Non-invasive and real-time measurement of microvascular barrier in intact lungs

Abstract

Microvascular leak is a phenomenon witnessed in multiple disease states. In organ engineering, regaining a functional barrier is the most crucial step towards creating an implantable organ. All previous methods of measuring microvascular permeability were either invasive, lengthy, introduced exogenous macromolecules, or relied on extrapolations from cultured cells. We present here a system that enables real-time measurement of microvascular permeability in intact rat lungs. Our unique system design allows direct, non-invasive measurement of average alveolar and capillary pressures, tracks flow paths within the organ, and enables calculation of lumped internal resistances including microvascular barrier. We first describe the physiology of native and decellularized lungs and the inherent properties of the extracellular matrix as functions of perfusion rate. We next track changing internal resistances and flows in injured native rat lungs, resolving the onset of microvascular leak, quantifying changing vascular resistances, and identifying distinct phases of organ failure. Finally, we measure changes in permeability within engineered lungs seeded with microvascular endothelial cells, quantifying cellular effects on internal vascular and barrier resistances over time. This system marks considerable progress in bioreactor design for intact organs and may be used to monitor and garner physiological insights into native, decellularized, and engineered tissues.

Keywords: Bioreactor; Ex vivo; Lung; Mathematical model; Microvascular; Tissue engineering.

Figure 1.. Design of bioreactor system, measurement methods, and mathematical models.

(A) A schematic of the key components of the lung bioreactor. The pulmonary artery (PA), pulmonary vein (PV), and trachea of a lung are all cannulated and attached to ports within the bioreactor. (B) A lumped parameter circuit analog of the lung in this bioreactor system. The peristaltic pump is represented as a current source. The pleural pressure, or fluid height in the bioreactor, is represented as electrical ground. (C) Key pressures and flows outside of the lung, as defined in the glossary and the main text. (D) Key pressures and flows inside of the lung for a model alveolus, as defined in the glossary and the main text. (E) Calculation of exit flow rates. The resistance of the venous tubing (R_tub,vein) is known and well-characterized. By measuring a pressure drop across it (P_vein,meas - P_pleur), Ohm’s law is used to calculate the flow rate out the vein (Q_vein). Identical methods are used for determining the tracheal flow rate (Q_trach = [P_trach,meas – P_pleur] / R_tub,trach). The pleural flow rate is found by subtracting the venous and tracheal flow rates from the arterial flow rate. (F) Measurement of average internal pressures, shown for the vein and trachea (top) and for the artery (bottom). To measure the average pressure at the end of the capillaries (P_cap,end), flow out the vein is briefly shut off (Q_vein = 0). The pressure measured at the vein (P_vein,meas) is now identical to the pressure at the end of the capillaries. The pressure immediately after the step change in flow is representative of the end-capillary pressure right before the flow is turned off. The average pressure in the alveoli (P_alv) is found in a comparable manner by briefly shutting off flow out the trachea (Q_trach = 0) and looking at the measured tracheal pressure (P_trach,meas) immediately after the change. The average pressure at the start of the capillaries (P_cap,st) is found in a comparable manner by briefly shutting off perfusion to the whole organ (Q_art = 0) and looking at the measured arterial pressure (P_art,meas) immediately after the change.

Figure 2.. Characterizing native and decellularized lungs.

(A) Experimental setup. N = 3 native and N = 3 decellularized lungs were set up in the bioreactor system then perfused at a perfusion rate ranging from 5 to 60 mL/min with DMEM or PBS, respectively. (B) Barrier Resistance versus perfusion rate. (C) Capillary Resistance versus perfusion rate. (D) Percent Venous Flow Rate versus perfusion rate. (E) Percent Tracheal Flow Rate versus perfusion rate. (F) Percent Pleural Flow Rate versus perfusion rate. (G) Start-Transmural Pressure versus perfusion rate. (H) End-Transmural Pressure versus perfusion rate. (I) Transpulmonary Pressure versus perfusion rate. Asterisks denote significance by a students t-test, corrected for multiple comparisons.

Figure 3.. Manifestations of total lung resistance in decellularized lung architecture.

The decellularized lungs with the lowest total resistance and highest total resistance are directly compared. (A-D) Representative H&E images, 10x (A, C) and 40x (B, D), from the lowest resistance lung (A, B) and highest resistance lung (C, D). Scale bars are 400 μm (A, C) and 50 μm (B, D). (E-H) Representative Movat Pentachrome images, 20x (E, G) and 40x (F, H), from the lowest resistance lung (E, F) and highest resistance lung (G, H). Nuclei and elastic fibers are stained black, collagen fibers are stained yellow, mucopolysaccharides are stained blue, and acidophilic tissue components such as muscle and fibrin are stained red. Scale bars are 200 μm (E, G) and 100 μm (F, H). (I-L) Representative immunofluorescence images, 40x, for collagen IV (I, K) and elastin (J, L) from the lowest resistance lung (I, J) and the highest resistance lung (K, L). Scale bars are 50 μm. Exposure was 600 ms (I, K) and 250 ms (J, L). (M-P) Representative TEM images from the lowest resistance lung (M, N) and the highest resistance lung (O, P). Scale bars are 2 μm.

