(6ai) Engineered Hydrogel Biomaterials for Mimicking Tumor Microenvironments and Controlling Cancer Cell Fate

Research Interests: Biomaterials, Cancer, Metastasis, Tumor Dormancy, Angiogenesis, Vascularization, Microfluidics, Biofabrication

Teaching Interests: Cell & Tissue Engineering, Cancer Tissue Engineering, Biomaterials, Biomimetics & Biofabrication, Polymer Synthesis & Characterization, Transport Phenomena

Metastatic breast cancer is one of the most predominant causes of cancer-associated mortality in USA and throughout the world.^1,2 Current challenges toward developing effective treatment strategies for metastatic breast cancer include, amongst others, targeting of dormant chemo‑resistant tumor cell subpopulations³ and preventing crucial events in the metastatic cascade (e.g. epithelial-mesenchymal transition, invasion, extravasation).⁴ These efforts are aimed at preventing incidences of secondary tumor recurrence and improving relapse-free periods in cancer patients.

My research focuses on developing engineered biomaterial-based models of the tumor microenvironment for investigation of disease mechanisms and anti-cancer drug development processes. The tumor microenvironment plays a crucial role in the malignant and metastatic progression of the disease, the importance of which is gradually being realized over the past few years.^5,6 The complex, heterogenous and altered microenvironment of the malignant primary site significantly affect progression of the disease and treatment outcomes. Hence, investigating this complex milieu of tumor cells, supporting cell types and cytokine signaling within extracellular matrices (ECM) is important in this context.

In previous studies, I employed biosynthetic hydrogels, poly(ethylene glycol diacrylate)-fibrinogen (PEG-Fb), as a tissue mimic for three-dimensional (3D) culture of various cancer cell types.^7,8 The biochemical and mechanical characteristics of the PEG-Fb matrix were modulated to varying degrees to modulate breast cancer cell behavior. Softer PEG-Fb matrices facilitated higher spreading, higher colony formation and higher proliferation of cancer cells when compared to stiffer PEG-Fb matrices.⁸ PEG-Fb hydrogels were also fabricated into microspheres encapsulating cancer cells to realize a high-throughput system for anti-cancer drug testing applications.⁷ The PEG-Fb hydrogel model was eventually implemented in a microfluidic-chip device where breast cancer and stromal cells were cocultured long-term with tumor-mimetic endothelial cell networks and therapeutic efficacy of doxorubicin and paclitaxel was evaluated in vitro.⁹ Both drugs demonstrated cancer cell type-dependent and vascular network-dependent variations in drug efficacy, which could potentially be useful toward developing a more physiologically relevant drug-testing model for cancer treatment. Overall, we developed various biofabrication strategies to establishing 3D models of the primary breast cancer and microfluidic models to study tumor-stromal-endothelial interactions for drug efficacy testing.

In our current study, we focus specifically on recapitulation of metastatic breast tumor dormancy in secondary organotropic niches. Disseminated tumor cells (DTCs) from the primary tumor often lie dormant in secondary organs (bone marrow, brain, lung liver etc.) for extended periods of time and can revert to an actively proliferating state upon appropriate stimulation by inflammatory or proangiogenic signaling. Targeting DTCs as an approach towards reducing metastatic outbreaks is difficult owing to their quiescent behavior, poorly understood cellular characteristics, and limits of clinical detection.^10,11 Tumor dormancy is manifested in one of two ways: cellular dormancy (G0-G1 arrest) or dormant micrometastasis (angiogenic dormancy). In our studies, we find that metastatic breast cancer cells cultured within three‑dimensional (3D) hydrogel matrices can be made to remain dormant or become invasive by tuning the matrix adhesivity, crosslinking density and overall matrix compliance. Our developed hydrogel platform can be used to specifically identify and target dormant tumor subpopulations which would otherwise have been difficult to isolate and study in currently established models.

