Vascular prostheses fabrication

Prostheses were produced from medical grade polyurethane solution by SBS method, as described elsewhere [12, 27]. Briefly, polyurethane ChronoFlex®C75A (Advanced Biomaterials, USA) was dissolved overnight in 1,1,1,3,3,3-hexafluoro-2-propanol (> 99%Fluorochem Ltd, UK) on magnetic stirrer. The polymer solution was transferred into syringe and fed to the inner nozzle of concentric nozzle system. The polymer solution flow rate was controlled by syringe pump. The inner diameters of inner and outer nozzles were 1.1 and 10 mm, respectively. Fibers were collected on rotating collector, 6 mm in diameter and 12 cm in length, mounted 10–30 cm away from the tip of inner nozzle. Prior to the SBS process, the collector was covered with a thin layer of 10% w/v poly(ethylene) glycol 2000 (Sigma Aldrich, Germany) solution in distilled water in order to simplify removal of the prosthesis from the collector surface. After the prosthesis deposition and its immersion (together with the collector) in distilled water for 2 min, the prosthesis was gently slid off the collector. The slight shrinking of the prostheses after the removal resulted in a final inner diameter of 5 mm.

Two variants of layered prostheses were produced: (a) Nano and (b) Micro. As shown in Fig. 1A, Nano prosthesis consists of the following layers: nanofibers (luminal), dense microfibers, microfibers, and aligned microfibers (outermost). Micro prosthesis consists of the following layers: dense microfibers (luminal), microfibers, and aligned microfibers (outermost). The SBS process parameters used for producing individual layers are shown in Table 1.

Fig. 1

A Layers arrangement in Nano and Micro prosthesis, (B) Cross section of Nano and Micro prosthesis’ wall, (C) stereoscopic image of Nano and Micro prosthesis, (D) macroscopic image of prostheses (Nano and Micro mix)

Table 1 SBS process parameters applied for each layer in Nano and Micro prostheses. The layer that is present in a given prosthesis type is marked with “ + ”, a layer that is absent is marked with “- “Morphology of the prostheses

The prostheses were cut open and flat samples with dimensions 0.5 × 0.5 cm were glued to the SEM stubs with conductive carbon adhesive tape. Samples of internal (n = 3) and external surfaces (n = 3) were prepared. To characterize cross-sectional sample’s morphology, samples of prostheses 0.5 cm in length (n = 3 for each type) were glued upright to SEM stubs. The samples were then coated with 15 nm of gold using sputter coater (K550 Emitech, Quorum Technologies). Images of every sample (n = 10) were taken at magnifications × 200, × 600, and × 5000 using scanning electron microscopy Phenom G1 (Phenom World). SEM images were used to determine fiber diameter, pore size, and prostheses thickness. To determine the percentage of fibrous area on the internal luminal surface of Micro prostheses, the percentage of fibrous surface was measured in n = 20 SEM images. In every sample, n = 100 fiber diameters were measured using Fiji software. For nanofibrous internal surface of Nano prostheses, pore size was determined using SEM images of luminal surface at magnification × 5000. For this, the threshold tool (Fiji software) was used to delineate the most surface pores and the area of n = 100 pores was measured using Fiji software The pores were approximated to be circular in shape and the pore size (diameter) was determined using the circle area formula.

3D view of cylindrical structures was provided by a stereoscopic microscope Leica M205 C (Leica Microsystems GmbH).

Porosity was determined individually for every prosthesis. Two prosthesis ends, 1 cm in length were cut off and weighted on analytical lab scale. Afterward, the samples (n = 2 for each prosthesis) were glued upright to SEM stubs and coated with 15 nm of gold as described above. SEM images (n = 6) at magnification 200 × were taken, and n = 30 wall thickness measurements were made for each sample to determine individual layers' thickness and total wall thickness. The results (sample weight (ms) and total wall thickness (\(_})\)) were averaged and used to determine prostheses porosity (\(\upvarepsilon )\) using formula:\(\upvarepsilon =\left(1-\frac}_}}_}\bullet }_}}}_}}\right)*100\%\), where \(_}\) is a density of polyurethane ChronoFlex®C75A, \(_}=1.2\mathrm/}^\) [28], \(}_}\) is a sample’s side surface determined using formula \(}_}=2\uppi (\mathrm+\updelta )\mathrm\), where r is a prosthesis inner radius, r = 0.25 cm and L is a sample length L = 1 cm. The results are presented as mean value ± SD.

Mechanical properties

Prostheses of 5 cm in length (n = 5 for each type) were placed in the pneumatic jaws of the testing machine Instron 3345 equipped with 50 kN static load cell. Prostheses were stretched at the rate of 10 mm/min until the break. Dedicated Bluehill software automatically determined maximum load, elongation at break, Young’s modulus, and ultimate tensile stress. The results are presented as a mean value ± SD.

