studied for its stiffness, related to the slow elastic force on the cells. There was an increase in tumor stiffness in PDAC due to massive ECM deposition, particularly collagen and hyaluronic acid. Increasing stiffness could alter cellular behavior and trigger internal signaling pathways. 6 Fibronectin is produced when the ECM becomes stiffer, binds to extracellular collagen, fibrin, heparan sulfate proteoglycans, and integrins.
Increased stiffness of the ECM increases cell adhesion to the ECM, connects the ECM to the cytoskeleton through local adhesion proteins, and increases cytoskeletal tension through Rho/ROCK signaling. 7 Integrin aggregation promotes tumor progression by recruiting focal adhesion signaling molecules, including FAK, Src, paxillin, Rac, Rho, and Ras. 8 Additionally, stiffening the ECM increases PI3K activity and tumor invasion. 9 The increase in matrix stiffness in PDAC promotes the activation of YAP/TAZ, which, in turn, stimulates the production of the ECM protein and profibrotic mediators. [10][11][12] Additionally, ECM stiffness and oncogene-mediated changes in cell mechanical properties are essential for reprogramming normal cells into tumor precursors. 13 Gelatin methacryloyl (GelMA) has been widely used in various biomedical applications, because they possess suitable biological properties and a wide range of physical properties that can be tailored. As the hydrolysis product of collagen, the main component of ECM in most tissues, the properties of GelMA hydrogels in three dimensions (3D) closely resemble those of native ECM due to the presence of cell-attaching proteins and matrix metalloproteinases. [14][15][16] GelMA is also versatile from a processing perspective. In response to light irradiation, it forms a hydrogel with tunable mechanical properties that mimic the native ECM.
Matrix stiffness is strongly associated with poor prognosis and PDAC. 6,17 Increasing stiffness in pancreatic cancer tissue occurs slowly and gradually. So far, there has been no report on developing an experimental model that can rapidly construct and stably maintain a stiffness gradient dimension both in vitro and in vivo. In

| GelMA hydrogels preparation
To generate hydrogels, 60% DS GelMA (GM60; EFL-GM-60, Yongqinquan Intelligent Equipment Co., Ltd., Suzhou, China) was resuspended at 5% and 20% (w/v) in PBS and incubated in a 55 C water bath until it was completely dissolved. 90% DS GelMA (GM90; EFL-GM-90, Yongqinquan Intelligent Equipment Co., Ltd., Suzhou, China) was resuspended at 10% (w/v) in PBS and incubated in a 55 C water bath until it was completely dissolved. The GelMA solution was then combined with the photoinitiator lithium phenyl-2, 4, 6-trimethylbenzoyl phosphinate (LAP) at a final concentration of 0.025% w/v. Finally, the solution was heated to 37 C in a water bath, followed by sterilization using a 0.22-μm filter, and then aliquoted and stored at À20 C.

| Rheological measurement
The rheological properties of 5% GM60, 10% GM90, and 20% GM60 hydrogel were explored on a rotating rheometer (TA-DHR-2TA, Instruments). To study the cross-linked behavior of visible light (405 nm) cross-linked behavior of the hydrogel, cross-linked visible light (UV405 nm, 30 s, 25 mw/cm 2 ) was performed in 30-60 s. The storage modulus (G 0 ) was measured at a fixed angular frequency of 5 rad/s at 37 C with constant 1% strain.

| Biocompatibility and cytotoxicity test
Calcein/PI Cell Viability/Cytotoxicity Assay Kit (C2015M, Beyotime Biotechnology, Shanghai, China) was used to evaluate the biocompatibility and cytotoxicity of 3D cell culture with adjustable extracellular matrix rigidity (1, 10, and 20 Kpa) based on GelMA. Cells were collected and re-seeded in 3D culture in a 24-well plate following Protocol A as described above. After 5 days of cell culture, following the manufacturer instructions of manufacturer, cell climbing slices were co-stained with Calcein/PI for 30 min at 37 C.
Cell climbing slices were taken out and conformally sealed to the slide. Cell survival was observed using a confocal microscope (FV3000, Olympus, Japan), and 3D cell images were reconstructed.

| Cell morphology, proliferation, and migration assay
The plate-cloning assay was used for cell proliferation and migration assay. Cells were collected and reseeded in 3D culture in a 6-well plate following Protocol B as described above. The morphological change of cells was observed under an inverted microscope. The formula calculated cell volume: volume = 3/4 Â π(3.14) Â length Â width. 2 After 5 days, the hydrogel was removed from the 6-well plate.
Cells that migrated from the hydrogel and adhered to a 6-well plate for 5 days of culture were fixed and stained with crystal violet. The visible colonies were then counted using FIJI Image J software (https://imagej.net/software/fiji/).

