Cardiovascular diseases rank among the primary causes of mortality globally and addressing them appropriately continues to be a significant challenge in clinical practice. Among these, calcific aortic valve disease represents one of the most challenging issues worldwide with a high associated mortality rate that of 41% and 52% after 5 years, for mild and severe aortic stenosis respectively. Due to a lack of approved pharmaceutical treatments, the only usable clinical approach is the surgical replacement of the valve with prosthetic heart valves [1, 2, 3]. Among the various options, the most used involves the implantation of bioprosthetic heart valve (BHV) substitutes, usually made of pericardial tissue of bovine or porcine origin [4]. Such BHVs are fixed with glutaraldehyde (GA) for the improvement of the mechanical properties, assuring sterility and stability, as well as the reduction of immunogenicity. These devices are usually implanted in older people ($>$60 years) in whom the expectation of the bioprosthetic function should be higher than the patients’ lifetime; indeed, BHVs are infrequently used in younger patients because of their propensity to calcify, a process that is even more rapid in this type of patient because of their more active calcium metabolism and a more aggressive immune system [5]. Historically, calcification of BHVs has been ascribed to multiple external elements, as, for example, variability of GA bounds to the extracellular matrix (ECM), stress-related breakdown, and internal factors such as collagenolytic process and calcium deposition induced by remaining phospholipids [6]. Recently, the intrinsic pathway mediated by the immune system has become noteworthy, primarily due to recognition of the $\alpha$-Gal xenoantigen trigger [7, 8].

The $\alpha$-Gal epitope (Gal$\alpha$1-3Gal$\beta$1-4GlcNAc-R) represents a unique polysaccharide structure naturally produced on glycolipids, glycoproteins, and proteoglycans in the vast majority of mammals [9]. Although present in non-primate mammalians, prosimians, and New World monkeys, it is not naturally expressed in Old World monkeys, apes, and humans that instead normally produce antibodies against it, due to the expression of the antigen on the bacterial wall of the resident gut flora [10, 11]. The continued stimulation from these bacteria forces the immune system to produce anti-Gal antibodies that represent 1–8% and 1–2.5% of the circulating IgM and IgG human immunoglobulins, respectively [12, 13].

The choice to use rats or mice for in vivo biomaterial assessment, is a common practice, due to their cost-effectiveness, ready availability, manageability, and clearly defined immunological patterns. These animals are frequently employed to study the chronic changes that affect contemporary or novel BHVs’ leaflets when placed in a subcutaneous dorsal area. Specifically, the subdermal model guarantees a good serum and blood exposition of the implant ensuring cellular migration and enabling a reliable evaluation of anti-calcification treatments [14, 15].

Nevertheless, implanting BHV leaflets (mainly sourced from bovine or porcine pericardium) into a wild-type (WT) mouse represents a concordant xenotransplantation scenario with respect to the $\alpha$-Gal xenoantigen. In this scenario, both donor and recipient species exhibit this epitope, thereby excluding the opportunity to examine the role of an immune response against $\alpha$-Gal or the presence of preformed anti-$\alpha$-Gal antibody and thus negating evaluation of the effects thereof instead, recognition occurs in situ within the $\alpha$-Gal negative patient.

This study described the creation of a knockout (KO) mouse, genetically modified to be $\alpha$-Gal deficient by suppressing the gene responsible for $\alpha$1,3-galactosyltransferase (GGTA1). This animal model should replicate more predictably the behavior of the human’s immune system and the response to the presence of the xenoantigen by acquiring the ability to produce anti-Gal antibodies [16].

The primary target of the experimentation was to estimate calcium deposition in isolated leaflets from a surgical commercial BHVs of bovine pericardial origin implanted for 1, 2, and 4 months in GGTA1 KO mice, comparing it with an investigation conducted simultaneously in WT mice. Moreover, the possible remnant $\alpha$-Gal epitopes in various tissues and organs of the KO mice were also evaluated, taking into consideration prior observations of residual reactivity in biallelic GGTA1-KO pig cells as a potential contributor to chronic rejection of GGTA1${}^{-/-}$ organs [17, 18].

All surgical and experimental procedures conducted on animals followed the guidelines outlined in the Guide for the Care and Use of Laboratory Animals, as issued by the US National Institutes of Health (NIH Publication 85-23, revised 1996). The utilization of a mouse animal model for experimental purposes received approval from the Italian Ministry of Health under project registration number 17E9C.154 and authorization number 542/2020-PR. The GGTA1 KO mouse animal model is the property of BCI - Biocompatibility Innovation Srl (Este, Padua, Italy). The model generation process was conducted in cooperation with Polygene Transgenetic (Rümlang, Switzerland), and the animals are presently homed at Charles River Laboratories Italia (Lecco, Italy).

