Thromboembolic events associated with currently available cardiovascular implants still remain a major problem in daily clinical routine. Thus, long-term anticoagulation or anti-platelet therapy is often required, which however permanently increases the risk for bleeding and decreases the overall patient’s quality of life. Therefore, a good hemocompatibility profile represents a key requirement in the development of novel cardiovascular implants (van Oeveren, 2013; Carnicelli et al., 2016).

Next-generation tissue-engineered (TE) replacements with regenerative capacities (Dahl et al., 2011; Weber et al., 2013; Driessen-Mol et al., 2014; Syedain et al., 2014; Syedain et al., 2015; Reimer et al., 2017; Syedain et al., 2017; Emmert et al., 2018; Lintas et al., 2018; Motta et al., 2018; Biermann et al., 2019; Boethig et al., 2019; Kirkton et al., 2019; Motta et al., 2019; Gerdisch et al., 2020; Fioretta et al., 2021; Syedain et al., 2021) have demonstrated their strong potential in numerous preclinical studies and first clinical pilot trials (Fioretta et al., 2021), and may therefore represent an ideal candidate to overcome the limitations of current prostheses. As one promising TE approach, we have recently introduced a biomimetic acellular tissue-engineered matrix (TEM), that is, manufactured from a polymer composite and an in vitro grown extracellular matrix (ECM), which can be engineered from different (human) cell sources (Emmert et al., 2018; Fioretta et al., 2021).

Importantly, prior to implantation, the TEM is decellularized leaving a cell-free construct which is thereby conceptually applicable to every patient (Dijkman et al., 2012). Preclinical evaluations of such decellularized TEMs in the context of heart valves and blood vessels have demonstrated their strong remodeling, and recellularization capacity, including continuous endothelialization over time (Weber et al., 2013; Driessen-Mol et al., 2014; Emmert et al., 2018; Lintas et al., 2018; Motta et al., 2018; Motta et al., 2019; Motta et al., 2020; Fioretta et al., 2021). As part of the remodeling process, the blood coagulation cascade is initiated immediately upon implantation of the TEM (Lynn et al., 2004; van Loon et al., 2013), where host cells are able to rapidly infiltrate to finally allow for the gradual formation of a functional endothelium, thereby ensuring a sufficient and safe performance of the TEM. However, the exact mechanisms of the endothelialization process (e.g., blood-borne and/or migration from the adjacent tissue) in such TEMs remain to be elucidated, especially as their endothelialization capacity may also be influenced by the patient’s individual regenerative potential. This aspect may be particularly relevant when using TEMs in the high-pressure circulation of the left heart (e.g., transcatheter aortic valve implantation, TAVI) which is known to carry a high risk for thromboembolic events.

Hence, from a translational perspective, the concept of TEM in vitro pre-endothelialization prior to implantation may be warranted for certain indications or specific patient populations with a high risk for thromboembolic events and may represent a good strategy to further enhance their overall hemocompatibility. Therefore, we here tested the hypothesis whether in vitro pre-endothelialization of human cell-based TEMs (hTEMs) using an endothelial progenitor cell (EPC)-based approach is feasible. We particularly focused on hTEM in vitro pre-endothelialization in the setting of transcatheter heart valves as these implants are prone to microstructural damage due to the required crimping maneuvers during delivery (Kiefer et al., 2011; Amahzoune et al., 2013; Alavi et al., 2014). Our group has previously developed several (h)TEM-based transcatheter valves (Weber et al., 2011; Weber et al., 2013; Driessen-Mol et al., 2014; Emmert et al., 2018; Lintas et al., 2018; Motta et al., 2018; Motta et al., 2019; Fioretta et al., 2021), and therefore we investigated if the integrity of an in vitro pre-endothelialized hTEM can also be preserved after such crimping procedures.

To test these hypotheses, we first evaluated the general in vitro pre-endothelialization capacity of EPC-seeded hTEMs in a high-throughput patch culture experiment. In a next step, we assessed the regenerative capacity of human blood-derived EPCs or human-derived umbilical vein endothelial cells (HUVECs) seeded hTEMs in a scratch assay experiment (Liang et al., 2007) under static and dynamic orbital conditions, which allowed us to mimic potential endothelial damages that may occur due to crimping of transcatheter valves. Thereafter, we generated hTEM-based transcatheter tissue engineered sinus valves (TESVs) which were then in vitro pre-endothelialized before they underwent crimping procedures to assess whether the endothelium remains stable. Finally, to enhance their translational relevance, the hTEM pre-endothelialization capacity was also assessed under xenogeneic-free culture conditions (i.e., supplemented with human platelet lysate (hPL)).

