1 INTRODUCTION

Pericardium of autologous, homologous, and xenogeneic origin has been used extensively in the clinic for cardiovascular tissue reconstruction. In particular, glutaraldehyde (GA)-fixed xenogeneic pericardium has been largely employed due to its substantial availability, for various applications such as pericardial closure in the prevention of cardiac herniation,1 ventricular outflow tract reconstruction in treatments of congenital and acquired heart defects,2, 3 carotid endarterectomy in vascular surgery,4 and, finally, valvular reconstruction.5, 6 Despite GA-treated xenogeneic pericardium has shown a better performance over other reconstruction strategies, such as synthetic grafts, its use has often been associated with inflammation, calcification, aneurysmal dilatation, and lack of in vivo regeneration, and finally, leading to reoperation.2, 7-9 Likewise, bioprosthetic heart valves (BHV), made of GA-treated bovine or porcine pericardium, have been associated with structural valve deterioration (SVD), leading to low valve durability especially in young patients (under 60).6, 10 The in vivo deterioration, inflammation and calcification have been associated not only with the GA-fixation but also recipient immune response to the antigenic components of xenogeneic tissues.11 In particular, BHV have been playing a major role in the study of tissue antigenicity due to their direct exposure to the immune system via the blood stream. The major xenoantigens identified as playing a key role in recipient immune response are the Galα1-3Gal (α-Gal) epitope and the non-human sialic acid N-glycolylneuraminic acid (Neu5Gc).12-15 Both xenoantigens are not synthesized in humans due to inactivation of the GGTA1 and the CMAH gene, respectively; nevertheless, circulating IgA/IgG/IgM antibodies against both xenoantigens are still expressed in humans.16 The presence of α-Gal epitope on xenogeneic tissue has been associated with hyperacute graft rejection, whereas that of Neu5Gc with acute graft rejection upon transplantation in humans.17 Recent studies reported the presence of another xenoantigen on porcine BHV, produced by the porcine beta-1,4-N-acetylgalactosaminyltransferase 2 (β4GALNT2).18 All these findings suggest the need for further investigation into the presence of other xenoantigens, especially among carbohydrates, considering the vast diversity of these biomolecules among different species.19

Besides carbohydrates, other compounds have been associated with BHV deterioration and calcification, that is, lipids; indeed, phospholipids have been reported to play a key role in the calcification process of pericardial BHV, as their phosphate heads act as binding sites for circulating calcium ions.20 Companies producing pericardial BHV have introduced new treatments for the removal of lipids in an attempt to inhibit valvular calcification.21-23

Decellularization has been employed in order to reduce the immunogenicity of allografts and xenografts, by removing cellular material. While decellularized human valves have shown excellent long-term results, reduced immunogenicity, and capability of growing with the patient,24-26 xenogeneic valves implanted in humans have shown rather dramatic failure.27, 28 Despite some of the known immunogenic compounds, such as cells and water-soluble compounds, have been reported to be removed with decellularization,29 other xenoantigens, such as carbohydrates present in ECM glycoproteins (including α-gal and Neu5Gc), cannot be removed unless using targeted treatments, such as enzymatic, or through the silencing of genes correlated with the expression of specific xenoantigens. As an example of gene silencing, in vivo short-term studies using GGTA1 knockout (GGTA1 KO) tissue, (lacking the expression of α-gal), transplanted in non-human primates, showed a lower immune response compared to standard biological heart valves.30 Further long-term in vivo study in non-human primate showed that hearts isolated from GGTA1 KO pig, expressing a human membrane cofactor protein (CD46) and human thrombomodulin, survived up to 945 days.31 Nevertheless, graft survival was improved using substantial immunomodulatory strategies. In vitro studies using porcine tissue from triple-knockout pigs for GGTA1, CMAH, and β4GALNT2 genes showed encouraging results in terms of decreased binding of human IgG/IgM to knockout tissue compared to wild type.32

Elseways, carbohydrates removal using enzymatic treatments has been limited to enzymes with specificity against the known immunogenic carbohydrates or specific types of glycosylations. Indeed, carbohydrases such as α-galactosidase or peptide:N-glycosidase F (PNGase F) have been applied for the removal of the α-Gal epitope and N-glycans, respectively, from xenogeneic tissues.33-35 Another carbohydrase with broader specificity, α-amylase from porcine pancreas, has been used for the production of porous foams derived from decellularized porcine left ventricle.36

