Acellular Tissue Engineered Vessels as Coronary Artery Bypass Grafts

Abstract

Coronary artery bypass graft (CABG) uses the patient's internal mammary artery and saphenous vein; however, unavailable or poor quality autologous vessels limit revascularization. This study addresses the critical need for alternative CABG conduits by evaluating a small diameter acellular tissue-engineered vessel ([sdATEV], 3.5 mm) in a primate model. Adult baboons (n = 5) underwent CABG to the right coronary artery (RCA) with an sdATEV. Patency, diameter, and cardiac function were longitudinally assessed by computed tomography angiography. All sdATEVs remained patent throughout the 6-month study. Computed tomography angiography demonstrated that the distal sdATEV diameter gradually remodeled to approximate the smaller baboon RCA. Histology and spatial transcriptomics revealed that sdATEVs recellularized with host endothelial and smooth muscle cells including a quiescent neomedia layer with gene expression patterns similar to the RCA, indicating that host cell ingrowth from the bypassed coronary artery regulates sdATEV diameter. These results suggest that the sdATEV is a durable alternative CABG conduit.

Keywords: coronary artery bypass graft; coronary computed tomography angiography; regenerative medicine; spatial genomics; tissue engineering; vascular biology.

Graphical abstract

Figure 1

sdATEV Production and Implantation in a Baboon CABG Model (A) Schematic of small diameter (3.5-mm) acellular tissue-engineered vessel (sdATEV) manufacturing process. (B) Formation of distal anastomosis of 3.5-mm diameter sdATEV to native right coronary artery (RCA) in adult baboon showing size mismatch to ∼1.5-mm diameter RCA (1.5-mm stainless steel coronary probe is placed in RCA). (C) Proximal anastomosis of sdATEV to aorta using parachute suturing technique. (D) Final aorta-to-RCA configuration of sdATEV on the heart prior to closure. CABG = coronary artery bypass graft.

Figure 2

sdATEV Luminal Remodeling Followed In Vivo by 3D CTA (A to C) Representative 3-dimensional (3D) computed tomography angiograms from 1 to 6 months showing luminal remodeling to mimic RCA diameter. White arrows indicate sdATEV. (D to F) Diagrammatic representation of luminal remodeling from 1 to 6 months. (G to I) Quantification of luminal diameter from 1 to 6 months across 5 animals; mean ± SD; ∗∗P < 0.010; ∗∗∗P < 0.001; 1-way analysis of variance with Dunnett post hoc test vs RCA. (J) Changes in luminal diameter (percentage of change labeled from 1 to 3 and 3 to 6 months), Significance (∗P < 0.050 and ∗∗P < 0.010) only found in comparison with respective 1-month measurements, all other comparisons not significant; 2-way analysis of variance with Tukey post hoc test. (K) Calculated wall shear stress (WSS) of RCA and sdATEV during study; mean ± SD. CTA = computed tomography angiography; D = distal; F = foot; H = head; L = left; M = midgraft; NS = nonsignificant; R = right.

Figure 3

sdATEV CABG Repopulation With Vascular SMCs and ECs (A and E) Hematoxylin and eosin (H&E)-stained sdATEV preimplantation shows extracellular matrix proteins but no nuclei (lack of hematoxylin, blue). (B and F) H&E stain of midgraft explanted at 6 months shows remodeling into multilayered tissue with surrounding adventitia and luminal neomedia tissue. (C and G) α-Smooth muscle actin (αSMA, red) and CNN1 (green) identifies smooth muscle cells (SMCs) in sdATEV wall and neomedia. (D, H) VWF (red) and CD31 (PECAM1, green) show endothelialization of sdATEV lumen (white arrows). Cell nuclei were stained blue with 4',6-diamidino-2-phenylindole (DAPI). (I) H&E stain of sdATEV-RCA anastomosis at 6 months shows neomedia originating from RCA. (J) Neomedia contains cells expressing αSMA (red) and CNN1 (green). (K) Endothelial cells (ECs) expressing VWF (red) and CD31 (green) line the lumen of both the RCA and sdATEV neomedia across the anastomotic interface (white arrows). L = lumen; other abbreviations as in Figure 1.

