Engineered Humanized Microenvironments Expose the Awakening Tumor Niche

The life of a disseminated tumor cell remains elusive mainly due to the lack of relevant experimental models. Successfully disseminated tumor cells are rare and can take months to decades to develop a clinically detectable mass.

Like Comment

One important and promising application in tissue engineering is to create functional human tissue analogs to reduce the gap between preclinical animal studies and clinics. In doing so, essential biological mechanisms, detailed disease progression, and candidate drug efficacy and toxicity can be studied with high patient fidelity and predictive power.  In the paper published in Nature Biomedical Engineering, we report a tissue-engineered metastasis model that can recapitulate and dissect the complex and dynamic processes of dormancy and recurrence of disseminated human tumor cells in the presence of human stromal and immune cells. Our study started with the development of an inverted colloidal crystal (ICC) hydrogel scaffold inspired by the inverse opal photonic crystal structure used in nanomaterials1,2 (Fig.1A). Fully interconnected, uniform size, and ordered spherical pore arrays were leveraged to improve standardization and reproducibility of in vitro 3D cell culture (Fig.1B). Hydrogel matrices with ICC structure exhibit anatomical and biophysical similarities to lymphoid organs and have demonstrated biological significance in simulating bone marrow3 and lymph nodes4 in vitro. Subdermal implantation of the ICC hydrogel scaffolds in mouse models facilitated rapid and extensive inter-scaffold angiogenesis5, 6. This feature promoted the survival and function of pre-seeded human bone marrow stromal cells. Non-degradable synthetic hydrogel maintained chronic inflammatory response and recruited systemically circulating immune cells, such as endogenous hematopoietic stem cells and tail-vein injected human leukemic cells. The implanted microenvironments also demonstrated attraction of circulating tumor cells released from an orthotopic xenograft tumor models and subsequent metastatic growth7. Comparison of stromal cell gene expression between metastatic positive and negative implants revealed upregulated inflammatory factors in scaffolds containing metastasis, indicating that inflammation and immune cell activity are important modulators of tumor metastasis. This result was in line with rapidly accumulating compelling observation that bidirectional interaction between disseminated tumor cells and their local microenvironment is critical in directing disseminated tumor cell dormancy and reactivation8-15. However, existing humanized mouse models for metastasis are limited to study the role of human immune cells in regulating metastatic dormancy and relapse.

Figure 1. Experimental strategy to create and monitor humanized tumor microenvironments. (A) ICC hydrogel scaffold. (B) In vitro seeding of human bone marrow stromal cells (hBMSCs) to create humanized microenvironment once implanted. Figures are adapted from our paper.

My advisor. Dr. Jungwoo Lee launched his independent research group at the University of Massachusetts- Amherst in 2014. I joined the lab as the founding graduate student member. Our initial goal was to create a more complete humanized mouse model by introducing human peripheral blood mononuclear cells (hPBMCs). When we combined this system with an orthotopic human prostate tumor expressing green fluorescent protein and luciferase, we realized another critical limitation of the traditional mouse model; lethal metastases quickly arise in vital organs and greatly reduce the experimental time frame (Fig.2A). This phenomenon makes it challenging to observe disseminated tumor cells that remain dormant for extended periods of time. Within the implantable microenvironments, metastases were rare in the primary host. However, when the intact microenvironments were transplanted to tumor-free hosts, we observed gradual metastatic tumor formation via bioluminescent imaging (Fig.2B). In mice that received human immune cells, we observed a reduction in tumor growth in the scaffolds 10 weeks after transplantation (Fig.2C).

Figure 2. Tissue-engineered model for studying human tumor metastasis. (A) Schematic to create humanized tumour microenvironments and monitor long-term evolution via serial transplantation. (B) Representative bioluminescent images at different time points in primary and secondary mice and gross image of explanted scaffolds at the end of the study. (C) Quantitative analysis of the metastatic growth of human disseminated tumor cells over 10 weeks in secondary mice with and without human immune cells. Figures are adapted from our paper.

We wanted to ensure that the scaffolds that remained negative for bioluminescent signal still retained tumor cells. Flow cytometry and thin tissue slicing are non-ideal for finding potentially rare cell populations. We were inspired by work was done recently in the field of tissue clearing, and deep tissue imaging and adapted the CLARITY protocol16 to the scaffold microenvironments (Fig.3A). Via confocal microscopy, we were able to image whole scaffolds and revealed singular and colonized disseminated tumor cells. We observed disseminated tumor cells in all six of the bioluminescent negative scaffolds imaged, indicating that tumor cells reproducibly remained in the scaffold microenvironment following transplantation and could remain in a singular state. When we counted the single and colonized tumor cells, mice that received hPBMCs had a higher ratio of colonized cells to single cells. This was a puzzling result since the bioluminescent imaging showed fewer positive scaffolds in mice with hPBMCs.