The hemocompatibility profile of current cardiovascular implants remains a major clinical challenge. To this end, ongoing research studies (e.g., GALILEO and ATLANTIS trials) are evaluating the impact of thrombosis on valve hemodynamics, valve durability, and its progression (Chakravarty et al., 2020). To prevent thromboembolic events related to the implantation of such substitutes, the routine use of anticoagulation or anti-platelet therapies is still required in many patients, while on the other hand such therapies come at a price of a permanently increased risk of bleeding complications (Leiria et al., 2011). In addition, especially in newborns and children, the use of such therapies is limited due to the risk of severe adverse events (e.g., heparin-induced thrombocytopenia) (Dabbous et al., 2014; Newall et al., 2018). Hence, when considering the significant patient population born with congenital cardiovascular malformations requiring treatment with currently-available cardiovascular implants (DiBardino and Jacobs, 2014; Jacobs et al., 2014), this still represents a significant and unresolved clinical problem around the globe. In addition, due to the embryonic toxicity of anticoagulation therapy (e.g., warfarin), also young pregnant women are not eligible for such therapies; and thus cannot be treated with a mechanical heart valve prosthesis (Alshawabkeh et al., 2016; Ayad et al., 2016).

hTEMs may represent a promising option to overcome the limitations of current cardiovascular prostheses. In fact, in numerous animal studies we have shown that (h)TEMs are mechanically mature enough to be safely applied in the cardiovascular system but, on the other hand, are still premature enough to allow for significant recellularization allowing the host cells to invade into their specific niche thereby continuously remodeling the hTEM into native-like tissue. However, despite their strong in vivo remodeling potential, the hemocompatibility profile of hTEMs could be further enhanced by using autologous EPC-based in vitro pre-endothelialization approaches, which may be particularly beneficial in vulnerable patient cohorts (e.g., patients with diabetes, children, athletes, pregnant women Weber et al., 2012).

A particular focus in our study was to test the feasibility of pre-endothelialize acellular hTEMs in the context of transcatheter heart valve applications as these are particularly exposed to surface damages due to the required crimping maneuver during valve delivery. Our study demonstrates that hTEMs may provide an ideal microstructural basis for the effective and stable in vitro formation, retention, and potential repair (in case of damage) of a confluent and functional human EPC-derived endothelium under both, static, and dynamic orbital conditions (e.g., pulmonary arterial shear forces). In our dedicated in vitro system, the application of shear forces did not have a negative impact on the protein expression in EPCs (i.e., cell-cell contact proteins, secreted factors, and matrix components). These findings suggest that hTEMs are able to sustain a functional endothelium under dynamic conditions. As suggested by previous studies (Dardik et al., 2005), laminar shear stress is believed to have an atheroprotective effect on ECs, whereas orbital (oscillatory) shear stress often correlates with areas of atherosclerosis. In vitro models of “activated” ECs are usually replicated by means of low magnitude oscillatory shear stress platforms, such as an orbital shaker, as the shear stress is not homogeneously applied throughout the culture well. We demonstrated that the hTEM functionally sustains the growth of EPCs when exposed to low shear magnitudes in an orbital shaker. Additionally, recent computational modeling analyses performed on TEMs-based TEHVs demonstrated that increasing pressure and shear stress conditions (i.e., aortic shear forces) corresponded to an improved preservation of valve functionality (Loerakker et al., 2016; Emmert et al., 2018). We could therefore hypothesize, that at higher magnitudes EPCs should positively respond and do not show major signs of activation. Our data further highlight the migration properties of EPCs on hTEMs in an in vitro simulated crimping procedure (i.e., scratch assay). In fact, we hypothesize that the particular composition of our hTEM may promote the adhesion properties of EPCs by allowing the expression of cell-matrix contact proteins (e.g., Vinculin), which in turn may enhance cell retention in situ. Moreover, vWF is a well-known mediator for inflammation, which in case of pre-endothelialized hTEMs, may be helpful to further stimulate native host cell recruitment in situ.

The principal feasibility to pre-endothelialize matrices such as polymers and xenogeneic or allogenic substitutes has been successfully demonstrated in several preclinical and clinical studies (Zilla et al., 1994; Shinoka et al., 1995; Lehner et al., 1997; Cebotari et al., 2006; Hibino et al., 2010; Roh et al., 2010; Quint et al., 2011; Tudorache et al., 2013; Meier et al., 2014; Melchiorri et al., 2015; Theodoridis et al., 2015; Ma et al., 2017; Drews et al., 2020; Schwarz et al., 2021). In comparison to non-endothelialized counterparts, they showed less neointima and thrombus formation. However, on the other hand, human ECs are also known to have limited capabilities to grow and expand across anastomotic regions (Berger et al., 1972; Zilla et al., 2007), thereby limiting the overall in vivo endothelialization capacity of tissue-engineered constructs once implanted.

hTEMs and other, conceptually comparable acellular technologies have demonstrated strong over time endothelialization following implantation in recent preclinical studies (Syedain et al., 2017; Biermann et al., 2019)^,. However, despite this potential, evidence in the human setting remains preliminary (Shirzad et al., 2017) and a systematical investigation in larger clinical studies are needed. Therefore, to date, precautionary anticoagulation or anti-platelet therapy would be still advisable (Lawson et al., 2016; Tudorache et al., 2016; Bobylev et al., 2019; Boethig et al., 2019; Kirkton et al., 2019).