Figure 4.. Tracking native lung injury.

(A) Experimental setup.N = 3 native lungs were set up in the bioreactor system then perfused at 40 mL/min with PBS. (B) Non-vascular resistances. Lumped barrier, pleural, and airway resistances versus time for one representative lung. (C) Vascular resistances. Lumped arterial, capillary, and venous resistances versus time for the same representative lung (D) Flow rates. Venous, tracheal, and pleural flow rates versus time for the same representative lung. (E) Extra-pulmonary pressures. True arterial, venous, and tracheal pressures versus time for the same representative lung. (F) Intra-pulmonary pressures. Start-capillary, end-capillary, and alveolar pressures versus time for the same representative lung. (G) Pressures differences. Start-transmural, end-transmural, and transpulmonary pressures versus time for the same lung. (H) Phases of decay. Global observations for five distinct phases of decay. Results for the other two lungs may be found in Supplemental Figure S4.

Figure 5.. Tracking engineered endothelialized lung culture over 24 hours.

(A) Experimental timeline. Experimental time zero was defined as the start of perfusion. The vein was seeded 2.5 hours prior to perfusion, while the artery was seeded 1.25 hours prior to perfusion. Perfusion was ramped up to 20 mL/min over 90 minutes, at which point it was maintained for 24 hours before taking down the lung. (B) Experimental setup. N = 3 engineered lungs seeded with 100 million endothelial cells were set up in the bioreactor system then perfused at 20 mL/min with culture medium. (C-D) Representative H&E images. 40x and 4x images from one EC-seeded lung after 24 hours of culture. Scale bars are 100 μm (C) and 1 mm (D). (E) Central resistances. Lumped barrier, capillary, and pleural resistances versus time. (F) Peripheral resistances. Lumped arterial, venous, and airway resistances versus time. (G) Flow rates. Venous, tracheal, and pleural flow rates versus time. (H) Pressures differences. Start-transmural, end-transmural, and transpulmonary pressures versus time. (I) Extra-pulmonary pressures. True arterial, venous, and tracheal pressures versus time. (J) Intra-pulmonary pressures. Start-capillary, end-capillary, and alveolar pressures versus time.

Figure 6.. Tracking engineered endothelialized lung culture over 4 days.

(A) Experimental timeline.Experimental time zero was defined as the start of perfusion. The vein was seeded 2.5 hours prior to perfusion, while the artery was seeded 1.25 hours prior to perfusion. Perfusion was ramped up to 20 mL/min over 90 minutes, at which point it was maintained for 96 hours before taking down the lung. Half of the culture medium was exchanged every 24 hours, coupled with a brief (< 5 min) cessation of perfusion and a corresponding ramp up of perfusion over 30 minutes afterwards. (B) Experimental setup. N = 3 engineered lungs seeded with 100 million endothelial cells were set up in the bioreactor system then perfused at 20 mL/min with culture medium. (C-D) Representative H&E images. 40x and 4x images from one EC-seeded lung after 4 days of culture. Scale bars are 100 μm (C) and 1 mm (D). (E) Central resistances. Lumped barrier, capillary, and pleural resistances versus time. (F) Peripheral resistances. Lumped arterial, venous, and airway resistances versus time. (G) Flow rates. Venous, tracheal, and pleural flow rates versus time. (H) Pressures differences. Start-transmural, end-transmural, and transpulmonary pressures versus time. (I) Extra-pulmonary pressures. True arterial, venous, and tracheal pressures versus time. (J) Intra-pulmonary pressures. Start-capillary, end-capillary, and alveolar pressures versus time.

Figure 7.. Comparing native, decellularized, and engineered lungs.

Various parameters are compared between native lungs, decellularized lungs, and engineered endothelialized lungs after four days of culture (EC – 96hr), all under identical bioreactor conditions. (A) Central resistances. Lumped barrier, capillary, and pleural resistances. (B) Peripheral resistances. Lumped arterial, venous, and airway resistances. (C) Flow rates. Venous, tracheal, and pleural flow rates. (D) Extra-pulmonary pressures. True arterial, venous, and tracheal pressures. (E) Intra-pulmonary pressures. Start-capillary, end-capillary, and alveolar pressures. (F) Pressure differences. Start-transmural, end-transmural, and transpulmonary pressures. Asterisks denote significance levels by an ordinary one-way ANOVA with multiple comparisons, with one asterisk indicating p < 0.05, two indicating p < 0.01, and three indicating p < 0.001.