| Western blot analysis based on a 3D cell culture system
Protein extraction in a 3D cell culture system was called tissue protein extraction. Cells were collected and reseeded in 3D culture in a 6-well plate following Protocol A as described above. After 5 days of cell culture, the media was removed and washed with PBS (2 Â 2 mL). Use a cell spatula to gently lift the hydrogel and place it in a 1.5 mL microcentrifuge tube.
The samples were then lysed in RIPA lysis buffer and fully ground by hand using an electric grinding rod. After sonication and centrifugation at 12,000g for 15 min at 4 C, the supernatants were collected.
The procedure remainder was performed as previously described. 18

| Immunofluorescence
Cells were collected and reseeded in 3D culture in a 24-well plate following Protocol A as described above. After 5 days, cells in 3D Fluorescence intensities were analyzed using FIJI ImageJ software (https://imagej.net/software/fiji/).

| Imaging procedure
Elastography was performed using the SuperSonic Aixplorer ultrasound system (France) with an SL15-4 linear transducer. When the pancreatic tumors are detected by echography, a study box is placed and a US wave is applied at different depths, compressing the tissue.
Three elastographic images obtained in the maximal diameter plane were taken of each lesion in the elastography mode of SWE. SWE values were determined by setting fixed regions of interest (ROIs) on the entire lesion and adjacent tissue. All SWE values were recorded and the mean maximal SWE values were used for assessment.

| Bulk RNA-seq assay
The bulk RNA-seq assay was performed by BGI (BGI, Beijing, China).
Total RNA was extracted from murine tumor tissues by Trizol and sequenced at MGISEQ2000. The reference genome was Mus_musculus.GRCm38. After the performance and quality of the RNA-seq was evaluated, a standard analysis pipeline was adopted to detect biologi-

| Space transcriptome
The space transcriptome (ST) data reported in this manuscript were obtained from published sources (GSE111672). 19 Seurat clustering was used Seurat R package (version 4.3.0). The carcinoma foci in the HE staining pictures were observed and analyzed by two professional pathologists according to the lists of cell gene markers (Table S2)

| Immunohistochemistry and immunofluorescence
Tumor tissues immunohistochemistry (IHC) and immunofluorescence (IF) were performed as previously described. 18 The average integrated optical density (IOD) and fluorescence intensities in five randomly selected areas for each group were calculated using FIJI ImageJ software (https://imagej.net/software/fiji/).

| Statistical analysis
Data were presented as mean ± SD. Statistical analysis was performed using GraphPad Prism 9.0 software via Student t-test and one-way ANOVA analysis. Differences were considered statistically significant at p < 0.05.

| Matrix stiffness adjustable hydrogels characterization
Van den Bulcke and co-workers introduced GelMA in 2000. 22 Due to its inherent bioactivity and physicochemical tailorability, it has gained considerable interest in tissue engineering. 14  F I G U R E 3 Legend on next page. complex mechanical environment in vivo. Cell viability and attachment are critically influenced by the average pore size of a biological scaffold. 24 A scanning electron microscope (SEM) was used to observe the 3D bulk structure of hydrogels. The hydrogels are highly porous with moderate thickening of the hole walls (Figure 2f).

| In vitro matrix stiffness adjustable 3D culture system
We applied two methods for matrix stiffness adjustable 3D cell culture in vitro (Protocol A and Protocol B; Figure 3a,b). Protocol A can control the 3D culture thickness and its contact area with the medium by controlling the size of the cell climbing slices or the material volume. Cells can be digested faster and harvested for other experiments (effective surface area enlargement), and microscope observation is much clearer using this method. This method is suitable for experiments with certain material thickness requirements, such as shooting confocal laser scanning microscopy. The advantage of Protocol B is its simplicity and provides an environment that resembles as closely as possible that of in vivo. Cells can migrate to the medium direction (nutrient-rich conditions) during culture. Therefore, if the experimental design is reasonable, this method can also be used to study cell migration ability.
Biocompatibility is an essential factor for the biomedical application of nanomaterials. GelMA is a promising material due to its superior biocompatibility and biodegradability. 25 Live/dead staining confirmed that the GelMA-based 3D culture system has excellent biocompatibility, and the cellular activity of PDAC cell lines was not significantly affected by different gelMA hydrogels of matrix stiffness (Figure 3c,d).

| The influence of matrix stiffness on cell morphology, proliferation, and migration of PDAC cells in vitro
Cellular deformation depends on the outer force minus the inner tension. 26  Through experiments similar to the principle of plate cloning, we found that the speed of cell migration is more prominent with increasing matrix stiffness. Cells migrated to culture medium are more likely to adhere and proliferate rapidly in monoclonal form (Figure 4g).
Together, these results reveal that high matrix stiffness could enhance cell proliferation and migration of PDAC cells. The results also confirmed the findings of the TMA analysis (high matrix stiffness had a high ratio of regional lymph node involvement and distant metastasis) ( Figure 1f).