Besides carbohydrates removal, enzymes have been largely applied for tissue decellularization. Notably, the removal of DNA fragments using nucleases (such as DNases, RNases, and benzonase),33 or lipids from fatty tissues using lipases and phospholipases has been of particular focus in the decellularization field. For example, lipase from porcine pancreas has been used for decellularization of human lipoaspirate, omentum, and others.37 Pancreatic phospholipase A2 has been used for phospholipids removal in cornea decellularization.38 Furthermore, proteases like trypsin, collagenase, dispase, and thermolysin have also been applied in decellularization for the cleavage of cell-cell and cell-ECM attachments.37

The approach for glycans removal suggested in our study is based on the use of enzymatic supplements, commonly applied for the treatment of digestive disorders, or enzymatic mixtures used in various industrial applications, that is, food processing, textile industries, laundry industry, etc.39 Such mixtures contain enzymes that target the cleavage of several types of glycosidic linkages among specific carbohydrates, lipids, and, some of them, also proteins. The major advantages of these products are their cheap cost, as well as their broad specificity, which could allow for an overall unspecific glycans and lipids reduction.

The application of digestive enzymatic supplements and mixtures for carbohydrates removal from decellularized material is totally novel; these mixtures could potentially decrease the immunogenicity of xenogeneic tissues, as well as the costs of tissue processing with a view of implementing low-cost, large-scale production of decellularized tissues for clinical application.

2 MATERIALS AND METHODS 2.1 Tissue collection and preparation 2.1.1 Pericardium from wild-type pigs

Porcine hearts still covered with pericardium were obtained from 6-month-old German Landrace pigs from the local slaughterhouse (Merhold GmbH) within 2 hours of slaughter. One GGTA1-KO heart, which served as lectin control, was received from Friedrich Loeffler Institute, Institute of Farm Animal Genetics in Mariensee, Germany. For transportation, hearts were immersed in ice-cool PBS solution containing 1% v/v of penicillin (10 000 UI/mL)/streptomycin (10 000 UI/mL) (Biozym Scientific GmbH, 882082) and 1 IU/mL gentamycin (Merck, A2712). The parietal layer of pericardium was isolated from the heart, and the excess fat and tissue were trimmed carefully. Some pericardia were left untreated and tested immediately after dissection for histology and cryo-embedding for immunohistochemistry/lectin stains. The remaining pericardia (n = 6) were directly decellularized. Samples of each were enzymatically treated according to different protocols, and processed for histological, immunohistochemical/lectin analysis.

2.2 Decellularization

For decellularization, a modified version of the protocol developed for the decellularization of porcine pulmonary valves40 was utilized. Each step of the protocol was carried out under agitation (180 rpm) in an orbital shaker (GFL 3031), using 2 mL per 1 cm2, unless otherwise stated. Briefly, pericardia were placed in 250 mL square Schott bottles and disinfected in an antibiotic cocktail (1 mL/cm2) made of 0.72 U/mL (0.2 mg/mL) polymyxin B sulfate (Sigma, P4932), 0.37 U/mL (0.05 mg/mL) vancomycin hydrochloride hydrate (Sigma, 861987), and 500 U/mL (0.5 mg/mL) gentamycin sulfate (Biochrom, A271-25), at 37°C for 1 hour. Then, tissues were washed in hypotonic buffer (10 mmol/L Tris, 2.7 mmol/L EDTA) for 24 hours (2 × 12 hours) at RT. Pericardia were then treated with 0.5% (v/v) Triton X-100 for 24 hours (2 × 12 hours) (Roth, 3051.4), followed by 0.5% (w/v) SDS for 24 hours (2 × 12 hours) (Roth, 2326.2), and washed 12 times with PBS, (12 hours for each wash) at RT. Samples were stored in 50 mL reaction tubes in PBS with 1% v/v penicillin/streptomycin (Biozym Scientific GmbH, 882082) at 4°C until further analysis.

2.3 Enzymatic treatments

After decellularization (§2.2), samples (1.5 × 1.5 cm) of pericardia were treated enzymatically applying protocols developed in-house (Table 1). Briefly, ZYC (ZyCarb, Houston Enzymes, HN006-120) and ZYP (Zyme Prime, Houston Enzymes, HN002-S90) digestive enzymatic supplements were applied at different dilutions and duration (Table 1). One capsule of enzyme, either ZYC or ZYP, dissolved in 1 mL of sterile PBS served as stock solution. Further dilutions were performed in sterile PBS.

