Generation of vascular chimerism within donor organs

Abstract

Whole organ perfusion decellularization has been proposed as a promising method to generate non-immunogenic organs from allogeneic and xenogeneic donors. However, the ability to recellularize organ scaffolds with multiple patient-specific cells in a spatially controlled manner remains challenging. Here, we propose that replacing donor endothelial cells alone, while keeping the rest of the organ viable and functional, is more technically feasible, and may offer a significant shortcut in the efforts to engineer transplantable organs. Vascular decellularization was achieved ex vivo, under controlled machine perfusion conditions, in various rat and porcine organs, including the kidneys, liver, lungs, heart, aorta, hind limbs, and pancreas. In addition, vascular decellularization of selected organs was performed in situ, within the donor body, achieving better control over the perfusion process. Human placenta-derived endothelial progenitor cells (EPCs) were used as immunologically-acceptable human cells to repopulate the luminal surface of de-endothelialized aorta (in vitro), kidneys, lungs and hind limbs (ex vivo). This study provides evidence that artificially generating vascular chimerism is feasible and could potentially pave the way for crossing the immunological barrier to xenotransplantation, as well as reducing the immunological burden of allogeneic grafts.

Figure 1

Ex vivo perfusion of rat and porcine organs and decellularization of the vascular tree. (a) Schematic representation of the hypothermic perfusion system used to decellularize rat and porcine organs. The circuit was driven by a peristaltic pump (Masterflex, Cole-Parmer) and controlled using a pressure transducer (Art-Line, Biometrix) and a monitor connected to the inflow cannula. (Illustration by Noa First Cohen). (b,c) Photographs of exemplary organs connected to the perfusion system: a rat kidney (b) cannulated using a 22G intravenous (IV) cannula placed in the abdominal aorta, and a porcine kidney (c) with Luer lock fittings placed into a single renal artery and two renal veins, and a 14G IV cannula placed into the ureter. Scale bars, 1 cm (B) and 5 cm (C). (d–s) Representative histological images of untreated and decellularized rat organs stained with hematoxylin and eosin (H&E): Low- and high-magnification of untreated aorta (d,f). Arrowheads show endothelial cells lining the luminal surface. Low- and high-magnification of decellularized aorta (e,g). Note the absence of the endothelial cell layer, while subendothelial tissue remains intact. Low- and high-magnification of untreated (h,j) and decellularized (i,k) rat kidneys. Note the appearance of acellular glomeruli with intact capillary wall structure, and preservation of epithelial cells of the surrounding collecting tubules. Representative images of untreated (l,n) and decellularized (m,o) lungs. Arrowheads show endothelial cells in untreated lung. Note the appearance of decellularized capillary surrounded by cellular alveoli (m) and de-endothelialized artery alongside cellular bronchiole (o). Low- and high-magnification of untreated (p,r) and decellularized (q,s) hind limb. Arrowheads show endothelial cells in untreated hind limb. PT, pressure transducer; P, pump; c, capillary; a, alveolus; bv, blood vessel; aw, airway. Scale bars, 50 µm (f,g,j,k,l,o,r,s), 100 µm (m,n) and 200 µm (d,e,h,i,p,q).

Figure 2

Isolation and characterization of human placental cells. (a,b) Microscopic images of human placental cells at low (a) and high (b) confluency. (c) Normal 46,XX karyotype of placental cells at passage 11. (d–f) Representative flow cytometry analysis, showing high expression level of EPC markers compared to unstained cells. (g) Placental cell migration in response to VEGF. Histogram is representative of cell counts in four random microscopic fields from two independent experiments (at passages 7 and 13). Bars indicate means ± SD (P < 0.01, non-parametric signed-rank test). Scale bars, 200 µm (a,b).

Figure 3

Cell seeding and generation of vascular chimerism in vitro. (a) H&E staining of human placental EPCs seeded on completely decellularized human umbilical vein. (b,c) H&E (B) and corresponding vWF immunohistochemistry staining (c) of human placental EPCs on de-endothelialized rat aorta. (d-f) SEM images of untreated (d), de-endothelialized and human EPC-seeded (f) rat aorta. (g) Corresponding high magnification SEM image of EPC-seeded aorta. Yellow arrows show cell-to-cell connections, yellow arrowheads show cell-to-ECM connections. Scale bars, 100 µm (a,c) and 50 µm (b).

Figure 4

Isolated perfusion and vascular decellularization of rat organs in situ. Illustrations of isolated organ perfusion technique (a–c), corresponding photographs (d–f) angiographic images (g–i), and histological analysis (j–l) of rat kidneys, lungs, and hind limbs, respectively. Scale bars, 100 µm (l) and 50 µm (j,k). (Illustrations by Noa First Cohen).

Figure 5

Normothermic perfusion, cell seeding and generation of vascular chimerism ex vivo. (a) Schematic representation of the normothermic perfusion system setup. The organ circuit was driven in a similar way to the hypothermic perfusion system described in Fig. 1a. The oxygenation circuit included a dialyzer used as membrane oxygenator (FX-10, Fresenius Medical Care) that was ventilated with 95% O₂/5% CO₂. Both the organ and the perfusate reservoir were placed in a heated water bath. (Illustration by Noa First Cohen). (b–d) Photographs of procured rat kidneys (b), lungs (c) and hind limb (d) together with their inflow and outflow cannulas, perfused ex vivo. (e–g) Corresponding H&E histological images. Arrows show human ECPs lining de-endothelialized rat blood vessels. (h–p) Seeded cells were further identified by immunostaining with anti-human mitochondria and anti-Ki67 antibodies, and FISH with DNA probe against human HER-2. Representative images showing Ki67 staining in kidneys (h) and lungs (i,j), human mitochondria staining in kidneys (k–m), and HER-2 staining in kidneys (n–p). Note positively-stained human cells rat blood vessels, alongside negatively-stained rat cells, generating vascular chimerism. White arrowheads show HER-2 positive cells displaying orange and green fluorescent dot signals inside cell nuclei stained with DAPI (blue). Dotted lines differentiate between positively-stained human EPCs and subendothelial tissue. Scale bars, 50 µm (e,h,i,l,m), 100 µm (f,g,j,k) and 25 µm (n–p).

Figure 6

Suggested pathway to immune tolerance. An illustration demonstrating the proposed methodology for generation of transplantable, immunologically-acceptable organs for future translation into clinical practice. Donated human placenta (A1) are used to isolate, expand and bank ABO- and HLA-screened, immunologically-acceptable cells (A2). Pig organs (A3) are procured and mounted on a perfusion machine (B) used to rebuild the organ’s vasculature using ABO- and HLA-compatible placental cells. The immunologically-acceptable organ is transplanted into a recipient (C). (Illustration by Noa First Cohen).