Recellularization of xenograft heart valves reduces the xenoreactive immune response in an in vivo rat model

Abstract

 

Open in new tabDownload slide

OBJECTIVES

Our aim was to address the role of autologous mesenchymal stem cell recellularization of xenogenic valves on the activation of the xenoreactive immune response in an in vivo rat model.

METHODS

Explanted aortic valve constructs from female Hartley guinea pigs were procured and decellularized, followed by recellularization with autologous Sprague-Dawley rat mesenchymal stem cells. Aortic valve xenografts were then implanted into the infrarenal aorta of recipient rats. Grafts were implanted as either autologous grafts, non-decellularized (NGP), decellularized and recellularized xenografts (RGP). Rats were euthanized after 7 and 21 days and exsanguinated and the grafts were explanted.

RESULTS

The NGP grafts demonstrated significant burden of granulocytes (14.3 cells/HPF) and CD3+ T cells (3.9 cells/HPF) compared to the autologous grafts (2.1 granulocytes/HPF and 0.72 CD3+ T cells/HPF) after 7 days. A lower absolute number of infiltrating granulocytes (NGP vs autologous, 6.4 vs 2.4 cells/HPF) and CD3+ T cells (NGP vs autologous, 2.8 vs 0.8 cells/HPF) was seen after 21 days. Equivalent granulocyte cell infiltration in the RGP grafts (2.4 cells/HPF) compared to the autologous grafts (2.1 cells/HPF) after 7 and 21 days (2.8 vs 2.4 cells/HPF) was observed. Equivalent CD3+ T-cell infiltration in the RGP grafts (0.63 cells/HPF) compared to the autologous grafts (0.72 cells/HPF) after 7 and 21 days (0.7 vs 0.8 cells/HPF) was observed. Immunoglobulin production was significantly greater in the NGP grafts compared to the autologous grafts at 7 (123.3 vs 52.7 mg/mL) and 21 days (93.3 vs 71.6 mg/mL), with a similar decreasing trend in absolute production. Equivalent immunoglobulin production was observed in the RGP grafts compared to the autologous grafts at 7 (40.8 vs 52.7 mg/mL) and 21 days (29.5 vs 71.6 mg/mL).

CONCLUSIONS

Autologous mesenchymal stem cell recellularization of xenogenic valves reduces the xenoreactive immune response in an in vivo rat model and may be an effective approach to decrease the progression of xenograft valve dysfunction.

INTRODUCTION

Valvular heart disease is a common diagnosis with an approximate prevalence of 2.5% of the total population having moderate-to-severe valvular heart disease, resulting in significant reduction in the quality of life for those patients affected [1, 2]. Xenograft tissue heart valves (XTHVs) made from bovine pericardium or porcine valve tissue are the most commonly implanted construct used for surgical valve replacements [3–5]. The development of structural valve deterioration of XTHVs is characterized by time-dependent onset, degeneration and/or haemodynamic dysfunction of the valve and in its most severe form requires reoperation [6]. Structural failure is strongly age dependent, making XTHVs suitable primarily for the elderly and less for children and young adults [7, 8]. Implantation of XTHVs is a form of xenotransplantation and is therefore subject to robust humoral and cellular immune system rejection, despite chemical crosslinking and anti-calcification treatments [9, 10]. Thus, we and others believe that the development of structural valve deterioration is a result of a chronic, immune-mediated rejection to the foreign XTHV tissue [9, 11, 12].

In an attempt to overcome the development of structural valve deterioration, several modifications to XTHVs have been suggested, including decellularization and recellularization of these scaffolds [13]. Decellularized xenografts have demonstrated inconsistent outcomes in both animal and clinical studies, suffering from tissue degradation, calcification and limited cellular regeneration [14, 15].

Subsequently, recellularization of decellularized xenografts has been proposed and investigated experimentally in both animal models and clinically in humans [16–18]. The use of autologous mesenchymal stem cell (MSC) to recellularize an acellular xenogenic scaffold is thought to skew the xenoreactive immune response towards an anti-inflammatory phenotype [19]. Numerous experiments have been performed to examine XTHV transplantation in a rat model [20, 21]; however, there is a paucity of data on what effect an autologous recellularized xenograft has on the immune response in the host.