Leakage and delamination tests

The leakage test was carried out as follows: prostheses of 4 cm length (n = 3 for each type) were mounted in a closed flow system connected to a peristaltic pump Zalipm PP1B-05A (Zalipm) and 0.9% NaCl solution was circulated in the system (through the prosthesis) for 1 h at a flow rate of 20 ml/min. During the test, samples were checked for any signs of leakage through the prostheses’ walls. After the leakage test, prostheses were dried at 20 °C for 24 h. Then, the samples were glued to SEM stubs with conductive carbon adhesive tape and covered with 15 nm layer of gold. Materials cross-sections were analyzed using scanning electron microscopy Phenom G1.

The above-described flow system was also used to test the permeability of the prostheses’ walls in contact with blood. Freshly drawn whole blood was connected to the flow system and the prostheses were perfused for 1 h at a flow rate of 20 ml/min. During that time, macroscopic observations were carried out to assess whether there is any blood leakage through the prosthesis’ wall.

Additionally, a static delamination test was carried out. Prostheses of 1.5 cm length (n = 3 for each type of prostheses and for each timepoint) were prepared and placed in 1.5 ml Eppendorf® test tubes fully filled with 0.9% NaCl solution. Test tubes were closed and placed in an incubator at 37 °C for 7, 14, or 30 days. After this time, the prostheses were dried at 20 °C for 24 h and investigated using scanning electron microscopy Phenom G1.

Hemocompatibility of materials

Blood tests were performed using fresh human blood from healthy volunteers. Blood was collected in 1.8 ml test tubes containing citrate (BD Vacutainer, Franklin Lakes, NJ, USA).

Static platelet adhesion

For static analysis, round shape samples (n = 2 for each type of material) were placed in 24-well plate with the luminal surface of the prosthesis facing up. In order to stabilize and flatten the material, each sample was placed in CellCrown (Sigma-Aldrich) inserts. Subsequently, 500 µl of 0.9% NaCl solution in ultrapure water was added to wells with samples and plate was incubated at 37 °C for 30 min. Then, NaCl solution was removed and 200 µl of platelet-rich plasma (PRP) was added to every well containing the samples. PRP was prepared using two “slow” centrifugations: 150 g for 14 min (first centrifugation) and 150 g for 12 min (second centrifugation). The platelet density in PRP was 1 × 106 platelets/µL. Plate with materials was incubated at 37 °C for 90 min. Next, PRP was removed, and samples were thoroughly rinsed with 0.9% NaCl to remove blood residues. Finally, samples were prepared for SEM analysis. Briefly, materials were incubated in 4% paraformaldehyde for 24 h at 4 °C. Next, the samples were dehydrated by 10 min immersion steps in 50, 60, 70, 80, 90, and 100% ethanol solution (EtOH), and for 20 min in 1:2 hexamethyldisilazane:ethanol (HMDS:EtOH), 2:1 HDMS:EtOH and 100% HDMS solution. Finally, the samples were glued to SEM stubs with conductive carbon adhesive tape (luminal surface of prostheses up) and covered with 15 nm layer of gold. The % of platelet-coated area was counted from SEM images of every sample, taken at 3000 × magnification. Additionally, pictures at magn. = 5000 × were taken in order to present the morphology of surface-adhered platelets in detail. The platelet adhesion assay was done in triplicate, with change of blood donor each time. For every sample n = 10 SEM images were taken. The average values for all materials were calculated from 60 images (10 images × 3 experiments × 2 samples).

In this assay, PTFE was cut from vascular prosthesis (FlowLine Bipore, Jotec) and used as a reference material that induces low platelet adherence.

Hemolysis

Round samples with diameter of 14 mm (n = 3 for each type of prosthesis) were placed in 48-well plate with the luminal surface of the prosthesis facing up. To separate erythrocytes from plasma, fresh blood was centrifuged at 700 g for 5 min and plasma was removed from blood tubes. Then, erythrocytes were diluted 20 × in ultracold PBS and 500 µl of erythrocyte suspension was added to wells with materials. PBS was used as a negative control and 0.2% TritonX-100 as a positive control. Triplicate samples were placed on a shaker at 300 rpm for 1 h, at 37 °C. Afterward, 600 µl of solution from every well was centrifuged at 700 g for 1 min, and 200 µl of supernatant was transferred triplicate to 96-well plate. The absorbance at 540 nm was measured using a plate reader Epoch Biotek (Biokom).

Hemolysis rate was calculated using the following formula:

$$HR=\frac_-_}_-_}*100\%$$

where: AS – sample absorbance, ACP – mean positive control absorbance, C– mean negative control absorbance.

Results are presented as mean hemolysis rate ± SD.