TABLE 1. Decellularization protocols and enzymatic treatments applied in this study Enzymatic mixture Protocol Dilution in PBS Incubation time Washing step in PBS ZyCarb (ZYC) ZYC 1 cps/5 mL 5 h; 24 h 3 × 10 min ZYC_1:10 1:10 45 min; 1.5 h 3 × 10 min ZYC_1:50 1:50 5 h; 24 h 3 × 10 min ZYC_1:100 1:100 5 h; 24 h 3 × 10 min Zyme Prime (ZYP) ZYP 1 cps/5 mL 2.5 h; 5 h; 24 h 3 × 10 min ZYP_1:10 1:10 5 h; 24 h 3 × 10 min ZYP_1:50 1:50 5 h; 24 h 3 × 10 min Phenol Assist (PA) PA 1 cps/5 mL 24 h 3 × 10 min Veggie Gest (VG) VG 1 cps/5 mL 24 h 3 × 10 min Carb Digest (CD) CD 1 cps/5 mL 24 h 3 × 10 min Rohament CL Rohament No dilution 24 h 7 × 10 min Abbreviations: CD, Carb Digest; cps, capsule; PA, Phenol Assist; VG, Veggie Gest; ZYC, ZyCarb; ZYP, Zyme Prime.

PA (Phenol Assist, Kirkman, 0835-090), VG (Veggie Gest, Enzymedica), and CD (Carb Digest, Kirkman, 0933-120) digestive enzymatic supplements were tested only by using one dilution, that is, one capsule dissolved in 5 mL of sterile PBS.

Rohament CL (AB enzymes, 9012-54-8) enzymatic mixture, generally used in food industry for fruit and vegetable processing, distilling, and brewery, was tested undiluted. Indeed, 5 mL of filter-sterilized Rohament CL were used for the treatment of one pericardial patch. The pH of Rohament CL was measured prior to treatment, and it corresponded to 3.75. Owing to this, tissues were washed until the pH of the wash solution reached a neutral value of 7.2-7.4 after washing (after 7 washes).

All enzymatic treatments were conducted in 50 mL reaction tubes with 5 mL of enzymatic solution (volume-to-surface ratio: 5 mL/2.25 cm²) at 37°C with agitation (150 rpm). Enzymatic solutions were filter sterilized prior to use (PES filter, 0.22, Sartorius). Washing steps, 10 minutes each, with 2 mL sterile PBS were performed at RT under agitation (150 rpm). After treatments, tissues were stored in 2 mL of sterile PBS. The composition of the enzymatic mixtures is described in Table 2.

TABLE 2. Composition of enzymatic mixtures and enzymatic supplements employed in this research study Enzymatic mixture Enzyme blend ZyCarb (ZYC) Amylase, Glucoamylase, Protease 4.5, Alpha-Galactosidase, Lactase, Lipase, and Xylanase Zyme Prime (ZYP) Amylase, Glucoamylase, Protease 4.5, CereCalase® (Hemicellulase, Glucanase, Phytase), Alpha-galactosidase, Lactase, and Lipase Phenol Assist (PA) Xylanase, CereCalase®, Glucoamylase, Phytase, Alpha-galactosidase, Beta-glucanase, Cellulase, and Amylase Veggie Gest (VG) Amylase (Thera-blend), Alpha-galactosidase, Glucoamylase, Cellulase (Thera-blend), Protease (Thera-blend), Maltase, Lactase, Invertase, Lipase (Thera-blend), Pectinase (w/Phytase), Hemicellulase, and Xylanase Carb Digest (CD) Glucoamylase, Transglucosidase (Isogest), Diastase, Lactase, Invertase, and Amylase Rohament CL Cellulase 2.4 Lectin histochemistry for staining of carbohydrate structures

Samples of wild-type native and decellularized pericardia w/o enzymatic treatment, as well as from GGTA1 KO pericardium embedded in Tissue-Tek® OCT medium (Sakura) were cut into sections (7 µm). These were investigated through lectins stains for overall glycan removal. All conducted lectin stains including the lectin specificity are given in Table 3.

