Porcine femoral and carotid arteries, provided by Life and Device S.r.l. (Turin, Italy), have been decellularized with a three-step protocol. In summary, the vessels have been treated with a 1 M NaCl, 8 mM CHAPS and 25 mM EDTA solution for 1 h at 37°C under stirring. After prolonged PBS rinsing, the matrices have been incubated with a second solution of 1 M NaCl, 1.8 mM SDS and 25 mM EDTA for 1 h at 37°C under stirring. Finally, in order to remove the residual DNA content, samples have been treated with a solution of 6.4 μM Deoxyribonuclease I from bovine pancreas, 0.1 M MgCl₂, and 0.9 M NaCl, and stirred for 16 h at room temperature (all reagents Sigma-Aldrich, United States). Decellularized vessels have been stored at −20°C before use.

Decellularized matrices have been enriched with 1 mg/ml polylysine (Sigma-Aldrich, United States). The process (sueGraft^®) used was optimized by TissueGraft S.r.l., all details are owned by the company.

To evaluate the efficiency of the enriching process, three different matrices were compared:

In order to verify the efficiency of the decellularization process, several analyses were performed (Wolf et al., 2012), including histology to evaluate the absence of cellular material using 4′, 6′-diamidino-2-phenylindole dihydrochloride (DAPI) and hematoxylin and eosin (H&E) staining. For the nuclear staining analysis (DAPI), samples were fixed in a 4% formaldehyde solution, then rinsed in PBS and soaked in DAPI solution (300 nM for 2 min). After an additional PBS rinsing, samples were observed under a fluorescence microscope.

For H&E staining, samples were fixed in 4% formaldehyde solution, dehydrated and paraffin embedded. Samples were then cut into sections of 5 μm, rehydrated and soaked in hematoxylin for 15 min. Following this, eosin solution (0.05% eosin in distilled water and acetic acid) was applied for 1 min. Finally, samples were dehydrated and observed with an optic microscope (Leica DM2500, Leica, Germany). Samples images have been acquired through a Leica DFC7000 T camera (Leica, Germany) and analyzed with Leica Application Suite X software (Leica, Germany).

In addition, the amount of DNA in each sample was also quantified through a DNA quantification assay. Native and decellularized vessel slices (5 mm × 5 mm) were lyophilized and digested for 16 h at 55°C in a buffer solution, containing 100 μg/ml proteinase K, 75 mM NaCl, 25 mM EDTA, 1% SDS, pH = 8. Samples were incubated with a 6 M NaCl solution for 30 min and centrifuged at 15,000 rpm; followed by 1 ml of isopropanol. In order to precipitate the DNA after centrifugation, the supernatant was removed, and 70% ethanol added. After evaporation of ethanol, 20 μl of ultrapure water was added and the amount of DNA measured with full-spectrum UV-visible measurements (NanoDrop 2000, Thermo Fisher Scientific, United States).

Finally, agarose gel electrophoresis was used to visualize the DNA fragmentation. Briefly, 50 ng of DNA from native and decellularized samples were mixed with EZ-Vision loading buffer (VWR, Italy), and loaded on a 1% agarose gel. After the electrophoresis, DNA was visualized with ChemiDoc Imaging System (Bio-Rad, Italy).

Mechanical properties of native and decellularized enriched matrices were tested. The Young’s modulus and tensile strength of the different samples was measured, briefly, Young’s moduli were quantified from the stress-strain curve obtained with the Instron 5564 (Instron Corporation, Norwood, MA, United States). System control and data analyses were performed using Instron^® Bluehill^TM 3 material-testing software (Instron Corporation, Norwood, MA, United States). Sample dimensions (length, width, and thickness) were measured in order to normalize the mechanical test parameters. Uniaxial tests were performed at room temperature and with a tension rate of 0.5 mm^∗s^–1.

Burst pressure was also measured. The test was performed using a hydraulic press, a high-precision digital pressure gauge and a PBS solution as a liquid. The sample was stressed by pushing the PBS until bursting, thus detecting the maximum operating pressure in laboratory conditions.

Mechanical properties were also evaluated after simulation of physiological conditions (dynamic condition test). Decellularized and polylysine-enriched vessels have been inserted in a circuit, composed of a peristaltic pump (Masterflex L/S, Cole-Parmer Instrument Company, United States), 10 mm diameter tubing (Cole-Parmer Instrument Company, United States), and a chamber containing the samples and filled with the PBS. Fluid speed was 15 ml^∗min^–1 and samples were tested for up to 30 days. Samples were weighed before and after the dynamic test in order to evaluate the samples’ weight loss after dynamic conditions were applied. Finally, both Instron and burst pressure tests, were performed after dynamic conditions.

