Volum 18, Issue 5 (2026)

Silk Fibroin-Based Polymer Patch with Antiproliferative Properties: Results of in vitro Tests

Prokudina E.S., Antonova L.V., Senokosova E.A., Kolomeets M.S., Khanova M.Yu., Koshelev V.A., Torgunakova E.A., Matveeva V.G., Kochergin N.A.

---

2026, Volum 18, Issue 5, page 19 Back

Introduction

Atherosclerosis, which underlies coronary artery disease, ischemic stroke, and peripheral artery disease, is the leading cause of morbidity and mortality worldwide [1, 2]. Patch angioplasty is an effective surgical method to correct hemodynamically significant stenoses caused by atherosclerotic plaques. Autologous vein patches are generally considered the “gold standard”; however, their use may be limited by widespread atherosclerosis or previous vascular interventions [3]. The use of synthetic patches (polytetrafluoroethylene and Dacron) can lead to neointimal hyperplasia and restenosis in the long term after implantation [3]. Biological patches made from xenopericardium are subject to calcification and degenerative changes [4].

The lack of an ideal vascular patch and the high rate of restenosis following arterial reconstruction highlight the need to create and develop new biocompatible materials capable of actively regulating reparative processes — in particular, suppressing neointimal hyperplasia without impairing complete endothelialization. Tissue engineering approaches enable to develop the matrices providing structural support, hemocompatibility, and controlled release of bioactive molecules [5–7]. The natural polymer silk fibroin (SF) is a promising material for producing tissue-engineered patches by electrospinning. It is biocompatible, has low immunogenicity, undergoes controlled degradation, and combines well with other polymers [8–10]. Combining SF with the synthetic polymer — thermoplastic polyurethane (PU) — can provide the resulting composite material with greater elasticity and strength while preserving its capacity for complete remodeling [10].

However, even biocompatible materials have been associated with the activation and proliferation of smooth muscle cells followed by neointimal hyperplasia leading to stenosis of the arterial vessel at the site of implantation of the artificial patch [11]. Therefore, to overcome the problem of neointimal hyperplasia, it is of current interest to use antiproliferative agents such as sirolimus (SRL) — an mTOR kinase inhibitor used in oncology and transplantation medicine, and as a stent coating [12]. It is expected that the sustained local release of SRL from polymer-matrix fibers can suppress smooth muscle cell hyperproliferation and inflammation without impairing endothelialization of the inner surface of the implanted patches. However, the literature contains insufficient data on the comprehensive evaluation of the efficacy of such materials as patches for arterial wall reconstruction.

The aim of the study was to fabricate polymeric patches based on a blend of regenerated SF and thermoplastic PU with SRL added and to investigate their surface structure, physical and mechanical properties, and in vitro biocompatibility.

Materials and Methods

Fabrication of polymeric patches. Polymeric patches were fabricated by electrospinning using a 10% solution in 1,1,1,3,3,3-hexafluoroisopropanol (CJSC “PiM-Invest,” Russia) based on a blend of regenerated SF and thermoplastic PU Tecoflex-EG 80A (Lubrizol, USA) in a mass ratio of 2:1 [10, 13]. SRL (C17314651; Shanghai Macklin Biochemical Technology Co., Ltd., China) was preliminary dissolved in dimethyl sulfoxide (Am-0231-0.1; Helicon, Russia) and then added to the electrospinning solution at final concentrations of 0.175% (0.35 mg/cm²), 0.35% (0.7 mg/cm²), 0.525% (1.04 mg/cm²), and 0.7% (1.39 mg/cm²) — the values in parentheses representing the calculated final SRL content in the polymer matrices after electrospinning. A composite SF/PU-matrix without SRL was used as the control. Electrospinning was performed using a NANON-01A apparatus (MECC CO, Japan) under the following conditions: needle outlet diameter — 22G, distance to the collector — 15 cm, collector rotation speed of 200 rpm, solution feed rate — 1.6 ml/h, voltage — 26 kV. A metal pin, 8.0 mm in diameter, was used as the collector.