2.3 Mouse Subcutaneous Implantation

GGTA1${}^{-/-}$ (C57BL/6, 6 weeks old) and WT (n = 16) mice were anesthetized and shaved; a subcutaneous pocket was created for each mouse in the dorsal area. A portion size of 1 cm $\times$ 1 cm was isolated from leaflets of commercial GA-fixed BHV of bovine origine, and implanted singularly in the pocket of each animal, and the wounds were closed with 6/0 nylon sutures. After 1, 2, and 4 months, the mice were euthanized in a CO${}_2$ atmosphere and samples were carefully collected (Fig. 1). As control to evaluate the potential contribution of the surgical insult to the immunological stimulation n = 3 KO mice (surgical insult cohort) underwent the same procedure without the implant of the leaflet.

Fig. 1.

Experiment timeline and schematic implant procedure. (A) Study timeline (Abs = antibodies). (B) Schematic representation of subcutaneous surgical implant of commercial BHV’s leaflet. BHV leaflets (L) were inserted in a subcutaneous pouch (P) in the dorsal area of the animal (STEP 1). Puoch was sutured with 6/0 nylon (STEP 2). Explanted commercial BHV leaflets (EXL) were carefully harvested after 1, 2, and 4 months of incubation (STEP 3). BHVs, bioprosthetic heart valves.

BHV degeneration appears as a multimodal process engaging mechanical, chemical, and immunological pathways, leading to structural degradation over different time frames [24]. To date, GA still represents the gold standard fixative and sterilizing agent for almost all BHVs available on the market. Unfortunately, GA possesses an intrinsic chemical instability that consequently leads to a host immune response which is the first step in an articulated series of mechanisms that result in tissue calcification. The deposition of calcium is mainly due to the presence of the free aldehydic and carboxylic groups, which can attract Ca${}^{2+}$ ions becoming then stuck in the fibers of the tissues. The Ca${}^{2+}$ ions deposited act then as a nucleation site for other ions that are attracted by Van der Walls forces slowly generating crystals. This process is exacerbated also by the fact that GA treatment induces chemical changes in proteins within the tissue, altering their structure and composition, and making the tissue even more susceptible to interactions with calcium. Such denaturation may expose previously hidden sites that are prone to calcium binding [25].

To mitigate the impact on the calcification process, BHV manufacturers have proposed various modifications to GA fixation protocols. These modifications include additional steps aimed at chemically stabilizing reactive chemical groups. Processes such as GA detoxification by urazole, diamine spacer extension, treatment with 2-amino oleic acid, or incubation in ethanol are among the strategies developed to enhance GA stabilization, to delay calcific tissue dystrophy [26, 27]. Besides BHV calcification, which usually occurs over a long period (average of 10 years in $>$60 years people), several degenerative processes have also been found to occur within a very short time frame, even hours, immediately following implantation. These rapid mechanisms are attributable to the action of the patient’s immune system towards the implant [28]. The Translink international collaborative study group has recently highlighted this aspect. The study confirms that preformed antibodies generate an immune response that exacerbates calcification processes due to remaining antigens on BHV tissue [7, 8]. The obtained results are in line with what was already observed in the Translink study. The absence of $\alpha$-Gal in the KO model was paired with a rapid (one-month) immune response that led to a deposition of calcium that was four times higher than the one observed in WT animals in the same time. This rapid and severe response mimics what is, unfortunately, widely observed in young patients that undergone valve replacement in which the prosthesis fails due to calcification in a very short period [29, 30]. As already reported, the remnants $\alpha$-Gal xenoantigens starts to increase the anti-Gal antibody titers about two weeks after the implantation of the bioprosthetic device. The antibody titers continue to rise over time reaching a peak in 3–4 months specially for IgG and IgM isotype. Once raised, the levels of these antibodies do not return to basal levels even after 5 years from implantation. This fact may be easily ascribed to the instability of the crosslink with GA. Such chemical instability, emphasized by the mechanical stress to which BHVs are subjected, is involved in the breaking down of the GA network, exposing previously masked xenoantigenic epitopes. In this way, a mild but continuous immunological stimulation is maintained with the subsequent antibody formation [31, 32]. After the BHV leaflet implant, the GGTA1 KO animal model exhibited an increase in the IgG and IgM anti-Gal antibody titers similar to what was already observed in humans [33].

