Towards Biohybrid Lung Development: Establishment of a Porcine In Vitro Model

Abstract

Lung transplantation (LTx) is the only curative therapy option for patients with end-stage lung diseases, though only available for chosen patients. To provide an alternative treatment option to LTx, we aim for the development of an implantable biohybrid lung (BHL) based on hollow fiber membrane (HFM) technology used in extracorporeal membrane oxygenators. Crucial for long-lasting BHL durability is complete hemocompatibility of all blood contacting surfaces, which can be achieved by their endothelialization. In continuation to successful in vitro investigations using human endothelial cells (ECs), indicating general feasibility, the appropriate porcine in vivo model needs to be prepared and established to fill the translational data gap prior to patient's application. Therefore, isolation of porcine ECs from carotid arteries (pCECs) was established. Following, pCECs were used for HFM endothelialization and examined under static and dynamic conditions using cell medium or heparinized blood, to assess their proliferation capacity, flow resistance and activation state, especially under clinically relevant conditions. Additionally, comparative hemocompatibility tests between native and endothelialized HFMs were performed. Overall, pure pCECs formed a viable and confluent monolayer, which resisted applied flow conditions, in particular due to physiological extracellular matrix synthesis. Additionally, pCECs remained the non-inflammatory and anti-thrombogenic status, significantly improving the hemocompatibility of endothelialized HFMs. Finally, as relevant for reliable porcine to human translation, pCECs behaved in the same way as human ECs. Concluding, generated in vitro data justify further steps towards pre-clinical BHL examination, in particular BHL application to porcine lung injury models, reflecting the clinical scenario with end-stage lung-diseased patients.

Keywords: biohybrid lung; endothelialization; hemocompatibility; hollow fiber membrane; porcine endothelial cells.

Figure 1

Characterization of isolated pCECs. (a) Histogram view of cultured pCECs, analyzed via flow cytometry for the expression of CD31 (red) versus the IgG control (blue); (b) Calcein staining (green) of the confluent and viable pCECs monolayer on TCP with a characteristic cobblestone morphology; Immunofluorescence microscopy of pCECs on TCP for the detection of (c) the EC-specific junction protein VE-Cadherin (green) and (d) the extracellular matrix protein fibronectin (green); (c,d) Corresponding nuclei were counterstained with Hoechst 33342 (blue); (e) Real-time qRT-PCR expression analysis of inflammatory (E-Selectin and ICAM) and thrombogenic state (TM and TF) marker genes with and without TNFα stimulation. Gene expression levels of the stimulated group were compared to its respective unstimulated control group using an unpaired t-test. TCP: Tissue Culture Plastic; stimulated: six hours TNFα exposure; unstimulated: control group without TNFα exposure. Results are given as mean with SD (n = 3) (* p < 0.05; ** p < 0.01; **** p < 0.0001).

Figure 2

Viability and immunofluorescence staining of pCECs on HFMs. (a) Calcein staining of viable pCECs (green) forming a confluent monolayer on the HFM; (a) Insert depicts higher magnification; (b,c) Immunofluorescence staining for the detection of (b) VE-Cadherin (green) and (c) the de novo synthesized extracellular matrix protein fibronectin (green); (b,c) Corresponding nuclei were counterstained with Hoechst 33342 (blue).

Figure 3

Gene expression analysis for the assessment of the inflammatory and thrombogenic genotype of pCECs on HFMs. Expression levels of the inflammatory (E-Selectin and ICAM) and thrombogenic state (TM and TF) marker genes were compared between untreated pCECs on TCP versus HFMs and pCECs on HFMs stimulated with TNFα. A one-way ANOVA with correction for multiple comparisons (Šídák test) was used to compare these three groups as indicated in the figure. TCP: Tissue Culture Plastic; HFM: Hollow Fiber Membrane; stimulated: six hours of TNFα exposure; unstimulated: control group without TNFα exposure. Results are given as mean with SD (n = 3) (* p < 0.05; ** p < 0.01; *** p < 0.001; ns = no significance).

Figure 4

Comparison of pCECs under static and flow conditions. (a,b) Calcein staining of viable pCECs (green) after 24 h of static (a) or dynamic (15 mL/min flow) (b) conditions; (c) qRT-PCR analysis of flow-exposed pCECs regarding inflammatory activation (E-Selectin and ICAM), shear stress response (KLF 2), thrombogenic state (TM and TF) and matrix synthesis (Col 4 A1). Gene expression levels of the flow-exposed group were compared to the static control groups using an unpaired t-test. For abbreviations of gene names, see Table 2. Results are given as mean with SD (n = 3) (* p < 0.05; ** p < 0.01; ns = no significance).

Figure 5

Analysis of EC monolayer integrity after culture medium or blood flow exposure for six hours. (a,b) Calcein staining (green) of pCECs on HFMs that (a) were kept statically or were exposed to (b) flow conditions using culture medium (60 mL/min); (c,d) Cell Tracker staining (green) of pCECs on statically cultivated HFMs (c) or on 60 mL/min blood flow exposed HFMs (d).

Figure 6

Thrombogenicity assessment of HFMs with and without pCEC monolayer after blood flow exposure. Macroscopic examination of (a) non-endothelialized and (b) endothelialized HFMs; (a) black arrow = macroscopically visible thrombus; SEM images of (c,e) non-endothelialized and (d,f) endothelialized HFMs; (e) white arrow: fibrin, red +: erythrocyte, red *: thrombocyte; (f) white arrow: cell-cell contact within confluent EC-monolayer; (g) D-Dimer and (h) thrombocyte count level changes in the non-endothelialized and endothelialized HFM group before and after blood flow exposure. An unpaired t-test was used to compare both groups. Results are given in mean difference [%] with SD (n = 3) (** p < 0.01).