The polymer patches removed from the pins were sterilized along with simultaneously performed cross-linking of the polymers in the material using gamma irradiation on an ILA-10 (Impulse Linear Accelerator) unit with a beam energy of 5 MeV and a power of 50 kW (Institute of Nuclear Physics, Siberian Branch of the Russian Academy of Sciences, Russia) at a total absorbed dose of 15 kGy.

Ultrastructural examination of the surface of the polymer patches. The ultrastructure of the polymer-patch surface was examined using scanning electron microscopy (SEM). The 0.5×0.5 cm specimens were coated with a 15-nm-thick Ag–Pd layer using an EM ACE 200 system (Leica Microsystems GmbH, Austria). The patch surfaces were examined with an S-3400N scanning electron microscope (Hitachi, Japan) at an accelerating voltage of 10 kV under high-vacuum conditions. Fiber diameter and pore size were measured in the SEM images obtained.

Study of the physical and mechanical properties of the polymer patches. The physical and mechanical properties of the polymer patches were studied in accordance with GOST 270-75 under uniaxial tensile loading using a Z-series universal testing machine (Zwick/Roell, Germany). A 50-N nominal-force load cell was used, the crosshead speed being 50 mm/min. There were 6 specimens in each group; the specimens were cut in the longitudinal direction of the patches. The specimens were prewetted with distilled water before testing. The following parameters were measured: maximum tensile stress (MPa), force applied to the specimen until failure (Fmax, N), relative elongation corrected with respect to the specimen failure pattern (%), and Young’s modulus (E, MPa).

The physical and mechanical parameters of the polymer patches were compared with those of native arteries: the sheep carotid artery (a. carotis) and the human internal thoracic artery (a. mammaria), which was taken during coronary artery bypass grafting from the patients who had provided voluntary informed consent for tissue collection.

The study was approved by the local ethics committee of the Research Institute for Complex Issues of Cardiovascular Diseases (protocol No. 3 dated March 16, 2026).

Evaluation of the hydrophilicity of the polymer-patch surface. Surface hydrophilicity of the polymer patches was evaluated by measuring the material water contact angle (the sessile-drop method) and using a camera at room temperature. The contact angle was calculated from six images for each type of polymer patch using ImageJ software (NIH, USA).

Study of the hemocompatibility of the polymer patches. The hemocompatibility of the polymer patches was studied based on the degree of erythrocyte hemolysis and platelet aggregation.

The degree of erythrocyte hemolysis was assessed after the contact of the polymer-patch specimens with fresh citrated donor blood using the method described previously [14]. Six specimens of each material type were tested. The results were expressed as the percentage of hemolysis.

Platelet aggregation after contact between the test polymer-patch specimens and donor blood plasma was evaluated in accordance with ISO 10993.4 using the previously described method [14]. Six specimens of each material type were tested. The maximum percentage of platelet aggregation was recorded.

In vitro study of biocompatibility of polymer patches. The biocompatibility of the polymer patches was investigated in vitro by assessing the cell density and proliferative activity of Ea.hy 926 endothelial cells cultured on the material surface, as well as by evaluating real-time cytotoxicity while endothelial cells were cultured in the presence of the material using a cell analyzer.

For all cell culture procedures, we used DMEM/F12 medium (11320033; Thermo Fisher Scientific, USA) supplemented with HAT (H0262; Sigma-Aldrich, USA), 10% fetal bovine serum (26140079; Thermo Fisher Scientific, USA), antibiotics (0378016; Thermo Fisher Scientific, USA), and amphotericin B (15290018; Thermo Fisher Scientific, USA). Ea.hy 926 cells (30,000 cells/well) were cultured on the test specimens (circular fragments 1.6 cm in diameter; n=6 per group) affixed to the bottoms of the wells of a 24-well plate under sterile conditions in a CO₂ incubator for 72 h. In the control group, the cells were cultured on the well bottom without specimens.

