(25c) Neutrophil-Vasculature Interactions Promote Pro-Recurrent Niche Formation Post-Radiotherapy

Background and Motivation: Locoregional recurrence remains a challenge for triple negative breast cancer (TNBC) patients and treatment options are limited due to lack of hormone receptor and HER2 expression. Patients with high neutrophil-to-lymphocyte ratios after radiotherapy (RT) are at significant risk for recurrence, yet the mechanisms underlying this risk factor and the role of neutrophils in recurrence remain unclear. Recent work demonstrates that RT alters the mammary adipose microenvironment, damaging the normal tissue and enabling chronic immune cell infiltration that attracts circulating tumor cells (CTCs) to reseed at the primary tumor site. Furthermore, RT-damaged vasculature in other cancer types has been hypothesized to promote an innate immune cell-driven cascade that leads to re-vascularization and ultimately primary tumor recurrence. Adipose tissue is highly vascularized, and we therefore hypothesize that following RT, irradiated vasculature primes neutrophils towards a pro-tumor phenotype and creates a pro-recurrent niche favorable for CTC outgrowth. To test this hypothesis, we characterized neutrophil and vascular changes in an in vivo TNBC recurrence model and used these findings to develop an in vitro microfluidic model of the mammary gland vasculature, allowing us to probe the effects of RT and neutrophil-vasculature interactions on tumor cell behavior.

Methods: To model recruitment of CTCs in vivo, BALB/c mice were inoculated with 4T1 TNBC cells into the mammary fat pad (MFP) to serve as a source of CTCs. After 1.5 weeks, contralateral MFPs were irradiated to 20 Gy. MFPs were resected at 10 days post-RT and processed for immunohistochemistry, flow cytometry, and bulk RNA-sequencing analysis of neutrophil phenotype and molecular and structural changes of the vasculature. To identify effects of the radiation-induced endothelial cell secretome—which has been shown to shift towards a pro-inflammatory profile post-RT—on neutrophil behavior, we cultured murine endothelial cells in flasks, irradiated them to 0 or 10 Gy, and collected conditioned media (CM) at 2 or 7 days post-RT. Bone marrow-derived neutrophils (BMDNs) from BALB/c mice were then cultured in the endothelial cell CM for 2 hours before staining with SYTOX green and analyzing via flow cytometry to quantify induction of neutrophil extracellular trap (NET) release (NETosis).

To further probe molecular mechanisms of neutrophil-vasculature interactions in a more biologically relevant model, we designed a microfluidic device to model the vascular endothelial barrier in mammary adipose tissue. Devices were fabricated from polydimethylsiloxane using standard soft lithography techniques and consisted of two horizontally stacked channels separated by a 3µm porous membrane to enable cell migration between compartments. Membranes were coated with collagen I and fibronectin to form the extracellular matrix. The top channel of each device was seeded with murine endothelial cells and cultured under flow to model the vascular endothelial compartment, and the bottom channels were seeded with murine mammary fibroblasts under static conditions to model the mammary tissue. Devices were irradiated to 0 or 10 Gy and cultured for 2 or 7 days post-RT. Neutrophils were stained with Far Red CellTrace, flowed into the top channel at 2 and 7 days post-RT, and allowed to interact with the endothelial barrier for 4 hours before flowing GFP⁺ 4T1 tumor cells into the top channel of devices. Neutrophil and tumor cell invasion was analyzed using fluorescent imaging. Changes in endothelial cell expression were analyzed by fixing the top channel of devices and performing immunocytochemistry for ZO-1 and NF-kB-related proteins. Permeability was assessed post-RT and post-neutrophil infiltration by adding 40 kDa FITC-dextran to the top channel of the devices and analyzing fluorescence intensity within the bottom channel after 1 hour.

Results: At 10 days post-RT, immunohistochemistry analysis of Ly6G⁺ cells and flow cytometry analysis of Ly6C^loLy6G⁺ cells revealed significantly higher neutrophil infiltration in irradiated MFPs compared to unirradiated MFPs. Furthermore, these neutrophils exhibited a pro-tumor phenotype evidenced by increased expression of CD84 and a shift in side scatter, suggesting activation of the neutrophils by the irradiated microenvironment. Bulk RNA-sequencing analysis of the MFPs also showed increased expression of genes related to NETosis, including Mpo, Elane, S100a8, S100a9, and Mmp8. Reactome pathway analysis of this sequencing data showed activation of tissue remodeling pathways such as non-canonical NF-kB, Hedgehog, and Notch signaling, which corresponded with increased vessel counts and a higher variability in vessel size and suggests post-RT vascular remodeling concurrent with pro-tumor neutrophil infiltration. BMDNs cultured in endothelial cell CM also exhibited higher induction of NETosis when cultured in CM from irradiated endothelial cells, highlighting the direct effects of the irradiated endothelial cell secretome on neutrophil behavior. In our microfluidic model, endothelial barriers showed increased permeability and decreased expression of ZO-1 after both RT and neutrophil co-culture, demonstrating evidence of damage by the infiltrating neutrophils. We also observed increased neutrophil adhesion to irradiated endothelial monolayers followed by increased tumor cell invasion. Neutrophils have been demonstrated to damage the vasculature while also inducing angiogenic signaling through NETosis and release of proteins such as S100A8/A9 and neutrophil elastase. These interactions could enhance the capability of tumor cell invasion through the vasculature, and are currently being evaluated in the microfluidic model in conjunction with activation of NF-kB signaling, which we observed in vivo.

Conclusions: These studies highlight the significant changes in neutrophils and vasculature that occur post-RT and begin to identify the mechanisms that underly the clinical association of a high neutrophil-to-lymphocyte ratio with TNBC recurrence. We demonstrate that the irradiated microenvironment—specifically secreted factors from endothelial cells—plays a large role in activating neutrophils and may shift them towards a pro-tumor phenotype and promote a pro-recurrent niche. We also provide an in vitro model that incorporates fluid flow and improves upon traditional cell culture models of the vasculature, ultimately serving to increase the translation of our findings to in vivo mouse models. Further studies will focus on the molecular changes induced by neutrophils on endothelial cells in this environment and evaluate signaling pathways that can be targeted in vivo to reduce CTC recruitment and recurrence.