Primary Human Lung Alveolus-on-a-chip Model of Intravascular Thrombosis for Assessment of Therapeutics

Abstract

Pulmonary thrombosis is a significant cause of patient mortality; however, there are no effective in vitro models of thrombi formation in human lung microvessels that could also assess therapeutics and toxicology of antithrombotic drugs. Here, we show that a microfluidic lung alveolus-on-a-chip lined by human primary alveolar epithelium interfaced with endothelium and cultured under flowing whole blood can be used to perform quantitative analysis of organ-level contributions to inflammation-induced thrombosis. This microfluidic chip recapitulates in vivo responses, including platelet-endothelial dynamics and revealed that lipopolysaccharide (LPS) endotoxin indirectly stimulates intravascular thrombosis by activating the alveolar epithelium, rather than acting directly on endothelium. This model is also used to analyze inhibition of endothelial activation and thrombosis due to a protease activated receptor-1 (PAR-1) antagonist, demonstrating its ability to dissect complex responses and identify antithrombotic therapeutics. Thus, this methodology offers a new approach to study human pathophysiology of pulmonary thrombosis and advance drug development.

Figure 1. Microengineered model of human pulmonary thrombosis-on-chip

a) A conceptual schematic of the human lung showing that the alveoli interact with the neighboring blood vessels during hemostasis or pulmonary dysfunction. b) Engineering drawing of the microdevice containing two PDMS compartments separated by a thin porous membrane that reproduces the microarchitecture of the alveolar-capillary interface. c) Graphic illustration showing the top compartment (1 mm wide and 1 mm tall) is cultured with human primary alveolar epithelial cells and the entire bottom chamber (1 mm wide and 250 µm tall) lined with human endothelial cells forming a lumen. Whole blood is perfused through the bottom chamber and thrombus formation is visualized using fluorescence microscopy from the bottom. d) Micrograph of human lung alveolar epithelial cells (ZO1, left; brightfield, right; Scale bar, 50 µm) and e) vascular endothelial cells (VE-cadherin, left; brightfield, right, Scale bar, 50 µm) f) Sideview of confocal micrographs showing junctional structures, after twelve days of co-culture, of a single layer of the primary alveolar epithelium at the top chamber (purple, stained with E-cadherin) and endothelial monolayers covering the entire surface of the lower chamber (green, stained with VE-cadherin), through which blood perfusion takes place. Scale bar: 100 µm.

Figure 2. Analysis of dynamical progression of thrombus formation in the alveolus chip

a) Fluorescence micrographs depicting a section of the imaged microchannel (Scale bar, 25 µm) showing platelet accumulation over time (left to right) on collagen (top), a healthy blood vessel (middle), TNF-α stimulated vessel (bottom) and b) platelet accumulation after 4 minutes of laser-induced injury on a mouse cremaster arteriole (top, bright field and bottom, fluorescence; Scale bar, 25 µm, top and bottom). c) Fluorescent micrographs of a large section of the vascular chamber showing luminal thrombus formation on bare collagen (topmost) and TNF-α stimulated endothelium in a dose dependent manner (bottom three; Scale bar, 100 µm). d) Sensitivity analysis of the platelet-endothelial dynamics algorithm (PBD index), showing that in conditions of vascular injury/collagen or healthy endothelium, the dynamical events characterizing clot formation are nearly absent but they increase in a TNF-α dose dependent manner. The PBD index is also sensitive to applied shear rate (n = 3, *P<0.05, 2-way ANOVA). e) ICAM-1 expression on the endothelial cells after stimulation of the vessel with TNF-α (n = 4, *P<0.05, unpaired t-test). f) Image colormap showing the coefficient of variability (calculated from the fluorescence image timeseries) within a laser-induced thrombus in vivo (left) and TNF-α induced thrombus on a vsacular lumen in vitro (right). Scale bar, 20 µm.

Figure 3. TNF-α induced endothelial disruption and thrombus formation in alveolus chip

Inside the lung alveolus-on-a-chip that was either left untreated (0 ng mL⁻¹) or treated with 5 or 100 ng mL⁻¹ of TNF-α, a) measurement of vascular permeability (fluorescence normalized by untreated vascular tissue; n=4), b) vascular ICAM-1 measured on the vascular surface (n=4), and c) measurement of PBD index after whole blood perfusion (n=3). a–c, *P<0.05, **P<0.01, 1-way ANOVA. d) Representative fluorescence micrographs showing platelet aggregates (red) and fibrin (yellow) at the end of blood perfusion through the alveolus-chip after TNF-α stimulation (Scale bar, 100 µm).

Figure 4. LPS-induced endothelial disruption and thrombus formation when LPS was added to the vascular channel of a chip lined only by vascular endothelium or to the epithelial channel of an intact lung alveolus chip

a) Vascular permeability measurements (fluorescence normalized by untreated vascular tissue; n=4) demonstrate that LPS only compromised barrier function when presented to the epithelium in an organ-level context. b) Representative confocal micrographs showing that VE cadherin-labeled endothelial cell-cell adhesions (green) open and retract when LPS was added to the chip containing both the epithelium and endothelium in an organ context (Organ), but not when added directly to the endothelium alone (tissue) (bar, 100 µm; arrows indicate gaps; blue, DAPI-stained nuclei). c) Quantification of averaged gap areas resulting from discontinuous VE-cadherin staining, as shown in b (n=4).d) Quantification of vascular ICAM-1 on the endothelial cell surface under the conditions shown in b (n=4). e) Measurement of PBD indices after whole blood perfusion through the control and LPS-treated lung alveolus chips (n=3). f) Representative fluorescence micrographs showing platelet aggregates and fibrin within the vascular channel at the end of blood perfusion through the alveolus chip (bar, 100 µm;a,c,d,e *P<0.05, **P<0.01, 2-way ANOVA). g) Quantification of cytokines released, as measured within the effluent of the vascular channel, showing significant differences in a subset of cytokines following 2h of LPS stimulation compared to untreated conditions (n=2).

Figure 5. Therapeutic effect of a PAR-1 inhibitor (parmodulin) in lung injury

a) PBD index measured in the chip after blood perfusion through the vascular channel of an alveolus chip previously exposed to LPS (100 ng mL⁻¹ for 2 hours) on the epithelial side, while the blood, the endothelium or both were treated with parmodulin (PM2) before perfusion. b) Vascular permeability (fluorescence normalized by untreated vascular tissue) measured inside the LPS-treated microdevice with or without treatment of parmodulin 2. c) Vascular permeability measurements in vivo after intratracheal delivery of LPS (1 µg per mouse) for 2 hours and systemic delivery of parmodulin (PM2, 30 µM) for 4 hours (a-c, n=3. *P<0.05, **P<0.01, 1-way ANOVA).