Density assessment of the cells adhered to the patches. After 72-hour culture, the cell density on each specimen was assessed by fluorescence microscopy. The cells were stained with the nuclear dye Hoechst 33342 (10 µg/ml; 14533; Sigma-Aldrich, USA; it stains the nuclei of all cells blue). The adherent cells were counted using an Axio Observer Z1 inverted microscope (Carl Zeiss, Germany). The relative number of adherent cells was calculated per 1 mm² of the matrix.

Assessment of the proliferative activity of adherent cells. The proliferative activity of the cells adhered to the polymer patch specimens was assessed using the Click-iT™ Plus EdU Cell Proliferation Kit for Imaging (C10637; Thermo Fisher Scientific, USA). After 72-hour preliminary incubation on the test material, the cells were stained according to the manufacturer’s instructions and then additionally stained with the nuclear dye DAPI (10 µg/ml; D9542; Sigma-Aldrich, USA). The specimens were analyzed using an LSM700 scanning confocal microscope (Carl Zeiss, Germany). The relative number of proliferating cells was calculated using the following formula: (number of proliferating cells in the field of view / total number of cells in the field of view) · 100.

Real-time cytotoxicity analysis. The present study was conducted using the xCELLigence RTCA DP system (ACEA Biosciences, USA). DMEM/F12 culture medium was added to the wells of specialized E-Plate 16 Ins ert plates (ACEA Biosciences, USA), and the baseline impedance was measured. Ea.hy 926 endothelial cells were then added to the wells at a density of 10,000 cells/well and incubated for 1 h to let the cells settle and attach. The cell index was automatically recorded every 15 min. Subsequently, two preprepared 0.5×0.5 cm matrix samples of each type were placed into the E-Plate 16 Ins ert wells (ACEA Biosciences, USA). The inserts were placed in to the plates containing the cells. The control consisted of wells containing the cells cultured in the same volume of culture medium. The cells were cultured with the samples for 72 h. The cell index was automatically recorded every 30 min. The cytotoxic effects of the test samples were evaluated based on the changes in the cell index dynamics.

Statistical analysis. The findings were statistically analyzed using GraphPad Prism software (USA). The normality of the data distribution was assessed using Shapiro–Wilk test. For evaluating the differences among three or more groups, the nonparametric Kruskal–Wallis test was applied (with Tukey post hoc test); for pairwise comparisons we used Dunn’s test. The differences were considered statistically significant if p<0.05. The findings were presented as the median and 25^th and 75^th percentiles — Me [Q1; Q3].

Results

Ultrastructure of the polymer patches. Scanning electron microscopy revealed that regardless of the incorporated SRL concentration, the surface structure of the SF-PU-SRL patches involved randomly interwoven flat fibers forming the pores of varying diameters (Figure 1; see the Table). Isolated irregular areas of the matrices were formed by fused polymer fibers.

Figure 1. SEM images of the SF-PU-SRL polymer patches with different SRL content: (а) SF-PU-SRL 0.175; (b) SF-PU-SRL 0.35; (c) SF-PU-SRL 0.525; (d) SF-PU-SRL 0.7. SEM — scanning electron microscopy; SF — silk fibroin, PU — polyurethane, SRL — sirolimus

Mean fiber and pore diameters and the surface porosity of SF-PU-SRL polymer patches, Me [Q1; Q3]

Polymer patch type	Fiber diameter (µm)	Pore diameter (µm)	Porosity (%)
SF-PU	2.19 [1.80; 2.60]	12.5 [8.5; 21.4]	50.1
SF-PU-SRL 0.175	3.54 [2.10; 4.0]	20.5 [16.4; 28.2]	43.9
SF-PU-SRL 0.35	3.75 [2.80; 5.90]* (р=0.015)	19.8 [16.6; 26.4]	43.3
SF-PU-SRL 0.525	2.68 [2.10; 3.40]	16.6 [8.9; 20.9]	45.1
SF-PU-SRL 0.7	3.18 [2.60; 4.10]	17.6 [13.2; 21.9]	44.1

The fiber diameter of the SF-PU-SRL polymer patches was, on average, 1.5 times greater compared with SF-PU composition; for SF-PU-SRL 0.35, the difference was statistically significant (p<0.05). Between-group differences in fiber diameter were not statistically significant (p>0.05). In terms of fiber diameter, the SF-PU-SRL 0.525 patch was the closest to the SF-PU control matrix (see the Table).