TABLE 3. Lectins used for staining carbohydrates on porcine pericardium Lectin Provider, catalogue number Labeling Dilution Specificity Stained protocol Ricinus Communis Agglutinin I (RCA I) Vector Laboratories, RL-1082 Rhodamine labeled 1:200 Galβ1-4GlcNAc and Galβ1-3Gal (weak) on both N- and O-glycans; GM162, 64 All ZYC and ZYP protocols Wheat Germ Agglutinin (WGA) Vector Laboratories, RL-1022 Rhodamine labeled 1:200 (GlcNAc)n and multivalent Sia on both N- and O-glycans; Neu5Gc and Neu5Ac59, 60; clusters of GalNAcα-Thr/Ser (Tn)81; GM162; IV3NeuAc-nLcOse4Cer and IV6NeuAc-nLcOse4Cer63 and other glycolipids61 All ZYC and ZYP protocols Soybean Agglutinin (SBA) Vector Laboratories, B-1015 Biotinylated 1:100 Terminal GalNAc (especially GalNAcα1-3Gal) on O-glycans; α- or β-linked GalNAc, Galα1-4Gal-Glc; GalNAcα-Thr/Ser (Tn)72, 73, 81; GM162; and other glycolipids76 ZYC_1:100_24h; ZYP_1:10_24h Wisteria Floribunda (WFA) Vector Laboratories, B-1355 Biotinylated 1:100

Terminal GalNAc (eg,

GalNAcβ1-4GlcNAc) and

Galβ1-3(−6)GalNAc on N- and O-glycans; GalNAcα-Thr/Ser (Tn)71, 74, 81, 82; glycolipids76

ZYC_1:100_24h; ZYP_1:10_24h Peanut Agglutinin (PNA) Vector Laboratories, B-1075 Biotinylated 1:100 Galβ1-3GalNAcα-Thr/Ser (T68), and to a lesser extent GalNAcα-Thr/Ser (Tn81) on O-glycans; GM165 ZYC_1:100_24h; ZYP_1:10_24h Maackia Amurensis lectin I (MAL I) Vector Laboratories, B-1315 Biotinylated 1:100 Siaα2-3Galβ1-4GlcNAc on N- and O-glycans; glycolipids76 ZYC_1:100_24h; ZYP_1:10_24h Jacalin Vector Laboratories, B-1155 Biotinylated 1:50 Galβ1-3GalNAcα-Thr/Ser (T75) and GalNAcα-Thr/Ser (Tn70, 71) on O-glycans ZYC_1:100_24h; ZYP_1:10_24h; PA; CD, Rohament Erythrina Cristagalli (ECL, ECA) Vector Laboratories, B-1145 Biotinylated 1:100 Galβ1-4GlcNAc on N- and O-glycans; glycolipids77, 78 ZYC_1:100_24h; ZYP_1:10_24h; PA; CD, Rohament Note The references for glycan structures found on N- and/or O-glycosylation are,66, 67 unless otherwise stated. The relevant references for glycolipids are specified in the table per each lectin. Abbreviations: Gal: galactose; GalNAc: N-acetylgalactosamine; Glc: glucose; GlcNAc: N-acetylglucosamine; GM1: monosialotetrahexosylganglioside; Neu5Ac: N-acetylneuraminic acid; Neu5Gc: N-glycolylneuraminic acid; Sia: sialic acid; T: Thomsen-Friedenreich antigen; Thr/Ser: threonine/serine; Tn: Tn antigen; ZYC: ZyCarb; ZYP: Zyme Prime.

Samples for RCA I and WGA fluorescence stain were incubated in blocking medium made of 1% w/v bovine serum albumin (BSA; VWR, 422361V) in PBS for 60 minutes at RT, whereas samples stained with biotinylated lectins were blocked in 10% v/v donkey serum (Sigma, D9663-10ml) for 60 minutes at RT. Incubation in lectins was carried out overnight at 4°C. The dilution buffer for RCA I and WGA lectins consisted of PBS/Tween (0.05% v/v), whereas that for biotinylated lectins was made of PBS supplemented with 1% w/v BSA.

Upon overnight incubation, samples treated with RCA I and WGA were washed 3 times in PBS/tween (0.05% v/v), and then incubated in 1 μmol/L DAPI solution (Life Technologies, D1306) for 15 minutes at RT, followed by mounting. Contrarily, Samples labeled with biotinylated lectins were further incubated with Avidin-FITC (Vectorlabs, A-2001) at a dilution of 1:150 for 2 hours at RT. Subsequently, tissue sections were washed 3 times in PBS, and then incubated in 1 μmol/L DAPI solution (Invitrogen, MW350.25) for 15 minutes at RT, followed by mounting.