We looked at vascular and immune cells in the surrounding disseminated tumor microenvironment to try to better understand how the microenvironment was changing with the introduction of hPBMCs. It became evident that single and colonized DTCs had distinct microenvironments that were conserved regardless of the presence of hPBMCs. Single DTCs remained far from vasculature with little immune cell interaction. Conversely, colonized DTCs appeared to be dependent on vascularization to grow and were infiltrated with human and native mouse immune cells. We looked further into tumor progression and noticed that as the disseminated tumor cells became a large metastasis, innate immune cells no longer infiltrated. Here, we conducted immunohistostaining of sequential sections with 10 different antibodies and compiled the images for the multiplex analysis. Overt metastatic tumors exhibit heterogenous microenvironments and quantitative analysis 8 different anatomical regions representing 5 different functional zones (Fig.3B). When taken as a whole, these results suggest that Ly6G+ neutrophils may be important in initial colonization events but could be detrimental in the long term for tumor cells (Fig.3C). Tumors that continue to grow to expel these cells and control the surrounding microenvironment. While we did not conduct intravital imaging in this study, the optical transparency of the biomaterial allows performing real-time imaging of disseminated tumor cells via surgical engraftment of a dorsal skin-fold window chamber.

Figure 3. Analysis of early and advanced stages metastatic environments substantiated the potential roles of immune cells in metastasis. (A) Gross images of explanted scaffolds before and after processing tissue clearing and confocal images of tissue cleared scaffolds displaying vasculatures and disseminated human tumor cells. (B) Multiplex immunohistostaining of an overt metastatic tumor and quantitative analysis captures 8 different regions with 5 different functional zones. (C) Ly6G+ neutrophils localized at the boundary of an actively growing overt scaffold and quantitative comparison of Ki67 overlap confirms that proliferating cells are cancer cells while neutrophils are recruited. Figures are adapted from our paper.

With the work presented in this article, we hope to draw further enthusiasm within the field of cancer biology toward unconventional methods to study complex problems such as metastasis. We are continuing to work on understanding the mechanisms that awaken dormant tumor cells and enhance our models to better represent physiological tissue such as the liver, lung, and bone marrow.  Implantable microenvironments represent an exciting new horizon to uncover the mysteries that remain about cancer and aid in the development of effective therapies. Lastly, we thank the National Cancer Institute for supporting our work (R00CA163671) and the National Science Foundation for my research fellowship (1545399).


Contributed by

Ryan Carpenter
Jungwoo Lee



1.         Kotov, N.A. et al. Inverted colloidal crystals as three-dimensional cell scaffolds. Langmuir 20, 7887 (2004).

2.         Lee, et al. Inverted colloidal crystals as three-dimensional microenvironments for cellular co-cultures. J Mater Chem, 16, 3558 (2006).

2.         Nichols, J.E. et al. In vitro analog of human bone marrow from 3D scaffolds with biomimetic inverted colloidal crystal geometry. Biomaterials 30, 1071 (2009).

3.         Stachowiak, A.N. & Irvine, D.J. Inverse opal hydrogel-collagen composite scaffolds as a supportive microenvironment for immune cell migration. J Biomed Mater Res A 85, 815 (2008).

4.         Madden, L.R. et al. Proangiogenic scaffolds as functional templates for cardiac tissue engineering. Proc Natl Acad Sci USA 107, 15211 (2010).

5.         Lee, J. et al. Implantable microenvironments to attract hematopoietic stem/cancer cells. P Natl Acad Sci USA 109, 19638-19643 (2012).

6.         Bersani, F. et al. Bioengineered Implantable Scaffolds as a Tool to Study Stromal- Derived Factors in Metastatic Cancer Models. Cancer Research 74, 7229 (2014).

7.         Yumoto, K. et al. Axl is required for TGF-beta2-induced dormancy of prostate cancer cells in the bone marrow. Sci Rep 6, 36520 (2016).

8.         Taichman, R.S. et al. GAS6 receptor status is associated with dormancy and bone metastatic tumor formation. PLoS One 8, e61873 (2013).

9.         Bragado, P. et al. TGF-beta2 dictates disseminated tumour cell fate in target organs through TGF-beta-RIII and p38alpha/beta signaling. Nat Cell Biol 15, 1351 (2013).

10.       Ruppender, N. et al. Cellular Adhesion Promotes Prostate Cancer Cells Escape from Dormancy. PLoS One 10, e0130565 (2015).

11.       Park, J. et al. Cancer cells induce metastasis-supporting neutrophil extracellular DNA traps. Sci Transl Med 8, 361ra138 (2016).

12.       Engblom, C. et al. Osteoblasts remotely supply lung tumors with cancer-promoting SiglecF(high) neutrophils. Science 358, eaal5801 (2017).

13.       Quail, D.F. & Joyce, J.A. Microenvironmental regulation of tumor progression and metastasis. Nat Med 19, 1423 (2013).

14.      Murgai, M. et al. KLF4-dependent perivascular cell plasticity mediates pre-metastatic niche formation and metastasis. Nat Med 23, 1176 (2017).

15.       Albrengues, et al. Neutrophil extracellular traps produced during inflammation awaken dormant cancer cells in mice. Science, 361,eaao4227 (2018)

16.       Chung, et al. Structural and molecular interrogation of intact biological systems, Nature, 497, 332 (2013).



Assistant Professor, University of Massachusetts-Amherst

Biomaterials & Tissue Engineering