The purpose of this study was to investigate the role of autologous MSC recellularization of xenogenic valves on the activation of the xenoreactive immune response in an in vivo rat model. We hypothesized that recellularization of decellularized guinea pig (DGP) aortic valve constructs with autologous rat MSCs would result in the attenuation of immune activation, with equivalent graft cellular infiltrate and immunoglobulin production compared to that produced by autologous rat aortic valve constructs.

METHODS

Experimental animals

Inbred female Hartley guinea pigs (age 2 weeks) and Sprague-Dawley rats (weight, 275–300 g) were purchased from Charles River (Quebec) and housed in the institutional animal care facility with food and water ad libitum for 1 week before experimentation in accordance with the guidelines of the Canadian Council of Animal Care.

Ethics statement

The animal use protocol used in this study was approved by the local Animal Care and Use Committee: Health Sciences 2 at the University of Alberta in Edmonton, Alberta, Canada (AUP2414).

Surgical technique

Aortic valve constructs were implanted into recipient animals according to a previously established infrarenal implantation model [21] (Supplementary Material, Fig. S1). Aortic valve constructs were removed from donor guinea pigs with a cuff of ventricular muscle and ∼5 mm of ascending aorta. The coronary arteries were suture ligated with 10–0 nylon suture. The aortic valve constructs were subsequently transplanted into the infrarenal abdominal aorta of recipient rats by using end-to-end anastomoses (10–0 nylon, Mani) during general anaesthesia with isoflurane.

Grafts were implanted as either xenogenic (non-decellularized guinea pig-to-rat) (NGP), decellularized guinea pig-to-rat (DGP) or recellularized guinea pig (RGP) grafts with autologous rat MSCs (n = 6–10/time/group). Implantation of autologous rat aortic valves (rat-to-rat) was used as a control for surgical inflammation (n = 6–8/time). Xenogenic and autologous grafts were harvested from donor animals and implanted without delay into recipient animals. Decellularized and recellularized grafts were treated as described below before implantation.

The recipient animals were euthanized 7 and 21 days after transplantation for graft harvest. The aortic valve construct was dissected free and immersed in 10% formalin for histology and immunology. At the time of euthanization, rats were systemically heparinized with the injection of 500 U of heparin directly in the inferior vena cava (IVC), exsanguinated and blood collected in 15-ml tubes. The blood was then centrifuged for 10 min at 1200 × g, the supernatant collected and frozen at −80°C until needed.

Mesenchymal stem cell isolation and culturing

Cadaveric autologous Sprague-Dawley rats were used to collect femoral bone marrow and subsequently isolate MSCs, as described previously [22]. The femoral bone marrow collected was directly plated onto T25 plastic cell culture flasks (Thermo Fisher Scientific, Waltham, MA, USA). The cells were incubated at 37°C with 5% CO₂ in RPMI 1640 (Thermo Fisher Scientific) culture media completed with 10% foetal bovine serum, 0.2% ascorbic acid and 0.2% primocin to only culture plastic adherent MSCs. The cells were washed with phosphate buffered saline (PBS) and the media changed at 1 day after initial plating. The media was then changed every 3 days afterward and monitored regularly until the flask had reached 80% confluence. A small sample of isolated and cultured MSCs was utilized to confirm surface phenotyping. MSC phenotype was confirmed by staining negatively for CD34 on post-recellularization immunohistochemistry by the methods described below.

Decellularization and recellularization

Aortic valve constructs were harvested from donor animals and underwent a decellularization and recellularization procedure described in full in Supplementary Material, Methods SA.

Confirmation of successful decellularization and recellularization was completed by 4′,6-diamidino-2-phenylindole (DAPI) staining. Tissue samples at each stage of processing including native, decellularized and recellularized samples were stained with DAPI to identify the presence of original cells, the absence of those cells after decellularization, and the presence of cells after recellularization. Staining for overall morphology (haematoxylin and eosin) and mesenchymal cells (anti-vimentin) was performed by the methods described in Supplementary Material, Methods SB.

Tissue analysis

Samples were examined with a light microscope, and images were captured with an Aperio CS2 digital pathology slide scanner (Leica Biosystems). Cell density was obtained from the sampling of the entire sample area from representative tissue cross-sections from each group. Cell counts were obtained from 5 high powered fields (HPFs), averaged per HPF and the mean cell count per HPF recorded from each animal. Cell counts were performed by 2 authors (J.J.H.K. and R.E.-A.) in a blinded fashion.