Cell viability of human umbilical vein endothelial cells EA.hy926 (ATCC^® CRL-2922^TM), cultured on decellularized and enriched matrices, was assessed. Cells were cultured with Dulbecco’s Modified Eagle’s Medium (DMEM – Euroclone, Italy) enriched with 10% fetal bovine serum (FBS), 2 mM glutamine and 100 U^∗mL^–1 penicillin (Euroclone, Italy). Endothelial cells were cultured at 5^∗10⁴ cells^∗cm^–2. In order to evaluate cell viability, after 1, 3, and 7 days of culture, a MTS assay (CellTiter 96^® AQueous Non-Radioactive Cell Proliferation Assay, Promega, Italy) was performed. Briefly, a 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium solution was added to each sample for 3 h. Absorbance was measured by UV-VIS spectrophotometry (Victor X4, PerkinElmer, United Kingdom), at a wavelength of 490 nm. Measures were proportional to cell viability.

In order to evaluate cell adhesion and proliferation, EA.hy926 cells were cultured on decellularized and polylysine enriched matrices for 3 and 7 days. Matrices were then observed through Hematoxylin-eosin staining, as described in Section “Evaluation of the Decellularization Procedure.”

Cell adhesion was evaluated after simulation of physiological conditions (dynamic condition test). EA.hy926 were cultured on the lumen side of decellularized and polylysine-enriched vessels for 24 h. Afterward, samples were inserted in a circuit, composed of a peristaltic pump (Masterflex L/S, Cole-Parmer Instrument Company, United States), 10 mm diameter tubing (Cole-Parmer Instrument Company, United States), and a chamber containing the vessels samples and filled with DMEM enriched with 10% FBS. Fluid speed was 15 ml^∗min^–1 and samples were tested for up to 3 days.

Monocyte viability and inflammatory response elicited by decellularized and polylysine-enriched matrices was evaluated. After 24 h, cell viability on samples was evaluated with MTT assay. Briefly, culture medium was replaced with a 5 mg/ml solution of Thiazolyl Blue Tetrazolium Blue (MTT, Sigma-Aldrich, Italy) in RPMI 1640 and incubated for 4 h at 37°C. Medium was removed and DMSO was added, in order to dissolve formazan salts. Absorbance was measured by UV-VIS spectrophotometry (Victor X4, PerkinElmer, United Kingdom), at a wavelength of 570 nm. Absorbance is proportional to cell viability.

To evaluate the effect of decellularized and enriched samples on monocytes activity and polarization, cells on samples and controls (M0, M1, and M2) were seeded with 2^∗10⁵ cells^∗cm^–2. Differentiation of monocytes in control samples was obtained adding 10% FBS-enriched RPMI 1640 medium (Euroclone, Italy) with human (hr)GM-CSF (50 ng⋅mL^–1, Sigma-Aldrich, Italy), to induce M1 macrophages polarization, and in 10% FBS-enriched medium containing hrM-CSF (50 ng⋅mL^–1 Sigma-Aldrich, Italy) to induce M2 macrophages polarization. These conditions have been maintained for 7 days.

After 7 days, culture medium from each sample was collected and TNF-α release was evaluated through ELISA assay (RAB0476, Sigma Aldrich, Italy). Analysis was performed according to manufacturer’s instructions. Absorbance was measured by UV-VIS spectrophotometry (Victor X4, PerkinElmer, United Kingdom), at a wavelength of 450 nm. Measures were proportional to TNF-α concentration.

Thromboelastography (TEG) is a technique that measures several parameters relative to blood coagulation by evaluating the viscoelastic properties of blood from the beginning of coagulation to the clot rupture with fibrinolysis. Particularly important parameters are the coagulation formation, its progression, maximum strength, and stability, which provides important information on coagulation, fibrinolysis, and platelet functionality. Blood samples were collected in Vacutainer (BD, United States) containing sodium citrate and tested at 37°C. Blood was exposed to the control graft (Dacron^®), the decellularized scaffold and the enriched scaffold for 30 min and subsequently analyzed using Thromboelastograph^® (TEG^® 5000 Thrombelastograph^® Hemostasis Analyzer System) and the Kaolin TEG standard protocol. Briefly, after 30 min 1 ml of blood was collected and placed in a vial containing kaolin. After few seconds, 340 μl of blood were sampled in the thrombelastograph with 20 μl of calcium chloride, in order to eliminate the anticoagulant effect of sodium citrate. Coagulation parameters were evaluated through thromboelastogram obtained at the end of the testing time.

In order to evaluate the efficiency of the decellularization protocol, DAPI staining was performed on native and decellularized tissue. Figure 1 shows the native tissue and the presence of cellular nuclei in native tissues while, after the decellularization treatment, however, no nuclei were present. H&E histology results confirmed the preservation of the extracellular matrix microstructure within the decellularized samples. No cellular nuclei could be detected in the decellularized tissue samples.

DNA quantification on native and decellularized tissues strengthened the previous results, indicating a successful DNA removal from the matrix. In contrast to native tissue, which contained 860 ± 300 ng DNA per mg wet weight (n = 3), decellularized tissues showed a significant decrease of DNA content to 15 ± 11 ng DNA per mg wet weight (n = 3). Agarose gel showed that no DNA strands longer than 100 bp were available on decellularized samples, while in native samples the DNA presence was verified.

