Modulating the platelet-mediated innate foreign body response to affect in situ vascular tissue engineering outcomes

Abstract

The success of implanted tissue-engineered vascular grafts (TEVGs) relies on the coordinated inflammation and wound healing processes that simultaneously degrade the scaffold and guide the formation of a neovessel. Dysregulated responses can lead to aberrant remodeling (e.g., stenosis), impacting the long-term outcome and functionality of the TEVG. We developed a TEVG that, despite demonstrating growth capacity in the clinic, exhibited an unexpectedly high incidence of stenosis, or narrowing of the graft lumen. This study identified platelet-mediated immune signaling via the lysosomal trafficking regulator (Lyst) as a key driver of stenosis. Lyst mutations significantly impaired platelet dense granule exocytosis yet preserved alpha granule secretion and adhesion to the biomaterial. Uncontrolled platelet aggregation, potentiated by dense granule signaling, results in the formation of a mural thrombus that remodels into occlusive neotissue. Importantly, inhibiting sustained platelet aggregation using the P2Y12 antagonist, prasugrel, is a successful strategy for optimizing neotissue formation and improving overall TEVG performance.

Fig. 1. Schematic overview of the foreign body response and TEVG outcomes.

A Schematic overview of the foreign body response over time, highlighting the involvement of various immune cells and the progression of neotissue formation on the TEVG lumen in the mouse model. The generic foreign body response is shown over 14 days, highlighting key cell types involved in inflammation and wound healing. B Progression of neotissue formation in TEVG. Two outcomes are demonstrated where (top) densely compacted platelets forming provisional matrix that mediate tissue regeneration without significant narrowing, and (bottom) excessive platelet aggregation allows for overgrowth of provisional matrix that is remodeled into stenotic or occlusive neotissue.

Fig. 2. Disrupting lysosomal trafficking regulator (Lyst) function attenuated the development of stenosis.

A Histological representation of short-term (2 weeks) and long-term (2 years) graft morphology. Both wild-type (C57BL/6J) and global Lyst mutant (Lyst Exon 52 deletion) grafts are compared to the native vein, demonstrating that TEVGs remodel into neovessels that resemble the native vein. Carstair’s stain: collagen – bright blue; platelets – gray; fibrin – bright red; muscle and cytoplasm – red; red blood cells – orange/yellow. B Micro-CT anatomical views of TEVGs. Enlarged images show the (top left) axial view (blue circle indicates graft cross-section), (bottom left) coronal view, and (top right) sagittal view. Colors indicate individual CT scan slices of 40 µm thickness. C 3D reconstructions of Micro-CT scans showing the IVC (blue) and aorta (red). From left to right, a WT control with a surgically naïve IVC, a WT mouse implanted with an interposition TEVG, and a Lyst mutant mouse implanted with a TEVG. Tortuous collaterals (orange arrow) are observed alongside the occluded TEVG (yellow bracket) in the WT surgical, while Lyst mutant TEVG (yellow bracket) remains patent. D Comparison of minimum luminal diameter (mm) between WT (n = 24, 25% group patency rate) and Global Lyst Mutant (n = 29, 92.3% group patency rate) (****p < 0.0001 | Fisher’s Exact Test) based on Micro-CT analysis.

Fig. 3. Reciprocal bone marrow transplant confirms that stenosis occurs through Lyst-mediated immune signaling.

A Schematic representation of the bone marrow transplant experimental timeline, showing key time points for bone marrow transplant (BMT), tissue-engineered vascular graft (TEVG) implantation, and TEVG explantation. TEVGs were implanted 5 weeks after bone marrow transplant. TEVGs were explanted 2 weeks later, 7 weeks after the bone marrow transplant took place. B Micro-CT comparison of graft patency in WT (left) mice transplanted with Lyst mutant bone marrow and Lyst mutant (right) mice transplanted with WT bone marrow. Axial, sagittal, and coronal views are included with the TEVG indicated in yellow. C Comparison of minimum luminal diameter demonstrates a significant increase in the WT (Lyst mutant BM) and significant decrease in the global Lyst mutant (WT BM) group (****p < 0.0001 | Fisher’s Exact Test). Dotted line indicates the starting diameter of the implanted graft. D Representative histological cross-sections of Carstair’s stained TEVGs from WT (left) mice transplanted with Lystmutant bone marrow and global Lyst mutant (right) mice transplanted with WT bone marrow are shown.

Fig. 4. Evaluating the effect of cell-specific Lyst mutations on the development of TEVG stenosis.