Rat IgG ELISAs

At the time of euthanization, rats were systemically heparinized with the injection of 500 U of heparin directly in the IVC, exsanguinated and blood collected in 15-ml tubes. The blood was then centrifuged for 10 min at 1200 × g, and the supernatant was collected and frozen at −80°C until the enzyme linked immunosorbent assay (ELISA) analysis, described in Supplementary Material, Methods SC.

Statistical analysis

Continuous data are expressed as mean$±$standard error of the mean. One-way analysis of variance (ANOVA) testing followed by Tukey’s multiple comparisons test when means between groups were equal, or Bonferroni’s multiple comparisons test when means between groups were not equal was used to compare individual groups. We performed the resource equation method to determine our sample size based on ensuring a degree of freedom of ANOVA >10. As we are comparing 4 groups throughout, this requires a minimum of 4 animals per group. All tests were considered significant with a P-value <0.05. All statistical analysis and figures were performed and generated using GraphPad Prism version 9.0.0.

RESULTS

Guinea pig aortic valve constructs can be successfully decellularized and recellularized with rat mesenchymal stem cells

Our decellularization and recellularization processes were assessed qualitatively using DAPI nuclear staining on non-decellularized guinea pig (NGP), DGP and RGP aortic valve constructs (Fig. 1A–C). As expected, the NGP contained a large number of cell nuclei along the leaflet edge (Fig. 1A), which were completely eliminated after decellularization (Fig. 1B). Following recellularization with rat MSCs, there was patchy evidence of nuclear material along the leaflet edge, qualitatively less compared to the NGP (Fig. 1C).

Gross histology was performed with H&E and vimentin staining. Autologous rat aortic valves are shown in Fig. 2A and B confirming the presence of cellular material and an intact extra cellular matrix (ECM). The presence of cellular material and an intact ECM structure were observed in the NGP tissue prior to decellularization (Fig. 2C and D). Following decellularization, there is a complete absence of cellular material; however, the ECM structure is grossly preserved (Fig. 2E and F). Recellularization with rat MSCs was performed and resulted in restoration of a confluent layer of MSCs lining the valvular tissue (Fig. 2G and H). This layer of cells stained negatively for CD34, consistent with an MSC phenotype (Supplementary Material, Fig. S2) [23, 24].

Taken together, these data suggest that complete decellularization followed by recellularization is achievable without any gross structural damage to the ECM.

Recellularization of xenograft heart valves reduces the cellular immune response similar to autologous tissue

To examine for the cellular immune response, the implanted aortic valve construct was harvested for histology at the time of animal euthanasia at the end points of 7 and 21 days. Figure 3A summarizes the granulocyte cell infiltration observed in the explanted grafts at 7 and 21 days. The number of infiltrating cells was compared between the 2 groups at each respective timepoint. We observed a significant increase in granulocyte cell infiltration in the xenogenic grafts compared to the autologous grafts after 7 days (14.3 $±$6.3 vs 2.1$±$0.6 cells/HPF, xenogenic vs autologous, P = 0.0002). Furthermore, we observed increased granulocyte cell infiltration in the xenogenic grafts compared to the decellularized grafts after 7 days (14.3$±$ 6.3 vs 7.9 $±$4.8 cells/HPF, xenogenic vs decellularized, P = 0.05). Finally, we observed equivalent granulocyte cell infiltration in the recellularized grafts compared to the autologous grafts after 7 days (2.4$±$0.8 vs 2.1$±$0.6 cells/HPF, recellularized vs autologous, P = 0.99).

After 21 days, we observed similar overall trends as compared to 7 days, however, with a reduction in the absolute number of infiltrating granulocytes. We observed a significant increase in granulocyte cell infiltration in the xenogenic grafts compared to the autologous grafts after 21 days (6.4$±$2.1 vs 2.4$±$0.65 cells/HPF, xenogenic vs autologous, P = 0.0007). Furthermore, we observed increased granulocyte cell infiltration in the xenogenic grafts compared to the decellularized grafts after 21 days (6.4$±$2.1 vs 3.9$±$1.7 cells/HPF, xenogenic vs decellularized, P = 0.05). Finally, we observed equivalent granulocyte cell infiltration in the recellularized grafts compared to the autologous grafts after 21 days (2.8$±$ 0.95 vs 2.4$±$ 0.65 cells/HPF, recellularized vs autologous, P = 0.99). Raw granulocyte cell counts for each sample analysed at both 7 and 21 days are displayed in Supplementary Material, Table S1.