Endothelial cell culture

Human umbilical vein endothelial cells (HUVECs) were isolated from freshly collected umbilical cords (kindly provided by the Dept. of Gynaecology, University Hospital Erlangen) and grown in supplemented endothelial cell growth medium (EGM-2, Promo Cell, Germany). Accutase solution was used for cell harvesting. Cells from passages 1 or 2 were used in experiments. All experiments were repeated 3 times, in each experiment the material was used in duplicate. The use of human material was approved by the local ethics committee at the University Hospital Erlangen (review number 14-85_3-B from 01.02.2022).

Static cell seeding on flat materials – 2D model

Flat samples were cut off from cylindrical grafts, sterilized with 70% ethanol, washed with sterile PBS, and placed in 24 well cell culture inserts. Then, materials were seeded with HUVECs (5 × 104 cells/sample) and incubated at 37 °C for 1, 3, and 7 days. Culture media were changed 24 h after seeding and then every second day.

To analyze cell viability, Alamar Blue assay was performed according to manufacturer’s protocol. Briefly, after 1, 3, or 7 days of cell culture, materials with cells growing on the surface were transferred to a new 24-well plate and gently washed with sterile PBS. Then Alamar Blue working solution was added to each well (500 µl/well) and incubated with samples at 37 °C for 18 h in the dark. The fluorescence of the Alamar Blue solution was measured at Ex./Em = 550/590 nm using a plate reader (SpectraMax iD3, Molecular Devices).

Magnetic cell seeding on cylindrical prostheses – 3D model

Cell seeding was also performed on cylindrical vascular prostheses. For this, all materials were cut to equal length of 5 cm. Samples were sterilized with 70% ethanol, washed with sterile PBS, and placed in transparent cell culture tubes. 1% agarose solution was used to fix the prostheses in a vertical position inside the cell culture tubes. Before cell seeding prostheses were preincubated with EGM-2 medium for at least 1 h.

HUVECs were seeded on the lumen of the prostheses using magnetic seeding technique. Cells were pre-incubated with superparamagnetic iron oxide nanoparticles (SPIONs) in cell culture flasks for 24 h at 37 °C as described before [29]. After incubation, the SPION-loaded cells were harvested and counted. HUVECs were suspended in the culture media and transferred into the luminal space of each prosthesis (1 × 106 cells/prosthesis). Immediately after transferring the cell suspension, the scaffolds were exposed to a radially symmetric magnetic field for 15 min using the VascuZell endothelizer (Vascuzell Technologia S.L., Madrid, Spain). The cell culture tubes with prostheses were then carefully removed from the endothelizer and placed in the incubator for 1, 3, or 7 days. The culture medium was changed 24 h after seeding and then every second day.

Cell staining and image analysis

After the given cell culture period, cells growing on the lumen surface were fixed with 4% buffered paraformaldehyde (Roth GmbH, Karlsruhe, Germany) and permeabilized with 0.2% Triton X-100 (Sigma-Aldrich, Munich, Germany) in PBS. F-actin filaments were stained by Alexa488-phalloidin (Invitrogen, Thermo Fisher) and visualized using fluorescence microscope Zeiss Axio Observer Z1 (Zeiss, Jena, Germany) at 10 × magnification. To observe cells growing inside cylindrical prostheses, the materials were cut along the longitudinal axis, pressed to the glass slides, and then visualized using multiple mode (2 × 5). Cell counting was performed using the ImageJ software (Fiji, version 1.47v).

Data analysis and statistical analysis2D cell culture model and platelet adhesion assay

Cell coverage was calculated in 6 circular samples with a diameter of 11 mm (2 replicates × 3 independent experiments). For each sample, at least 3 microscopic images (magnification = 10x) were taken in randomly selected places and the cell coverage was calculated for every image. The average coverage was then calculated for each sample and the resulting boxplot was based on these 6 average values for all 6 samples. A boxplot in a%-b% range indicates that in a given group of materials, there was at least one sample with a% coverage and at least one sample with b% coverage.

3D cell culture model

Cell coverage was calculated in 5 cylindrical samples (diameter 6 mm, length 5 cm) from 3 independent experiments). For each sample, at least 2 multi-tile scan microscopic images (magnification = 10x) were taken. Each “tile” represents the standard analysis area at 10 × and the multi-tile scans covered the surface of 2 tiles (prosthesis circumference) × 5 tiles (prosthesis length), corresponding to an area of approx. 1.3 mm × 4.5 mm. For each sample, 2 multi-tile scans were performed and the results were averaged. Based on 5 averaged values for all 5 samples a boxplot was plotted in a%-b% range, indicating that in a given group of materials, there was at least one sample with a% coverage and at least one sample with b% coverage.

The results of the other measurements (mechanical analysis, delamination assay, hemolysis) were presented as mean values ± SD. Statistical significance of differences was analyzed using single-factor or two-factor analysis of variance (ANOVA) for p < 0.05 with post-hoc Tukey’s test (OriginPRO 2020b).