The pore diameter of the SF-PU-SRL patches also exceeded that of the SF-PU control material by an average of 1.5 times. The mean pore diameter had no significant difference depending on SRP content in the patches. Generally, SRL added to the polymer patches reduced the material porosity from 50.1% (SF-PU) to 43.3% (SF-PU-SRL 0.35); however, no linear relationship was revealed between porosity and SRL content (see the Table).

Thus, SRL added to the polymer patches caused no significant changes in their ultrastructure. At the same time, the SF-PU-SRL 0.525 polymer patch was the most similar in ultrastructure to the SF-PU comparison patch.

Physical and mechanical properties. The strength of the polymer patches was evaluated by measuring stress (ultimate strength) and the force applied to the specimen before the onset of failure (Fmax). The analysis of stress showed the patches containing SRL to exceed the comparison materials in this parameter; SF-PU-SRL 0.175 and SF-PU-SRL 0.35 demonstrated higher stress values than both native arteries (p<0.05; Figure 2 (a)). However, the SF-PU-SRL 0.525 and SF-PU-SRL 0.7 specimens showed virtually no difference in this parameter compared with the a. mammaria (see Figure 2 (a)). In addition, the Fmax of the polymer patches with these compositions was 3.5 times higher than that of the a. mammaria (p<0.05; Figure 2 (b)). All polymer patches had tensile strengths comparable to that of the a. carotis (see Figure 2 (b)).

Figure 2. Physical and mechanical properties of the polymer patches: (а) stress; (b) force applied to the specimen before the onset of its failure (Fmax); (c) relative elongation; (d) Young’s modulus (E). SF — silk fibroin, PU — polyurethane, SRL — sirolimus, a. carotis (ovine) — sheep carotid artery, a. mammaria — human internal thoracic artery

The polymer patches demonstrated greater relative elongation than the a. mammaria; for SF-PU-SRL 0.525, the difference was statistically significant (p<0.05; Figure 2 (c)). In contrast, the a. carotis exhibited greater elasticity than both: the polymer patches (particularly, SF-PU-SRL 0.7 (p<0.05)) and the a. mammaria itself (see Figure 2 (c)).

The analysis of Young’s modulus showed the SF-PU-SRL 0.175 and SF-PU-SRL 0.35 patches to be significantly stiffer than both native arteries (p<0.05) and the SF-PU comparison material (Figure 2 (d)). Conversely, the SF-PU-SRL 0.525 and SF-PU-SRL 0.7 compositions were not different from the a. mammaria, indicating the greater similarity to the native arteries (see Figure 2 (d)).

Ultimately, the analysis of the complex of physical and mechanical characteristics showed the SF-PU-SRL 0.175 and SF-PU-SRL 0.35 polymer patches to have increased strength along with excessive stiffness compared with native vessels. At the same time, all polymer patches were more elastic than the human internal thoracic artery (a. mammaria). Thus, from the standpoint of mechanical matching to the native arteries, the SF-PU-SRL 0.525 and SF-PU-SRL 0.7 polymer patches were optimal.

Hydrophilicity. The hydrophilicity of the polymer patches was studied by measuring the contact angle using the sessile-drop method. SRL added to the polymer matrix increased the contact angle and, consequently, reduced the hydrophilicity of the material. No statistically significant differences between the groups were found (p>0.05); however, the SF-PU-SRL 0.525 matrix demonstrated the highest contact angle, whereas the SF-PU-SRL 0.7 patch had the lowest val ue (Figure 3).