Figure 3B summarizes the CD3+ T-cell infiltration observed in the explanted grafts at 7 and 21 days. We observed a significant increase in CD3+ T-cell infiltration in the xenogenic grafts compared to the autologous grafts after 7 days (3.9$±$ 1.7 vs 0.7$±$0.2 cells/HPF, xenogenic vs autologous, P = 0.0008). Furthermore, we observed equivalent CD3+ T-cell infiltration in the xenogenic grafts compared to the decellularized grafts after 7 days (3.9$±$1.7 vs 2.7$±$1.6 cells/HPF, xenogenic vs decellularized, P = 0.33). Finally, we observed equivalent CD3+ T-cell infiltration in the recellularized grafts compared to the autologous grafts after 7 days (0.6$±$0.2 vs 0.7$±$0.2 cells/HPF, recellularized vs autologous, P = 0.99). After 21 days, we observed similar overall trends as compared to 7 days, however, with a reduction in absolute number of CD3+ T cells. We observed a significant increase in CD3+ T-cell infiltration in the xenogenic grafts compared to the autologous grafts after 21 days (2.8$±$1.2 vs 0.8$±$0.22 cells/HPF, xenogenic vs autologous, P = 0.03). Furthermore, we observed a similar CD3+ T-cell infiltration in the xenogenic grafts compared to the decellularized grafts after 21 days (2.8$±$1.2 vs 3.3$±$1.9 cells/HPF, xenogenic vs decellularized, P = 0.89). Finally, we observed equivalent CD3+ T-cell infiltration in the recellularized grafts compared to the autologous grafts after 21 days (0.7$±$0.24 vs 0.8$±$0.22 cells/HPF, recellularized vs autologous, P = 0.94). Raw CD3+ T-cell counts for each sample analysed at both 7 and 21 days are displayed in Supplementary Material, Table S2.

Figure 4 illustrates representative gross histology for granulocytes with H&E staining and T cells with anti-CD3 staining in autologous (Fig. 4A and B), NGP (Fig. 4C and D), DGP (Fig. 4E and F) and RGP (Fig. 4G and H) explanted tissues. There is a significant burden of granulocytes in the NGP and DGP groups, with extensive cellularity within the media and adventitia of the vessel wall and valvular thickening (red arrow in Fig. 4C). Similarly, there is a greater burden of T cells in the NGP and DGP groups as shown in Fig. 4D and F. There is normal laminar architecture and no valve thickening demonstrable in the autologous and RGP groups.

Taken together, these data suggest that autologous MSC recellularization of xenograft heart valves reduces the cellular immune response to amounts similar to that produced in response to autologous tissue.

Recellularization of xenograft heart valves reduces the humoral immune response similar to autologous tissue

To examine for the humoral immune response, the recipient rats were exsanguinated at the time of animal euthanasia at both end points of 7 and 21 days and serum frozen for use in quantitative IgG ELISA analysis. Figure 5A summarizes the total serum immunoglobulin results observed after 7 days of implantation. We observed a significant increase in total serum immunoglobulin production in the xenogenic grafts compared to the autologous grafts after 7 days (123.3$±$17.0 vs 52.7$±$5.6 mg/ml, xenogenic vs autologous, P = 0.03). Furthermore, we observed equivalent total serum immunoglobulin production in the xenogenic grafts compared to the decellularized grafts after 7 days (123.3$±$17.0 vs 96.4$±$30.1 mg/ml, xenogenic vs decellularized, P = 0.93). Finally, we observed equivalent total serum immunoglobulin production in the recellularized grafts compared to the autologous grafts after 7 days (40.8$±$7.6 vs 52.7$±$5.6 mg/ml, recellularized vs autologous, P = 0.99). Fig. 5B summarizes the total serum immunoglobulin results observed after 21 days of implantation, revealing similar overall trends as seen after 7 days. We observed a trend towards an increase in total serum immunoglobulin production in the xenogenic grafts compared to the autologous grafts after 21 days (93.3$±$3.2 vs 71.6$±$4.8 mg/ml, xenogenic vs autologous, P = 0.07). Furthermore, we observed equivalent total serum immunoglobulin production in the xenogenic grafts compared to the decellularized grafts after 21 days (93.3$±$3.2 vs 90.5$±$6.8 mg/ml, xenogenic vs decellularized, P = 0.99). Finally, we observed significantly reduced total serum immunoglobulin production in the recellularized grafts compared to the autologous grafts after 21 days (29.5$±$15.0 vs 71.6$±$4.8 mg/ml, recellularized vs autologous, P = 0.001). Raw serum immunoglobulin concentration for each sample analysed at both 7 and 21 days is displayed in Supplementary Material, Table S3.

