Two-Channel Compartmentalized Microfluidic Chip for Real-Time Monitoring of the Metastatic Cascade

Abstract

Metastases are the primary cause of death in cancer patients. Small animal models are helping in dissecting some of key features in the metastatic cascade. Yet, tools for systematically analyze the contribution of blood flow, vascular permeability, inflammation, tissue architecture, and biochemical stimuli are missing. In this work, a microfluidic chip is designed and tested to replicate in vitro key steps in the metastatic cascade. It comprises two channels, resting on the same plane, connected via an array of rounded pillars to form a permeable micro-membrane. One channel acts as a vascular compartment and is coated by a fully confluent monolayer of endothelial cells, whereas the other channel is filled with a mixture of matrigel and breast cancer cells (MDA-MB-231) and reproduces the malignant tissue. The vascular permeability can be finely modulated by inducing pro-inflammatory conditions in the tissue compartment, which transiently opens up the tight junctions of endothelial cells. Permeability ranges from 1 m/sec (tight endothelium) to 5 m/sec (TNF- at 50 ng/mL overnight) and up to  10 m/sec (no endothelium). Fresh medium flowing continuously in the vascular compartment is sufficient to induce cancer cell intravasation at rates of 8 cells/day with an average velocity of  0.5 m/min. On the other hand, the vascular adhesion and extravasation of circulating cancer cells require TNF- stimulation. Extravasation occurs at lower rates with 4 cells/day and an average velocity of  0.1 m/min. Finally, the same chip is completely filled with matrigel and the migration of cancer cells from one channel to the other is monitored over a region of about 400 m. Invasion rates of 12 cells/day are documented upon TNF- stimulation. This work demonstrates that the proposed compartmentalized microfluidic chip can efficiently replicate in vitro, under controlled biophysical and biochemical conditions, the multiple key steps in the cancer metastatic cascade.

Advanced Microfluidic Models of Cancer and Immune Cell Extravasation: A Systematic Review of the Literature

“…Shear stress was also shown to negatively correlate with the adhesion of breast cancer cells to the endothelium, since a shear stress of 0.16 dyn/cm 2 within an endothelialized channel reduced cancer cell adhesion compared to a half-shear stress value (0.08 dyn/cm 2 ). This trend was present also when an inflammatory stimulus was added, once again showing the prevalent effect of shear stress over inflammation, although a higher number of cells adhered to the endothelium compared to a healthy condition (Mollica et al, 2019 ).…”

“…Inflammation is a body response activated by toxic stimuli and pathological conditions, such as infections or tissue injuries (Medzhitov, 2008 ), but it is also a hallmark of cancer, involved in the development and progression of malignancies (Diakos et al, 2014 ). Inflammation influences cell extravasation mainly through a modulation of endothelial properties, decreasing the tightness of endothelial junction (Mollica et al, 2019 ) and leading to a higher permeability of the endothelial barrier (Chen et al, 2013 ; Menon et al, 2017 ; Mollica et al, 2019 ). Chemotaxis is fundamental for the homeostatic trafficking of immune cells and for the enrollment of leukocytes to infection and inflammation sites (Luster, 2001 ).…”

“…With regards to the investigation of the role of inflammation in extravasation, microfluidic models consent to generate gradients of inflammatory factors, usually TNF-α. These factors can be directly diluted into the culture medium (Chen et al, 2013 ; Menon et al, 2017 ) or incorporated into a gel (Mollica et al, 2019 ). The latter approach better mimics the physiological origin of the inflammatory stimuli, which usually takes place deep in the tissue, and eventually reaches the endothelial barrier.…”

“…The latter approach better mimics the physiological origin of the inflammatory stimuli, which usually takes place deep in the tissue, and eventually reaches the endothelial barrier. When supplemented directly in the medium, TNF-α is usually added in a concentration of 10 ng/mL, whilst its addition in the gel needs a higher concentration to reach a biological effect (Mollica et al, 2019 ). It is fundamental to apply the correct dose of inflammatory factors, considering that they can be cytotoxic when used at high concentrations, both on extravasating cells and on endothelial cells.…”

“…As for the biological effect, also the cytotoxic effect depends on the mode of application and on the timing. A concentration of 10 ng/mL directly imposed on endothelial cells was reported to be toxic, leading to cell death and ruptures in a self-assembled microvasculature (Chen et al, 2013 ), while no toxic effect occurred at higher concentration of TNF-α (50 ng/mL) when the molecule was included in a gel (Mollica et al, 2019 ).…”