Figure 4

Low Cell Proliferation in sdATEV Wall and Neomedia at 6 Months (A to G) Six-month sdATEV explants from baboons with low host reactivity (n = 3) had few Ki67⁺ cells coexpressing αSMA (white arrows show coexpressing cells), and none coexpressing CD3, or CD20 with negligible Ki67⁺ cells in the neomedia. (H to N) sdATEV explants from baboons with high host reactivity (n = 2) exhibited more Ki67⁺ cells, predominantly CD20⁺ B cells and CD3⁺ T cells in both the sdATEV wall and neomedia. (O) No Ki67⁺ cells within the baboon native RCA. (P) Cellular proliferation in the sdATEV wall (2.8%) was higher than native RCA (0%, P < 0.010). (Q) Highly reactive animals had greater neomedia cellular proliferation (P < 0.050) than that of low-reactivity animals and native RCA, which were not significantly different. Mean ± SD; ∗P < 0.050; ∗∗P < 0.010; Kruskal-Wallis test with Dunn post hoc test. Abbreviations as in Figures 1 and 3.

Figure 5

SMC and EC Protein Expression Indicative of Quiescent sdATEV Neomedia Representative immunohistochemistry staining of 6-month sdATEV midgraft explants (left 2 columns) and RCA (right columns). (A to D) ECs express both VWF and endothelial nitric oxide synthase (eNOS) on the lumen of the sdATEV and RCA. (B to H) Prostacyclin PGI₂ synthase (PTGIS) is expressed by CD31⁺ luminal ECs as well as some CD31⁻ cells within the wall of both the sdATEV and RCA. (I to L) Expression of contractile SMC marker myocardin was colocalized with αSMA in the neomedia and sdATEV wall, whereas smoothelin (M-P) was colocalized with αSMA predominantly in the neomedia and not the sdATEV wall. Both myocardin and smoothelin colocalized with αSMA in the RCA, which matches the sdATEV neomedia and indicates its origination from the RCA media. (Q to T) TGFβ1, which may be associated with synthetic SMCs, was only found in the sdATEV and RCA adventitia. Abbreviations as in Figures 1 and 3.

Figure 6

Spatial Transcriptomics Reveals Similarity of Explanted sdATEV to Native RCA Cross-sectional images and gene expression profiles of sdATEV midgraft (A to C) and native RCA (D to F) from the same animal. H&E stained (A, D) images of tissue sections analyzed by spatial transcriptomics with overlays of computationally predicted clusters (B, E) at 32-μm resolution. (Clusters colored independently between B to C and E to F). (H to K) Log-normalized sum of gene expression within the sdATEV for gene sets derived from RCA clusters at 8-μm resolution. (H) Adventitia genes localize primarily to the outer sdATEV wall and neomedia interface (arrowheads). (I) Media genes primarily expressed in neomedia and outer sdATEV wall (brackets). (J to K) EC genes expressed along lumen (K, arrowheads) and at neomedia interface. Bars = 1 mm for low magnification and 200 μm for high magnification. Gene sets listed in Supplemental Table 4. Abbreviations as in Figures 1 and 3.

Figure 7

Comparative Gene Expression in Explanted sdATEV and Native RCA (A) Spatial transcriptomic gene expression patterns within 6-month sdATEV and RCA. Dot size indicates percentage of 8-μm resolution transcriptomic spots expressing the gene, coloration represents expression relative to ACTB. (B) Computationally predicted gene clusters overlayed on H&E-stained sections to define tissue regions in A. (C-D) Expression of vascular SMC genes (8-μm resolution) and (E to F) PCNA (32-μm resolution). (G to J) EC gene expression (32-μm resolution) with insets (black boxes, I-J) depicting coexpression (green) of EC genes (yellow) with endothelium shear stress genes (blue). (K-N) Expression of fibrillar collagens (8-μm resolution) with insets (black boxes, M to N) illustrating co-expression (arrowheads, green) of fibrillar collagens (blue) with matrix metalloproteases (yellow). Scale plots (log₂₎ represent sum of gene expressions in the selected family. Bars = 1 mm, low magnification; 200 μm, high magnification. Gene sets listed in Supplemental Table 4. Abbreviations as in Figures 1 and 3.