Mechanically reinforced biotubes for arterial replacement and arteriovenous grafting inspired by architectural engineering

Abstract

There is a lack in clinically-suitable vascular grafts. Biotubes, prepared using in vivo tissue engineering, show potential for vascular regeneration. However, their mechanical strength is typically poor. Inspired by architectural design of steel fiber reinforcement of concrete for tunnel construction, poly(ε-caprolactone) (PCL) fiber skeletons (PSs) were fabricated by melt-spinning and heat treatment. The PSs were subcutaneously embedded to induce the assembly of host cells and extracellular matrix to obtain PS-reinforced biotubes (PBs). Heat-treated medium-fiber-angle PB (hMPB) demonstrated superior performance when evaluated by in vitro mechanical testing and following implantation in rat abdominal artery replacement models. hMPBs were further evaluated in canine peripheral arterial replacement and sheep arteriovenous graft models. Overall, hMPB demonstrated appropriate mechanics, puncture resistance, rapid hemostasis, vascular regeneration, and long-term patency, without incidence of luminal expansion or intimal hyperplasia. These optimized hMPB properties show promise as an alternatives to autologous vessels in clinical applications.

Fig. 1.. The appearance and physical properties of hPB prepared in the rat subcutis.

(A) Schematic diagram of the architectural engineering–inspired generation of mechanically reinforced biotubes. (B) The morphological characterization of hPS with different fiber winding angles. Stereoscopic images of cross sections (i) and outer surface (ii) of hPS. SEM images showing fiber fusion at crossover points of hPS (iii, red arrow indicates fusion site). (C) Process images of dorsum wound healing after SI in rats. (D) The morphological characterization of hPB. Stereoscopic images of cross sections (i) and outer surface (ii) of hPB. H&E staining of hPB cross sections showed complete tissue capsule formation in all three kinds of hPB (iii). (E and F) Quantitative analysis of wall thickness (E) and inner diameter (F) of nPS, hPS, nPB, and hPB based on stereoscopic cross-sectional images (n = 5). (G) Stereoscopic observation of suturability of TB, hPB, and rAA. (H) Quantitative analysis of suture retention (n = 5). Statistical significance was calculated by two-way ANOVA with Tukey’s test. The symbol “#” denotes the comparison between different groups, ^#P < 0.05, ^###P < 0.001. (I) The burst pressure of rAA, TB, nPB, and hPB (n = 5). The red line denotes 1600 mmHg, the threshold for burst pressure of vascular replacements. Statistical significance was calculated by two-way ANOVA with Tukey’s test. The symbol “+” indicates the comparison in the same group, ⁺P < 0.05. (J and K) Quantitative analysis of the radial (J) and axial (K) mechanical properties of rAA, TB, and the hPB with three different fiber winding angles (n = 5). Statistical significance was calculated by one-way ANOVA with Tukey’s test. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 2.. Evaluation of the potential of hMPB as autologous grafts in rAA replacement model.

(A) Subcutaneously prepared hPBs were used as autologous abdominal artery replacements in rats. (B) CDU images of hMPB-V1m and hMPB-V3m and (C) corresponding measurement of luminal diameter. (D) Quantification of in vivo compliance of implanted hMPB and rAA. (E) Lumen and enface views of hMPB-V1m and hMPB-V3m. (F) H&E images of hMPB-V1m, hMPB-V3m, and rAA cross sections; quantification of hMPB neointima thickness shown alongside. Black dashed lines represent PS wall boundary (hMPB, hMPB-V1m, and hMPB-V3m) or boundary between tunicae media and adventitia layers (rAA). (G) EC coverage was visualized and calculated on the basis of anti-eNOS IF staining of hMPB longitudinal sections before and after VI. (H) EC coverage of lumen surfaces of hMPB-V1m and hMPB-V3m observed by SEM. (I) Co-stained cross sections show the distribution of ECs (eNOS, red) and SMCs (MYH, green) in hMPB-V1m, hMPB-V3m, and rAA. White dashed lines represent PS wall boundary (hMPB, hMPB-V1m, and hMPB-V3m) or the boundary between tunicae media and adventitia layers (rAA). (J) MYH⁺ SMC layer thickness calculated from eNOS/MYH co-staining. (K) Representative curves of physiological functions of hMPB, hMPB-V3m, and rAA in response to vasodilators and vasoconstrictors. (L) Quantification of constriction in response to KCl and AD; (M) Quantification of relaxation in response to ACh and SNP. Samples were preconstricted with AD, and then responses to vasodilators were assessed. Data are representative of n = 5 rats. Unpaired Student’s t test (D and G) and one-way ANOVA followed by Tukey’s test (C, F, J, L, and M) were performed. *P < 0.05, **P < 0.01, ***P < 0.001. NS, not significant.

Fig. 3.. The availability of hMPB in different species at 30 days after SI.

(A) Tailor-made hMPB could be fabricated in subcutaneous space of rabbit. (B) hMPB prepared in subcutaneous space of canine had excellent flexibility. (C) hMPB prepared in subcutaneous space of sheep still maintained original skeleton structure despite oblique cutting, clamping, and twisting. H&E staining of cross sections showed that subcutaneously implanted hMPS in rats, canine, and sheep have been filled with autologous cells and tissue at 30 days after operation. IF staining with anti–α-SMA antibody showed that the lumen surface of hMPB prepared in different species were all covered by a thin layer of α-SMA⁺ cell with circumferential alignment, and microvasculature was present throughout hMPB. The white dashed lines represent the boundary between inner α-SMA⁺ cell tissue layer and hMPS wall in the hMPB.