Thanks to their low cost, ready availability, ease of handling, and well-defined immune parameters, small animal models such as rats and mice are a common and practical choice for in vivo study of biomaterials. However, in the majority of the cases, the selected models are WT, preventing the possible immunological implications in the studied process. Conversely, the use of an $\alpha$-Gal KO model permits also to evaluate the implication of the immuno-mediated mechanisms alongside intrinsic and extrinsic ones that are already assessed by using WT models. Noteworthy, the findings from this study serve to reaffirm the existence of immunologically active $\alpha$-Gal antigen in BHVs currently utilized in clinical practice, as determined previously through an alternative analytical method [23].

The GGTA1 gene, silenced in the model under investigation in this study, encodes for $\alpha$ 1,3 GT galactosyltransferase enzyme that catalyzes the formation of the $\alpha$-Gal epitopes. This reaction involves the formation of an $\alpha$-(1,3)-glycosidic bond between a galactose and an N-acetyllactosamine (LacNAc) (Fig. 4) acceptor molecule present on both protein and lipid on cells’ plasma membrane [34]. However, this is not the only enzyme to catalyze this reaction. Milland and colleagues [35] have described the presence of isoglobotrihexosylceramide synthase (iGb3s) in GGTA1 KO mice that bond galactose exclusively to glycosphingolipids (Lac). This is the reason why a residual presence of $\alpha$-Gal epitopes has been still detected in biallelic GGTA1 KO pig animal model and this could be implicated in the rejection of GGTA1${}^{-/-}$ organs found exploring xenotransplantation in non-primate models. This study reported a residual presence of $\alpha$-Gal in the KO model, varying concerning the analyzed tissue district (Table 1), which is in line with what can be found in the literature.

Fig. 4.

Chemical structure of the two different types of the $\alpha$Gal epitopes synthesized by GGTA1 (NAc) or iGb3 (Lac) enzyme. The presence of NAc or Lac determines the immunogenicity of the whole structure.

From the collected data, $\alpha$-Gal results not completely silenced in the brain, and skin, which are tissues that featured a high presence of ceramides. As mentioned above, the iGb3s enzyme generates $\alpha$-Gal epitope exclusively capping lipids, mainly ceramides. For this reason, it is possible to observe residual $\alpha$-Gal epitope where an alternative pathway may restore function. In addition, the conduction of nerve impulses in mammals, both centrally and peripherally, is enabled by the presence of the myelin sheath that surrounds axons. The myelin sheath is a lipid-rich membrane of which 30% is composed of galactosylceramides (GalCer). Interestingly, GalCers’ levels are particularly high in the brain where they were seen to be also higher than the ones observed in WT models [36]. Several studies have highlighted the importance of ceramides in the brain: they mediate signal transduction and cell adhesion and are critical for the development of nervous tissues; this should explain why significant numbers of iGb3s-$\alpha$-Gal epitopes are also present in KO mice, the presence of which is unavoidable for proper brain and neuronal function [37]. Prominent expression of ceramides is also present in the stratum corneum of the epidermis of the mammals, which represents the outermost layer. It is composed of flattened, enucleated keratinocytes and a unique extracellular lipid matrix produced by differentiating keratinocytes and it presents a permeability barrier towards water, chemicals, and microorganisms [38]. A significant presence of ceramides is also found in the myocardium, liver, lungs, and kidneys [39]. The residual presence of $\alpha$-Gal detected in this study is in line with the study of Shao and colleagues who describe a reduction in the expression of the antigen between 5.19% and 21.74% in GGTA1-KO mice, apart from the skin and brain where, due to a higher concentration of ceramides, the reduction is much smaller [40].

It is important to underline that the presence of iGb3s in the KO model is not sufficient to trigger an immune response that involves the destruction of the cells presenting the epitope. This fact was validated by the work of Murray and colleagues who found that it is necessary to have a minimum basal level of $\alpha$-Gal for antibody-mediated rejection to occur in a GGTA1 KO mouse model [41].

In 2016 Butler and colleagues [42] demonstrated that the silencing of the porcine iGb3S gene did not impact assessments of expected acute rejection in pig-to-human and pig-to-primate xenotransplantation showing that iGb3S does not play a role in antibody-mediated rejection. The $\alpha$-Gal contribution due to the presence of iGb3S is not appreciable by heat-inactivated human and baboon sera antibodies when incubated with GGTA1 KO or GGTA1/iGb3S double KO pig tissue [42]. This is the reason, even if residues of $\alpha$-Gal antigen are still present, the GGTA1 KO mouse was revealed as an adequate animal model for evaluating the calcification propensity of currently adopted BHV tissues.