Samples were stored at 4°C in the dark, until further analysis. Images were captured using the F37-542 Nikon filter for Rhodamine labeled lectins, the F37-483 Nikon filter for FITC labeled lectins, and the F39-377 Nikon filter for DAPI, with a Nikon TE300 Eclipse light microscope, incorporating a Nikon 191 Digital Sight DS-U3 camera controller. Pictures were processed through the NIS-Elements D Microscope Imaging Software (Nikon 192 Instruments) and merged for the lectin and DAPI stains using ImageJ software (NIH).

2.5 Histological assessment

Samples of native, decellularized, and enzymatic treated pericardia were fixed in 3.5% (v/v) formalin (Fisher) for 24 hours, dehydrated, and then embedded in paraffin. Sections (6 μm) were cut, dewaxed, and stained. For visualization of elastic fibers, Elastica van Gieson's picrofuchsin staining (Morphisto, 12739) was performed. After incubation for 15 minutes in Resorcin-Fuchsin, sections were washed in tap water for 1 minutes. Samples were then treated in Weigert´s iron hematoxylin for 5 minutes, for staining cell nuclei, washed for 10 minutes in tap water. Subsequently, sections were treated for 10 seconds in 1% (v/v) hydrochloric acid alcohol, then washed in deionized water for 5 seconds, and finally treated in van Gieson Picrofuchsin for 10 minutes, followed by washing in deionized water for 5 secconds. Lastly, samples were mounted with Eukitt mounting medium. Pictures were processed through the NIS-Elements D Microscope Imaging Software (Nikon 192 Instruments). Elastic fibers stained black-purple, cell nuclei brown-black, and connective tissue red.

2.6 Uniaxial tensile test

Uniaxial tensile test was performed on fresh, decellularized, and decellularized treated with ZYC at 1:100 at 24 hours and with ZYP at 1:10 at 24 hours, porcine pericardium. Six strips from each of the four groups, measuring 10 mm long and 5 mm wide, were cut out. Samples’ thickness was measured prior to testing at three different points along sample´s length using a Mitutoyo thickness gauge, and then averaged.

Pericardial specimens were subjected to low strain rate uniaxial tensile loading to failure in an Instron tensile testing machine (5967 Dual Column 196 Series, 100N load cell), at a pre-load of 0.01N and an extension rate of 20 mm/min. Testing took place in PBS at a temperature of 37°C.

The stress-strain behavior of each specimen was analyzed by means of the following four parameters, calculated as described by Korossis et al41: ultimate tensile strength, failure strain, maximum force, and Young’s Modulus (measured in the collagen phase, Coll-E).

2.7 Burst pressure test

Burst pressure test was performed on fresh, decellularized, and decellularized treated with ZYC at 1:100 at 24 hours and with ZYP at 1:10 at 24 hours, porcine pericardium. The test setup consisted of a water tank, a manometer, a custom-made fabric holder, and a compressed air valve. The water tank, which was attached to the manometer and to a compressed air valve, was filled to three quarters of its total volume with water. Pericardial samples with a size of 1 cm2 were placed in the tissue holder, which was then mounted on the lid of the water tank. The tank was gradually pressurized using compressed air. The maximum pressure that could be held by the sample prior to bursting (burst pressure) was automatically recorded with the manometer.

2.8 Statistical analysis

Results were expressed as mean ± 95% confidence interval (CI). ANOVA with Tukey post-test was performed to analyze significant differences among native and decellularized samples, ZYC at 1:100 at 24 hours and ZYP at 1:10 at 24 hours in uniaxial tensile test. Significant differences between ZYC at 1:100 at 24 hours and ZYP at 1:10 at 24 hours in the burst pressure test were analyzed using an unpaired t test. The number of samples used for both uniaxial tensile test and burst pressure test is specified in the Figures. Statistical significance was determined at .05 alpha level.

3 RESULTS

Control groups were used for lectins histochemistry and histological assessment. For lectins stains, in particular, native pericardium from wild-type pig served as positive control, whereas decellularized pericardia from wild-type pig served as control in order to compare the loss of carbohydrates before and after enzymatic treatments. Decellularized pericardium from GGTA1 KO pig served as a further group of comparison.

Regarding histological assessment, native and decellularized pericardia from wild-type pigs were used as control groups.