Fig. 4.. hMPB and ePTFE graft implantation in cCA replacement model.

(A) hMPB and ePTFE graft images before VI and immediately after VI (n = 3). (B) Representative CDU, angiography, CT 3D reconstruction, and MRI images of hMPB and ePTFE grafts at 7 months after VI (yellow arrows indicate suture sites). (C to E) Ultrasound-based measurement of flow velocity (C and D) and luminal diameter (E) of hMPB and ePTFE grafts after VI (n = 3). Statistical significance was calculated by two-way ANOVA with Tukey’s test. Symbol (+) indicates the comparison in the same group, ⁺P < 0.05. (F) Quantification of in vivo compliance of the hMPB-V7m and ePTFE-V7m as well as their adjacent proximal and distal cCA (n = 3). Statistical significance was calculated by two-way ANOVA with Tukey’s test. Symbol (+) indicates the comparison in the same group, ⁺⁺P < 0.01, ⁺⁺⁺P < 0.001. (G) hMPB-V7m and (H) ePTFE-V7m exterior view (i) and five cross-sectional pieces (ii). Green boxes indicate zoomed exterior views. Green dashed boxes indicate sutured end and magnified view. Red dashed boxes indicate the pieces taken from the middle of the graft for observation of graft lumens.

Fig. 5.. hMPB and ePTFE grafts remodeling at 7 months after VI into cCA.

(A) Schematic diagram of corresponding analysis for five vessel samples of hMPB-V7m and ePTFE-V7m. (B) H&E staining of cross sections. (C) Co-IF staining images with vWF and α-SMA antibody. (D) The contractile SMCs were visualized using IHC staining with MYH antibody. (E) Quantification of contractile SMCs thickness based on MYH IHC staining images of cCA and hMPB-V7m (n = 3). (F) WB analysis of α-SMA and MYH protein expression of corresponding samples. (G) Quantification of α-SMA and MYH protein from the WB analysis (n = 3). (H) SEM images showed the degradation of hMPS within hMPB-V7m. (I) Molecular weights analysis of hMPS and PCL from extracts of hMPB and hMPB-V7m by GPC (n = 3). Statistical significance was calculated by one-way ANOVA with Tukey’s test. *P < 0.05, ***P < 0.001. (J) Stress-strain curves of corresponding samples. (K) Representative curves of physiological functions of corresponding samples displaying different degrees of sensitivity to vasodilators and vasoconstrictors. (L) Quantification of constriction in response to KCl and AD (n = 3). Statistical significance was calculated by one-way ANOVA with Tukey’s test. *P < 0.05, **P < 0.01. (M) Quantification of relaxation in response to ACh and SNP (n = 3).

Fig. 6.. hMPB implantation in a sheep AVG model.

(A) Subcutaneous implant of hMPS in sheep neck according to prepositioned tunnel to prepare hMPB connecting the carotid artery and jugular vein (n = 3). (B) Macroscopic observations before and after 16G hemodialysis needle punctures. (C) Quantitative analysis of hMPB mechanical properties after 0, 8, 16, and 24 punctures (i to iii). Straight-across (0°) and oblique (45°) suture retention of hMPB and sCA were assessed and statistically analyzed (iv) (n = 5). (D) Macroscopic observation and CDU showed that after hMPB autografts as AVG, the hMPB segment still integrated with subcutaneous tissue, allowed for immediate needle puncture, and achieved hemostasis with 5 min of applied pressure. (E) C-model color ultrasound showed that no blood flowed from puncture sites after pressing. (F) Representative ultrasound images of hMPB as AVG in sheep at 3 months (hMPB-A3m). (G and H) Ultrasound-based measurement of proximal, middle, and distal diameter of hMPB (G) and flow rate (H) over the course of implantation (n = 3). (I) Explant exterior at 3 months showing loose connective tissue and microvasculature formation. (J) Low-magnification (top) and high-magnification (bottom) lumen view at 3 months. Top panel showed that the entirety of the hMPB lumen was free of thrombus formation; bottom panels showed absence of stenosis at sutured ends and no obvious changes at puncture site. One-way ANOVA followed by Tukey’s tests (C, i to iii, G, and H) and unpaired Student’s t test (C, iv) were performed; significance indicated by *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 7.. The histological analysis and infection detection of hMPB as AVG for 3 months in the sheep model.

(A) Schematic diagram of corresponding analysis of five sample sections of hMPB as AVG after implantation between jugular vein and carotid artery for 3 months. (B) H&E staining images of cross sections. (C) Quantitative analysis of neointima thickness of hMPB-A3m based on H&E images (n = 3). Statistical significance was calculated by one-way ANOVA with Tukey’s test. (D) The regeneration of ECs and contractile SMCs was observed by Co-IF staining with vWF and CNN. The white dashed lines represent the PS wall boundary (hMPB and hMPB-A3m) or the boundary between tunicae media and adventitia layers (sCA). Bottom-layer images represent the high magnification of the red area within top-layer images. (E) The thickness of CNN⁺ was calculated on the basis of low-magnification (top layer) co-IF staining images (n = 3). Statistical significance was calculated by one-way ANOVA with Tukey’s test. NS indicates not significant. (F) Gram-Twort staining showed no bacteria survival in hMPB, hMPB-A3m, and sCA. Bottom-layer images represent high magnification of the black area within top-layer images.