Samples were analyzed, as shown in Figures 2A,B, the survey spectra of decellularized and polylysine-enriched samples show differences, with the obtained surface composition reported in Table 2. Decellularized scaffolds showed peaks due to carbon, oxygen and silicon, and just a very weak signal from nitrogen. XPS tests just the outermost molecular layers of materials (the sampling depth is about 5–8 nm), and the obtained data likely reflects surface side-effects due to processing. Silicone is a common contaminant of surfaces and the peak of Si is a common finding in surface analysis of biological samples. There was also a lower than expected N/C and N/O ratio, which was probably due to processing aids (see also discussion of ATR-IR results) remaining on or combined with the sample surface. In contrast, the XPS spectrum of polylysine-enriched samples showed a strong peak due to nitrogen. Hence, the sample surface showed being enriched with a nitrogen-containing compound, clearly supporting the presence of polylysine. Further information was provided by ATR-IR analysis, as reported in the next section.

ATR-IR analysis of decellularized (red), polylysine-enriched (blue), and reference polylysine (green) in the amide band region of the IR spectrum (Figure 2C). The sampling depth of ATR-IR, using a diamond crystal, is around 1.6 mm, that is this technique is less surface-sensitive than XPS. All samples show the typical spectral features due to amide bonds, as expected. Concerning differences between tested samples, the Amide II band at about 1,550 cm^–1 shifts toward lower wavenumbers from decellularized to polylysine-enriched (arrow), centering close to the Amide II band of the polylysine reference. Also, around 1,400 cm^–1 the shape of the band (arrow) changes from decellularized to polylysine. Taken together, XPS and ATR-IR results strongly suggest that successful enrichment with polylysine of the decellularized sample is obtained through this process.

In Figure 3A the structure of decellularized and polylysine-enriched matrices are shown through confocal microscopy. The polylysine-enriched sample shows a structure of fractures better defined than in the decellularized sample, and the roughness is narrower than the latter. In general, the enriched sample shows more vertical oriented fractures, while in decellularized the lines are not so straight composing soft curves. The images suggest that the enrichment treatment changes the micro and macro collagen structure, orientating and bridging collagen fibers, as can be seen in Figure 3B, which confirms the data obtained from confocal microscopy. Indeed, decellularized sample’s tissue seems thinner and hollower with respect to both native and polylysine enriched matrix. The thickness and the wall microfiber structure of polylysine enriched matrix seems adequately similar, suggesting that the fiber reorganizing after enrichment process leads to a structure comparable to native sample.

Endothelial cells were seeded in the vessels lumen and cell proliferation rate was measured after 1, 3, and 7 days. Figure 5A shows comparable cell proliferation rates for decellularized and polylysine-enriched samples after 7 days. Endothelial cells’ adhesion and proliferation were also evaluated, using hematoxylin-eosin and DAPI staining. Results show that after 3 days, cells adhered on both decellularized and polylysine-enriched samples, while after 7 days of culture cells formed an almost continuous monolayer only on polylysine-enriched surfaces (Figure 5B). Finally, cell adhesion was evaluated after 3 days of dynamic conditions (Figure 5C). DAPI staining shows almost no cell nuclei on decellularized samples, indicating that cells do not remain adherent to that matrices. On the other hand, polylysine-enriched samples show cell nuclei almost forming a continuous monolayer, suggesting that cell adhesion is optimized on enriched matrices.

Monocyte viability on decellularized and enriched matrices was evaluated after 24 h after seeding. After 24 h, results show viable monocytes growth on control, decellularized and enriched matrices. Monocytes in contact with decellularized and enriched matrices showed an excellent viability. Thus, we proceeded analyzing monocytes polarization after 7 days of culture. Adequate media were used to induce macrophages polarization in M1 and M2, and used as control with respect to results obtained on macrophages growth on decellularized and enriched samples. As shown in Figure 6, M1 control released a significant amount of TNF-α, expressing an inflammatory phenotype, as expected. All the other conditions, both in M0 and M2 controls, as well as on decellularized and enriched matrices, TNF-a release is not significant, confirming the absence of inflammatory environment (Figure 6).

We evaluated the materials’ hemocompatibility using thromboelastography (TEG). Among the coagulation parameters, reaction time (R) indicates the time required to induce the clot formation. The matrices treated with polylysine have a significantly increased R-value with respect to Dacron and decellularized matrices. Fibrinogenic activity is rather decreased on enriched matrices, as a result the clot strength is lower than both matrices used as reference. The coagulation time parameter is directly correlated with the required time for the clot formation, in order to reach the maximum strength value before starting the dissolution process. This value increases in the samples treated with polylysine reflecting a benefit in terms of functionality. In addition, platelet aggregation confirms the lower rate in the clot formation process (Figure 7).

These results clearly indicate a favorable influence on coagulation parameters in the presence of polylysine enrichment in terms of hemocompatibility.