A The breeding schematic used to create each of the cell-specific Lyst mutant murine models where Pf4-Cre was used to drive “megakaryocyte/platelet specific” Lyst Ex52 deletion, S100A8-Cre was used to drive “neutrophil specific” Lyst Ex52 deletion, and Lyz2-Cre was used to drive “macrophage specific” Lyst Ex52 deletion. Created in https://BioRender.com. B Comparison of minimum luminal diameter between WT (25% group patency rate), Lyst Ex52 Floxed (32.1% group patency rate), neutrophil (43.5% group patency rate), and macrophage specific Lyst mutant (34.5% group patency rate) strains. Measurements were obtained using Micro-CT scans. There is no significant difference in the minimum luminal diameter between any of these groups. Fisher’s Exact comparisons were made to compare each strain to WT, and the “ns” indications have corresponding p values of (i) Flox p = 0.760, (ii) Neutrophil p = 0.227, and (iii) Macrophage p = 0.554. The dotted line indicates the starting diameter of the implanted graft.

Fig. 5. A platelet specific Lyst mutation disrupts dense granule exocytosis and improves TEVG patency outcomes.

A Comparing minimum luminal diameter (left) and volume (right) of TEVGs in WT, global Lyst mutant, and platelet specific Lyst mutant mice. Measurements were obtained using reconstructed Micro-CT scans. Significant increases in luminal diameter and volume were observed in the platelet specific Lyst mutants (96.7% group patency rate) compared to WT (25% group patency rate) (**** p < 0.0001 | Fisher’s Exact Test). Platelet specific Lyst mutants were not significantly different than global Lyst mutants (***p < 0.0001 | Fisher’s Exact Test). The dotted line indicates the starting diameter or volume of the implanted graft. B Comparing platelet count per cubic millimeter of whole blood in the WT, global Lyst mutant, and platelet specific LYST mutant mice. C Flow cytometry analysis highlighting p-selectin and integrin αIIbβ3 expression before and after collagen stimulation. Both Lyst Ex52 Floxed and global Lyst mutant platelets had p-selectin^highαIIbβ3^high populations observed after collagen stimulation (red box). Platelet aggregation (Max Aggregation - Light Transmission %) and dense granule secretion (luminescence AUC) of floxed, global Lyst mutant, and platelet Lyst mutant platelets in response to (D) thrombin, (E) ADP, and F collagen. (Left) As measured using light transmission, Lyst mutant platelets displayed normal aggregation in response to (d) thrombin (p = 0.88 | One-way ANOVA) and E ADP (p = 0.28 | One-way ANOVA) but impaired aggregation in response to (f) collagen (p < 0.0001 | One-way ANOVA). (Right) As measured using luminescence, Lyst mutant platelets displayed impaired dense granule release in response to D thrombin (p < 0.0001 | One-way ANOVA) and F collagen (p < 0.0001 | One-way ANOVA). No platelets released dense granules in response to (E) ADP alone. G Histological and scanning electron microscopy (SEM) analysis of TEVGs explanted at 3 days post-implantation from WT and global Lyst mutant mice. The images show differences in luminal narrowing and platelet deposition between the two groups, highlighting the impact of Lyst mutations on early graft outcomes. Carstair’s staining was used for histological analysis, while SEM images provide detailed views of the graft’s surface.

Fig. 6. Disrupting P2Y12 using a knockout model and prasugrel fully inhibit the development of stenosis.

A Comparing minimum luminal diameter (left) and volume (right) between WT, P2Y12 KO, and prasugrel-treated mice. Both P2Y12 KO and prasugrel-treated mice had increased luminal diameters and volumes compared to WT mice (****p < 0.0001 | Fisher’s Exact Test). “ns” indicates no statistically significant difference. Top dotted line indicates the starting diameter or volume of the implanted graft. B Representative immunohistochemical staining for CD31 (endothelial cells) from WT and prasugrel-treated mice (n = 14) compared to the native vein. Scaffold and lumen are indicated. C Percentage of the TEVG lumen that was endothelialized in WT and prasugrel-treated WT mice. There was no difference between the WT and prasugrel-treated mice (p = 0.56 | Mann Whitney U test). D–F Platelet aggregation (Aggregation - Light Transmission AUC) and dense granule secretion (luminescence AUC) of WT and P2Y12 KO platelets in response to C thrombin, (D) ADP, and E collagen. (Left) As measured using light transmission, P2Y12 KO displayed decreased aggregation in response to C thrombin (**p = 0.008 | Welch’s t-test), (D) ADP (**p = 0.002 | Welch’s t-test), and E collagen (***p = 0.0002 | unpaired t-test). (Right) As measured using luminescence, P2Y12 KO platelets displayed unchanged dense granule release compared to WT in response to C thrombin (p = 0.16 | Welch’s t-test) and E collagen (p = 0.07 | unpaired t-test). No platelets released dense granules in response to D ADP alone. G Carstair’s stained representative section of a TEVG implanted in a P2Y12 KO mouse and explanted at 3 days. SEM images allow for visualization of platelet deposition and fibrin at the surface of the scaffold. The lumen and scaffold are labeled.

Fig. 7. ADP-mediated platelet signaling forms the provisional matrix that guides neotissue formation, and in excess, the development of stenosis.

A schematic of the proposed mechanism for early provisional matrix formation, which influences graft patency. Lyst-mediated release of ADP from dense granules contributes to thrombus/provisional matrix stability at 3 days.