Recently, hPL has been validated and suggested as a clinically relevant alternative to the FBS used for standard cell culture procedures, especially in the setting of wound healing (Bieback, 2013; Martínez et al., 2015; Shirzad et al., 2017). In fact, recent studies reported reduced inflammation, improved bone regeneration, and improved hard and soft tissue wound healing when hPL was added as medium supplement (Chiara Barsotti et al., 2013; Ramos-Torrecillas et al., 2014), further highlighting the potential of hPL for clinical application.

In our study, EPCs pre-endothelialization of hTEMs cultured with hPL demonstrated the capacity to form microvessel-like structures, suggesting the ability of EPCs to rearrange and reorganize to induce angiogenesis (Fortunato et al., 2016). This finding might be attributable to specific factors that are present in hPL such as vascular endothelial growth factor (VEGF, a sub-family of platelet-derived growth factors), which is able to induce the formation of blood vessels (Shanskii et al., 2013; Ramos-Torrecillas et al., 2014). Interestingly, the results presented in this study suggest an anti-thrombotic effect of EPC seeded hTEMs. Here, ECFCs gene expression showed an upregulation of the genes involved in vascular tone control (eNOS), in angiogenesis (VEGF), and hemostasis and EPCs activation (vWF).

Finally, we were able to demonstrate that the concept of EPCs pre-endothelialized hTEMs is also feasible in the setting of a transcatheter heart valve geometry, and in particular, that such pre-endothelialized hTEM-based transcatheter valves can safely undergo crimping maneuvers which are mandatory for transcatheter valve delivery.

This is an important finding, as to date, only surgical tissue-engineered heart valves have been investigated as possible starter matrices for pre-endothelialization studies (Zilla et al., 2008; Stassen et al., 2017). Moreover, due to the increasing clinical relevance of transcatheter valve replacements, their forthcoming suitability for low-risk and young patients (Arora et al., 2017; Kenny and Hijazi, 2017; Fioretta et al., 2018), and our recent advances in the development of transcatheter tissue-engineered heart valves, our results are of high translational relevance, as it is of key importance to understand whether such prostheses could cope with in vitro pre-endothelialization procedures.

Crimping maneuvers are indeed very frequent “folding techniques” in order to enable transcatheter delivery of cardiovascular prostheses such as stented heart valves, vascular conduits or occluders (Jung and Choi, 2018; Mack et al., 2019). Crimping of current cardiovascular substitutes down to diameters of 14–18 Fr has been demonstrated to cause microstructural damages which could potentially result in an accelerated failure of the prosthesis (Kiefer et al., 2011; Amahzoune et al., 2013; Alavi et al., 2014). Notably, the non-regenerative nature of the fixed xenogeneic bioprostheses may not enable the complete recovery of the initial surface structure, thereby favoring platelets adhesion and long-term calcifications. In our study, we have successfully demonstrated that 1) pre-endothelialized hTEMs can be adjusted to a TESV geometry; and 2) TESVs sustained human EPCs adhesion following the crimping maneuver. Importantly, due to the strong regenerative potential of EPCs together with their migration capacity once seeded onto our hTEMs, we may hypothesize that potential damages that may have occurred during crimping may be repaired in vivo following implantation.

This study focused on the in vitro evaluation of EPC behavior when seeded onto hTEMs, by means of simplified in vitro techniques (e.g., scratch assay). In vivo model system showing the direct functionality of pre-endothelialized hTEMs once in contact with animal/human blood was not assessed and was beyond the scope of the study. Future studies should investigate how hTEMs pre-endothelialized with EPCs will impact on the mechanical and functional cardiovascular environment, also taking into consideration blood flow hemodynamics and immune cell response. In this context, in-depth understanding of EPCs anti-thrombotic molecular pathways over time could be analyzed to better predict EPCs behavior in vivo. In addition, taking into account the impact of the cell source (i.e., allogenic or autologous) and the existing inter-patient variability (e.g., age, health status, environmental factors) to better understand the dynamic of EPCs behavior in vivo and improve their translation to clinical application, an increased patient number could be implemented, also including EPCs isolated from diseased patients.

This study demonstrates the feasibility to develop an EPCs pre-endothelialized hTEM-based construct as starting matrix for cardiovascular applications. EPCs were successfully retained on hTEMs and showed over time migration properties as well as expression of typical endothelial cells markers such as VE-Cad, CD31, and vWF, under both static and dynamic orbital conditions. The translational potential of such matrices was further proven by the implementation of xenogeneic-free culture protocols and the application of physiological pulmonary shear forces with the characterization of a mature endothelium. Additionally, proof-of-concept pre-endothelialization of clinically-relevant TESVs was successfully demonstrated and the retention of EPCs endured also after crimping procedure, which opens up new avenues towards next-generation pre-endothelialized TESVs, that may provide an enhanced hemocompatibility profile, which may therefore be particularly beneficial in certain indications and for selected patients at high risk for thromboembolic events. Together, our findings provide further important insight into the potential of hTEMs as a next-generation technology for the development of cardiovascular implants.