3.1 Lectins histochemistry and histology of ZYC- and ZYP-treated samples 3.1.1 Initial treatment with ZYC and ZYP (at 1 cps/5 mL)

WGA and RCA I lectins stains were performed to investigate carbohydrates removal after treatment with ZYC and ZYP at 1 cps/5 mL, at 5 and 24 h. The pictures for the stains are shown in Figure 1 (Figure 1, WGA [A, a-g] and RCA I [B, a-g]).

Lectin histochemistry of ZYC- and ZYP-treated groups (1 cps/5 mL). Native (A, B, a) and decellularized (A, B, b) pericardia from wild-type pigs, decellularized pericardium from GGTA1 KO pig (A, B, c), and ZYC- and ZYP-treated pericardia from wild-type pig stained with WGA (A, a-g) and RCA I (B, a-g), and counterstained with DAPI. Carbohydrates binding lectins and cell nuclei were stained red and blue, respectively. Scale bars indicate 100 μm. NP: native pericardium; DP: decellularized pericardium; DP_GGTA1 KO: decellularized pericardium from GGTA1 KO pig; DP_ZYC_5h/24h: ZYC treatment for 5 h/24 h; DP_ZYP_5h/24h: ZYP treatment for 5 h/24 h

Decellularized tissues (Figure 1 A, B, b, c) showed a mild decrease in signal intensity for both WGA and RCA I compared to native tissue (Figure 1 A, B, a). On the contrary, a pronounced decrease in signal intensity in comparison to native and decellularized tissues was observed for ZYC-treated group at both time points, for both WGA and RCA I stains (Figure 1 A, B, d, e). ZYP-treated group showed a strong signal reduction after 24 hours treatment for both WGA and RCA I stains (Figure 1 A, B, g), and after 5 hours for WGA (Figure 1 A, f); nevertheless, a mild stain was still observed at 5 hours for RCA I (Figure 1 B, f).

The results for the Elastica van Gieson staining are reported in Figure 2 (Figure 2 a-f). Native pericardium showed a compact structure, thinner than decellularized tissue (Figure 2 a and b, respectively). Gaps were observed in decellularized sample (Figure 2b); elastic fibers were observed in both control groups. In contrast to this, decellularized and enzymatic treated tissues resulted in a cell-free matrix (Figure 2 b-f). ZYC treatment highly impaired tissue structure at both time points (Figure 2 c, d), in particular, after 24 hours; whereas ZYP-treated tissue demonstrated an intact and compact histoarchitecture at 5 hours time point, with elastic fibers loss after 24 hours (Figure 2 e, f). Furthermore, tissue appeared thinner after ZYP treatment at both time points (Figure 2 e, f).

Histological characterization of ZYC- and ZYP-treated groups (1 cps/5 mL). Native (a), decellularized (b), and ZYC- and ZYP-treated pericardia from wild-type pig stained with Elastica van Gieson. Scale bars indicate 100 μm. NP: native pericardium; DP: decellularized pericardium; DP_ZYC_5h/24h: ZYC treatment for 5 h/24 h; DP_ZYP_5h/24h: ZYP treatment for 5 h/24 h

Owing to the promising results in terms of carbohydrates removal, in particular at 24 hours, but due to the impairment of the ECM at these dilutions, further dilutions and time points were investigated for both ZYC and ZYP in order to identify the best compromise in terms of glycans removal and maintenance of ECM components.

3.1.2 ZYC and ZYP tested at different dilutions and time points

WGA and RCA I lectins stains were performed to investigate carbohydrates removal after treatment with ZYC and ZYP at various dilutions and time points, as reported in Figure 3 for WGA stain (Figure 3 a-n) and Figure 4 for RCA I stain (Figure 4 a-n).

WGA lectin histochemistry of ZYC- and ZYP-treated groups at different dilutions and time points. Native (a) and decellularized (b) pericardia from wild-type pigs, decellularized pericardium from GGTA1 KO pig (c), and ZYC- (d-i) and ZYP- (j-n) treated pericardia from wild-type pig stained with WGA, and counterstained with DAPI. Carbohydrates binding lectins and cell nuclei were stained red and blue, respectively. Scale bars indicate 100 μm. NP: native pericardium; DP: decellularized pericardium; DP_GGTA1 KO: decellularized pericardium from GGTA1 KO pig; DP_ZYC_1:10_45m/1.5h: ZYC treatment for 45 m/1.5 h; DP_ZYC_1:50_5h/24h: ZYC treatment for 5 h/24 h; DP_ZYC_1:100_5h/24h: ZYC treatment for 5 h/24 h; DP_ZYP_2.5h: ZYP treatment for 2.5 h; DP_ZYP_1:10_5h/24h: ZYP treatment for 5 h/24 h; DP_ZYP_1:50_5h/24h: ZYP treatment for 5 h/24 h