Taken together, these data suggest that autologous MSC recellularization of xenograft heart valves reduces the humoral immune response to amounts similar to that produced in response to autologous tissue.

Graft patency and recipient survival

To facilitate the clinical translation of recellularized xenograft heart valves, the assessment of graft patency and overall survival is required; thus, we assessed these outcomes in our study. Several prior studies have established that vascular xenografts fail due to aneurysm formation and resulting thrombosis [25, 26]. Interestingly, increased rates of graft thrombosis and subsequent mortality have been observed in a similar rat in vivo model [27].

We examined the explanted aortic valve construct at the time of animal euthanasia at 21 days for luminal patency (Fig. 6). All animals survived to 7 days. Those animals that did not survive to 21 days were excluded. Overall survival to the 21-day end point was lower in the decellularized xenograft group (67%; 4/6), compared to non-decellularized (100%; 6/6) and recellularized grafts (100%; 6/6). Survival in the autologous group was similar to that observed in both the non-decellularized and recellularized groups (83%; 5/6). Similarly, grafts in the decellularized group were more likely to have completely thrombosed (50%; 2/4), compared to non-decellularized (33%; 2/6) and recellularized grafts (0%; 0/6). Thrombosis in the autologous group was similar to that observed in the recellularized group (0%; 0/5).

DISCUSSION

We investigate the role of autologous MSC recellularization of xenogenic valves on the activation of the xenoreactive immune response in an in vivo rat model. Our results demonstrated that guinea pig aortic valve constructs can be successfully decellularized and recellularized with rat MSCs. Next, we confirmed that recellularization of xenograft heart valves reduces the cellular and humoral immune response similar to autologous tissue, with a reduction in graft cellular infiltrate and total serum immunoglobulin production. Finally, we determined that recellularization of xenograft heart valves improves graft patency and recipient survival compared to non-decellularized and decellularized xenograft. These data highlight the immunomodulatory effects of autologous MSC-recellularized tissue on the recipient immune response, thus preventing recipient immune recognition of these aortic valve constructs.

Several prior studies have been performed using a similar aortic valve implantation rat model that have established its viability and usefulness in assessing the immune responses to transplanted tissues [20, 21, 28, 29]. Legare et al. [29] performed allograft aortic valve implantation using an infrarenal aortic model, demonstrating that the allografts underwent T-cell-mediated immune rejection leading to structural failure. The onset of structural failure was rapid, beginning within 7 days and progressively increasing over time. In another study by Meyer et al. [21], decellularization of rat allograft aortic valves was performed followed by infrarenal aortic implantation. The investigators found that allografts induced a significant increase in CD3+ T-cell infiltrate and a significant production of immunoglobulin. This robust immune response to the allografts was significantly reduced by decellularization. Building on this, Manji et al. [20] analysed the immune response to xenografts and decellularized xenografts using a guinea pig-to-rat aortic valve model. The xenografts elicited a significant increase in immune cellular infiltrate and immunoglobulin production, compared to autologous valves, at 7 and 20 days and this immune response could be reduced with decellularization. These prior studies served to establish the feasibility of a xenograft transplant model with in vivo implantation in a rat and validate methods of assessing the immune responses to these tissues.

We sought to build on the previously performed experiments by performing autologous MSC recellularization of XTHVs and studying the impact on the immune response in the recipient animal. We focused on T-cell and granulocyte infiltration since these cells have been consistently demonstrated to correlate to the development of valve structural failure [9, 11]. The present analysis also focused on the humoral immune response, by quantifying total serum immunoglobulin production. Thus, the current study provides a more balanced assessment of the immune response to XTHVs than has been previously reported. Finally, in an attempt to provide more clinically translatable data, we measured overall survival rates and graft patency. The clinical utility of decellularized xenografts is limited, primarily due to severe immune rejection and thrombogenesis [14, 21, 30]. We have demonstrated that autologous MSC recellularization leads to not only a reduction in cellular and humoral immune response similar to autologous tissue but also an improvement in graft patency. This is an important finding since it reinforces the concept that decellularized xenografts have limited clinical utility and that recellularization may provide an improvement on this strategy.