| The influence of matrix stiffness in vivo
Many fabrication methods have been proposed for using microgels in cell therapy, controlled drug release, and disease modeling. 15,25,[27][28][29] Because of its excellent biocompatibility (as shown above), we believe that GelMA should also be suitable as a material for in vivo experi- Therefore, we tried to build an adjustable extracellular matrix rigidity model of PDAC hydrogels based on gelMA hydrogel. Figure 5a presents the general workflow for this modeling. The weight of animals weight was monitored, and no toxicity effect was observed ( Figure 5b). In situ tumors have a regular tumor morphology, and F I G U R E 3 In vitro matrix stiffness adjustable 3D culture system. (a and b) Schematic representation of two methods our established for preparing in vitro 3D culture system; (c and d) Live/dead staining (Calcein/ PI staining) of CFPAC-1 and KPC cells after culture in a GelMA 3D culture system of 1, 10, and 20 kPa for 5 days (Calcein staining as green and PI staining as red, Scale bar = 100 μm). Data presented in the graphs represent means ± SD. ***p < 0.001; **p < 0.01; *p < 0.05; ns p > 0.05. matrix stiffness significantly promotes pancreatic tumor growth (Figure 5c-e). By physical palpation or using imaging modalities such as magnetic resonance imaging, computerized tomography and elastography, cancer can be detected by taking advantage of the stiffness of solid tumors compared to healthy tissue. 32,33 To study the actual tumor matrix stiffness in vivo in animal models, we measure tumor elastography in live mice using the Aixplorer SWE imaging system (SuperSonic Imagine, Aux-en-Provence, France) 2 weeks after modeling (Figure 5f). The results confirm that the model can effectively establish and maintain a stable and gradient matrix stiffness environment in vivo (Figure 5g). Based on tumor size, we estimated that the matrix stiffness after 1:1 dilution of Matrigel should be between the low and medium stiffness groups (1-10 Kpa), which was also confirmed by the ultrasound elastography results. Compared to Matrigel-based models, GelMA hydrogel-based adjustable extracellular matrix rigidity orthotopic tumor transplantation murine models have the following advantages. First, it is relatively cheap, has controlled ingredients (no growth factors), is easy to store, and can be used for short-term storage at room temperature (37 C). Second, compared to temperature curing, light curing conditions are stable and adjustable, the batch effect is smaller, and the repeatability is higher. Third, it has no significant cytotoxicity, biological toxicity has a high tumor formation rate and can build and maintain a stable matrix stiffness microenvironment.

| Multiple signaling pathway pathways are involved in the regulation of stiffness-dependent PDAC progression
To explore the mechanisms of stiffness-dependent PDAC progression, we performed a bulk RNA sequencing using tumor tissues of Increasing matrix stiffness also resulted in thickening of cytoskeletal organization stress fibrils and increased actin and α-tubulin expression (actin staining as green and α-tubulin staining as red); (f) The clone-forming assay shows that changes in matrix stiffness affect cell clone-forming abilities; (g) Schematic of a 3D cell culture system based on cross-linked GelMA hydrogels with visible light (405 nm) cross-linked GelMA hydrogels. Data presented in the graphs represent means ± SD. ***p < 0.001; **p < 0.01; *p < 0.05; ns p > 0.05. that high matrix rigidity tumor microenvironment (TME) was highly associated with extracellular matrix organization, multiple signaling pathway pathways, and immune response (Figures 6a-c). The results illustrated that Wnt signaling pathway, Hippo signaling pathway, PI3KÀ Akt signaling pathway, epithelial-mesenchymal transition (EMT), TGFβ signaling. cancer-related pathways were upregulated in the high matrix rigidity group. Extracellular matrix organization, skeletal system morphogenesis. stiffness-related pathways were also upregulated. An immune response such as leukocyte migration, response to chemokine, chemokine production, cytokineÀcytokine receptor interaction, interferon alpha response, and interferon-gamma response was downregulated in the high matrix rigidity group.
To further corroborate the role of matrix rigidity in tumor immunosuppressive microenvironment, we analyzed immune infiltration in pancreatic ductal adenocarcinoma microenvironment using public ST data. 19 Consistent with the bulk RNA-seq results above, in terms of spatial location, the extracellular matrix significantly inhibits the infiltration of anti-tumor immune cells (effect T, NK, M1) into the tumor area, while the distribution of cells that promote immunosuppression (M2, CAF, monocyte) is not affected by the matrix, even significantly colocalized with tumor cells (Figure 6d,e).
Both transcriptomic and public ST data indicated that stiffness was closely related to PDAC progression and anti-tumor immunosuppression.

| Extracellular matrix rigidity promotes EMT and the tumor immunosuppression microenvironment
Increasing evidence points to the role of EMT in fibrosis, cancer progression, metastasis, and drug resistance. 34,35 TGF-β was a pleiotropic cytokine and modulated various physiological processes, including immunological reaction, cell proliferation, and EMT. TGF-β promoted tumor growth by inhibiting antitumor immunoreaction in the tumor microenvironment and promoted tumor metastasis via EMT. 36 We further assessed EMT status, TGFβ1, and PDL1 expression in the 3D F I G U R E 6 Bulk RNA-seq and space transcriptome to the mechanisms of stiffness-dependent PDAC progression. Adjustable orthotopic extracellular matrix rigidity tumor transplantation murine models of PDAC were built in immune-competent C57BL/6J mice. After 2 weeks, tumor tissues were collected for bulk RNA-seq assay.

| CONCLUSIONS
Matrix stiffness is strongly associated with a poor prognosis in PDAC.