Since other pathological conditions have been linked to variations in circulating anti-Gal antibodies in humans, this animal model could also prove useful for the development of novel diagnostic approaches. Our research group has recently demonstrated a variation in the levels of circulating anti-Gal antibodies in Alzheimer’s disease patients compared with healthy subjects [43]. Since Alzheimer’s disease is a condition in which the mechanisms are not yet fully established and the treatments not optimized, the possibility of being able to study it in a reliable animal model seems promising and could open new diagnostic rather than curative approaches. In addition, Dahl and colleagues [44] have seen that GGTA1 KO mouse models show glucose intolerance and reduced pancreatic $\beta$-cell function, opening up the possibility of studying diabetic diseases with this model. The same model also showed rapid development of cataracts, 3–6 weeks after birth, probably because of glucose intolerance, thus mimicking what occurs in humans [45, 46]. This KO model is also useful in the study of wound healing, as it has been shown that the application of $\alpha$-Gal-containing nanoparticles in situ increases endothelial migration and accelerates wound closure times [47]. Finally, such a model could help to better understand the mechanisms underlying the onset of the $\alpha$-Gal syndrome. This syndrome, which most often initiates when a Lone Star tick bites a subject, is an allergic reaction that sometimes manifests in severe forms that can be life-threatening. It is triggered when the patient encounters $\alpha$-Gal-containing products, such as red meat, milk, and even some drugs. This syndrome has been growing in importance in the United States in recent times due to an increasing number of diagnosed cases so much so that FDA approved the use of $\alpha$-Gal KO pigs for human food and potential therapeutic uses [48, 49].

Given the increasing attention being focused on $\alpha$-Gal and its repercussions on human health, the possibility of being able to use large KO animals such as cattle and pigs for medical as well as food purposes opens up new perspectives for treatment, especially in the cardiovascular field. Indeed, animals, genetically engineered to not express $\alpha$-Gal and other xenoantigens such as Neu5Gc [50, 51], could be a promising source of organs and tissues, as recently demonstrated by the first two KO pig heart transplants in human patients which successfully avoided the hyperacute rejection reactions [52, 53].

Small animal models, such as mice and rats have been widely used for years as a reliable tool for in vivo assessment of calcification propensity of biomaterials through subcutis implantation. This approach is technically easier and cheaper than implanting entire bioprostheses in large animals moreover, mice present an accelerated calcium metabolism that can mimic what is observed in clinical specimens especially in young patients [54]. However, the implanted biomaterial is not subjected to continuous blood flow and pressure or to any other mechanical stress to which a BHV may be subjected that can influence the calcification response.

The implantation of commercial BHV leaflets has been shown to activate the $\alpha$Gal KO murine immune system, leading to a substantial generation of anti-Gal antibodies of both IgM and IgG classes. This immune response in mice appears to closely resemble the reactions observed in humans, as extensively documented in the literature. These findings underscore crucial factors in assessing the effectiveness of anti-calcific treatments, emphasizing the importance of selecting small animal models, favoring the KO model. This choice may be mandatory to prevent the potential risk of dramatically underestimating the impact of the immune-mediated reactions to xenoantigens.

The authors declare that all datasets on which the conclusions of the manuscript depend are included in the study and are available to the readers.

FN and AG designed the research study. GS and FN performed the research. RM and CG analyzed the data. GS and FN wrote the manuscript. All authors have participated sufficiently in the work to take public responsibility for appropriate portions of the content and agreed to be accountable for all aspects of the work in ensuring that questions related to its accuracy or integrity. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

All animal experiments and surgical procedures adhered to the guidelines outlined in the Guide for the Care and Use of Laboratory Animals, as published by the US National Institutes of Health (NIH Publication 85-23, revised 1996). The utilization of a mouse animal model for experimental purposes received approval from the Italian Ministry of Health under project registration number 17E9C.154 and authorization number 542/2020-PR.

The GGTA1 KO mouse animal model was kindly supplied by Biocompatibility Innovation Srl.

This research received no external funding.

Alessandro Gandaglia, Giulio Sturaro, and Filippo Naso are employees of Biocompatibility Innovation Srl. Robert J. Melder is a scientific consultant for Biocompatibility Innovation Srl. Cesare Galli declares no conflict of interest.