In the tumor dormancy model, MDA-MB-231 metastatic breast cancer cells were cultured within tunable PEG-based hydrogels containing the proteolytic-degradable peptide sequence, GGGPQGIWGQGK (PQ), and the integrin-ligating RGDS peptide. The RGDS concentration and co-monomer N-vinyl pyrrolidone (NVP) concentration were varied to modulate matrix adhesivity and crosslinking density respectively. Cells encapsulated within hydrogels with 1-10 mM RGDS displayed high viability and low apoptosis over 15 days, though cells in non-adhesive hydrogels (0 mM RGDS) underwent high cell death. Viable cell density and proliferation varied as a function of both NVP and RGDS incorporation. Soft (2.5 kPa, no NVP) hydrogels containing 5-10 mM RGDS displayed higher proliferation, higher cell density, higher metabolic activity, and the formation of large, invasive colonies. However, stiff (5.7-8.6 kPa, with NVP) hydrogels induced low proliferation, higher metabolic quiescence and a constant viable cell density over time irrespective of matrix adhesivity. Cells within dense matrices either remained solitary or formed small multicellular foci. Overall, matrix crosslinking density and adhesivity were found to jointly regulate four different fates for metastatic cancer cells: restricted survival, single cell dormancy, balanced proliferation/death and invasive growth. By tuning the matrix characteristics, DTCs were maintained in a dormant, state with high viability, low proliferation, low apoptosis, and reduced metabolic activity over 15 days in culture. Eventually, we aim to combine our established techniques of generating organ-mimetic vasculature within microfluidic devices^12-14 with the developed tumor dormancy model to investigate mechanisms of extravasation and dormancy within secondary organotropic niches (bone marrow, lung, brain etc.).

Overall, our research paves the way for designing improved engineered models to study cancer metastasis and associated pathological conditions in vitro. These models can be used to study cancer-specific mechanisms, cell-microenvironment crosstalk and increased chemo-resistance of cells. Eventually, these models could be used for predicting efficacy of new drug candidates and for improved translation towards in vivo and pre-clinical studies. In the future, I plan to extend these models and findings towards investigating specific vascular diseases, obesity, systemic inflammation and their association with cancer metastasis through microenvironmental crosstalk.¹⁵ Specifically, I am interested in investigating the role of obesity-induced inflammation in disrupting normal vasculogenic development and in fueling tumor growth (breast, colorectal, liver etc.) through activated cytokine signaling mechanisms.¹⁶ Ultimately, these initiatives would help establish causal relationships between rising obesity trends and cancer incidences, and thereby facilitate the development of novel therapeutic interventions.

References:

1. Gomis et. al., Molecular Oncology 2017;11:62-78.
2. Giancotti, Cell 2013;155:750-764.
3. Aguirre-Ghiso et. al., Nature Medicine 2013; 19(3): 276–277.
4. Brabletz et. al., Nature Reviews Cancer 2018; 18: 128-134.
5. Whiteside, Oncogene 2008; 27:5904-5912.
6. Place et. al., BMC Breast Cancer Research 2011; 13:227
7. Pradhan et. al., Biomaterials 2017; 115:141-154.
8. Pradhan et. al., Journal of Biomedical Materials Research Part A 2017; 105:236-252.
9. Pradhan et. al., Scientific Reports 2018; 8:3171.
10. Aguirre-Ghiso et. al., Nature Medicine 2013;19:276-277.
11. Almog, Cancer Letters 2010;294:139-146
12. Heintz et al., Advanced Healthcare Materials 2016, 5(17):2153-2160.
13. Heintz et. al., Journal of Visualized Experiments. 2017, 119.
14. Pradhan et. al., Advanced Healthcare Materials 2017, 6(24):1700681
15. Deng et. al., Annual Review of Pathology 2016; 11:421-449.
16. Picon-Ruiz et. al., CA Cancer Journal for Clinicians 2017; 67:378-397.