RCA I lectin histochemistry of ZYC- and ZYP-treated groups at different dilutions and time points. Native (a) and decellularized (b) pericardia from wild-type pigs, decellularized pericardium from GGTA1 KO pig (c), and ZYC- (d-i) and ZYP- (j-n) treated pericardia from wild-type pig stained with RCA I, and counterstained with DAPI. Carbohydrates binding lectins and cell nuclei were stained red and blue, respectively. Scale bars indicate 100 μm. NP: native pericardium; DP: decellularized pericardium; DP_GGTA1 KO: decellularized pericardium from GGTA1 KO pig; DP_ZYC_1:10_45m/1.5h: ZYC treatment for 45 m/1.5 h; DP_ZYC_1:50_5h/24h: ZYC treatment for 5 h/24 h; DP_ZYC_1:100_5h/24h: ZYC treatment for 5 h/24 h; DP_ZYP_2.5h: ZYP treatment for 2.5 h; DP_ZYP_1:10_5h/24h: ZYP treatment for 5 h/24 h; DP_ZYP_1:50_5h/24h: ZYP treatment for 5 h/24 h

Overall, all ZYC-treated groups showed an impressive decrease in signal intensity for WGA stain at all dilutions and time points (Figure 3 d-i), whereas for the RCA I stain some mild signal was still visible for ZYC at 1:10 at 45 minutes and 1.5 hours (Figure 4 d, g), and at 1:50 and 1:100 at 5 hours (Figure 4 e, f). Nevertheless, treatment with ZYC for 24 hours at 1:50 and 1:100 showed the best results in terms of carbohydrates loss for the RCA I stain (Figure 4 h ,i).

The Elastica van Gieson staining pictures for ZYC treatment are reported in Figure 5 (Figure 5 c-h). Overall, histological analysis showed that ZYC treatment was too harsh to the ECM of tissue when tested at low dilutions (1:10, all time points; Figure 5 c, f); whereas at higher dilutions, the connective tissue structure appeared more preserved (1:50 and 1:100, all time points; Figure 5 d, g and e, h, respectively). Nevertheless, elastic fibers were not visible at all the dilutions and time points tested, whereas they were quite visible in native and decellularized tissues (Figure 5 a, b). In addition to this, ZYC-treated tissues looked thinner compared to decellularized tissue prior to enzymatic treatment, in particular at a dilution of 1:10 (Figure 5 c, f).

Histological characterization of ZYC- and ZYP-treated groups at different dilutions and time points. Native (a), decellularized (b), and ZYC- and ZYP-treated pericardia from wild-type pig stained with Elastica van Gieson. Scale bars indicate 100 μm. NP: native pericardium; DP: decellularized pericardium; DP_ZYC_1:10_45m/1.5h: ZYC treatment for 45 m/1.5 h; DP_ZYC_1:50_5h/24h: ZYC treatment for 5 h/24 h; DP_ZYC_1:100_5h/24h: ZYC treatment for 5 h/24 h; DP_ZYP_2.5h: ZYP treatment for 2.5 h; DP_ZYP_1:10_5h/24h: ZYP treatment for 5 h/24 h; DP_ZYP_1:50_5h/24h: ZYP treatment for 5 h/24 h

Regarding ZYP treatment, most of dilutions and time points showed a prominent signal decrease for the WGA stain (Figure 3 d-i), with the exception of the 1:50 dilution, at 5 hours, which showed a mild signal (Figure 3 k); nevertheless, still lower than decellularized tissue (Figure 3 b). The most pronounced decrease in signal for the RCA I stain was observed for the 1:10 dilution at 24 hours (Figure 4 m). The other dilutions and time points demonstrated a decreased signal compared to decellularized tissue, with the 1:50 dilution, at 5 hours, showing a signal intensity similar to decellularized tissue (Figure 4 k). The results for the Elastica van Gieson stain for ZYP-treated samples are described in Figure 5 (Figure 5 i-m). Overall, tissues looked thinner than decellularized sample (Figure 5 b); nonetheless, elastic fibers were still visible at all dilutions and time points, with the exception of 1:10 dilution, at both 5 and 24 hours (Figure 5 i,l).