Our findings suggest that recellularizing a decellularized xenogenic matrix with autologous MSCs results in the same lack of pro-inflammatory cytokine production as that seen by autologous tissue. It has been well established that MSCs mediate their immunomodulatory effects by interacting with cells from both innate and adaptive immunity, skewing the xenoreactive immune response towards an anti-inflammatory phenotype [19]. The reduction in immune activation seen in the recellularized tissue may be attributed to a combination of factors. First, as outlined above, MSCs possess inherent immunomodulatory capabilities that may be dampening the recipient immune response. Second, the mechanical effect of ‘masking’ the antigenic ECM may also be preventing the recipient immune system from recognizing this tissue. The mechanism of immune rejection in our model is likely a combination of hyper-acute and acute rejection. This is a result of the fact that guinea pig to rat transplantation is considered discordant, meaning that pre-formed effectors are present in the serum of the recipient at the time of engraftment. Our results support this notion with the development of significant immunoglobulin production within 7 days of implantation. Prior studies using animal models to assess the immune response to XTHVs have used concordant models, hindering the ability to apply these findings in humans where XTHV implantation is most commonly discordant [16, 17].

Limitations

The model used in this study is non-functional, meaning that the transplanted aortic valve construct does not open and close with systole and diastole. The non-functional nature of our model limits the relevance of our model to clinical valve replacement. Our results should therefore be interpreted with caution and a functional model of xenograft valve replacement is necessary to allow for the broader application of our findings. This results in the thrombus formation in the sinus of Valsalva, but not within the entire graft lumen. It is possible that non-specific foreign-body inflammation may be occurring in our model in addition to a specifically generated immune response. However, the distribution of immune cellular infiltrate in a similar pattern to that seen in live cellular xenograft rejection coupled with the production of immunoglobulin supports a specific immune response. Further studies are needed to define the long-term effects of this process, both in vitro and in vivo and to elucidate the mechanism of immunomodulation by the autologous MSCs.

CONCLUSION

Our results suggest that attenuation of immune activation in response to an autologous MSC RGP aortic valve construct is likely a result of the recellularized tissue not being recognized by the recipient immune system, suggesting that the foreign xenogenic aortic valve is being successfully masked from immune recognition. Importantly, we determined that recellularization of xenograft heart valves is associated with improved graft patency and recipient survival compared to non-decellularized and decellularized xenograft. Preventing this initial immune activation with autologous MSC recellularization may be an effective approach to decrease the recipient immune response and improve graft patency, potentially leading to improved long-term durability.

SUPPLEMENTARY MATERIAL

Supplementary material is available at EJCTS online.

Funding

This study was supported by the University Hospital Foundation, Edmonton, Alberta, and Edmonton Civic Employees Charitable Assistance Fund, Edmonton, Alberta.

Conflict of interest: none declared.

Data Availability Statement

All relevant data are within the manuscript and its Supporting Information files.

Author contributions

Sabin J. Bozso: Data curation; Formal analysis; Investigation; Methodology; Writing—original draft; Writing—review & editing. Jimmy J.H. Kang: Formal analysis; Investigation. Ryaan EL-Andari: Formal analysis; Methodology. Nicholas Fialka: Data curation; Methodology. Lin Fu Zhu: Investigation; Methodology; Project administration. Steven R. Meyer: Methodology. Darren H. Freed: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Resources; Supervision; Writing—review & editing. Jayan Nagendran: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Resources; Supervision; Validation. Jeevan Nagendran: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Supervision.

Reviewer information

European Journal of Cardio-Thoracic Surgery thanks David Schibilsky and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.

REFERENCES

ABBREVIATIONS

 
• DGP

  Decellularized guinea pig

 
• HPF

  High powered field

 
• MSC

  Mesenchymal stem cell

 
• NGP

  Non-decellularized guinea pig

 
• RGP

  Recellularized guinea pig

 
• XTHV

  Xenograft tissue heart valve