Figure 3. Contact angle of the polymer patches. SF — silk fibroin, PU — polyurethane, SRL — sirolimus

Hemocompatibility. The polymer patches caused virtually no erythrocyte hemolysis after the contact with fresh citrated blood. The degree of erythrocyte hemolysis was equal to “0” in all study groups.

Maximum platelet aggregation after the contact between the polymer patches and blood plasma was assessed. The following values were obtained for the materials under study: platelet-rich plasma, 80.0 [79.4; 85.0]%; SF-PU, 87.1 [81.4; 91.1]%; SF-PU-SRL 0.175, 84.2 [82.0; 86.7]%; SF-PU-SRL 0.35, 85.9 [81.0; 89.8]%; SF-PU-SRL 0.525, 87.6 [83.6; 89.1]%; and SF-PU-SRL 0.7, 83.0 [81.3; 88.1]%.

The evaluation of the ability of the patches to induce platelet aggregation after the contact with blood plasma showed no statistically significant differences between the different types of patches or compared with platelet-rich plasma.

Thus, all polymer patches under study demonstrated equally optimal hemocompatibility properties, which is important for materials intended for long-term contact with blood.

Biocompatibility of the polymer patches. The study of the matrix properties of the polymer patches included the assessment of the number of adherent Ea.hy 926 endothelial cells on the material surface (cells/mm²). The cell density on all polymer patch samples was 10-fold lower than on the control plastic (p<0.05). The highest cell density was observed on the SF-PU-SRL 0.35 sample, whereas the values in the other groups were comparable and did not exceed an average of 10³ cells/mm² (Figure 4 (a)).

Figure 4. Biocompatibility parameters of the polymer patches: (а) cell density; (b) percentage of proliferating nuclei; (c) cell index; (d) cell doubling time. SF — silk fibroin, PU — polyurethane, SRL — sirolimus

Since the material contained SRL, which has antiproliferative activity, it was important to evaluate the proliferative potential of each type of polymer patch. The proliferation of endothelial cells on the surface of all polymer patch samples, including the original SF-PU matrix, was lower than on the surface of the culture plastic (p<0.05; Figure 4 (b)). As expected, SRL added to the polymer patches reduced the relative number of proliferating cells: in the SF-PU-SRL 0.175 and SF-PU-SRL 0.35 samples, it decreased to 8 and 4%, respectively, whereas the cell proliferation ceased on the SF-PU-SRL 0.525 and SF-PU-SRL 0.7 patches (see Figure 4 (b)).

Real-time cytotoxicity analysis showed that, in the presence of patches containing SRL, the cell index of Ea.hy 926 endothelial cells was lower than that of endothelial cells cultured in the presence of the SF-PU control patch (Figure 4 (c)). No significant differences in patch cytotoxicity were detected depending on the concentration of SRL incorporated into the patches. As expected, the cell population doubling time was shorter both in the control wells without patch-containing inserts and in the wells containing the inserts with patches without sirolimus (Figure 4 (d)). Cell population doubling in the groups with SRL-containing patches took 1.7–2.0 times longer, confirming the antiproliferative activity of SRL incorporated into these patches (p>0.05; see Figure 4).

Thus, based on the obtained data, the SF-PU-SRL 0.525 patches demonstrated optimal antiproliferative properties at the lowest SRL concentration, exerting no cytotoxic effects on endothelial cells, making them promising for further investigation.

Discussion

The problem of neointimal hyperplasia and the subsequent restenosis in the area of vascular implants is highly relevant, since it affects the outcome and efficacy of the initial surgical intervention [15]. A modern strategy for controlling neointimal hyperplasia may involve the development of polymeric vascular patches with antiproliferative properties. Such approach would enable locally affect the pathological process minimizing systemic adverse effects.

We previously demonstrated a matrix based on an SF and PU polymer composition to be a promising platform for developing vascular patches because of its optimal physical and mechanical properties, which are similar to those of native arteries; its porous fibrous structure, which is favorable for the colonization by endogenous cells; and its high hydrophilicity and thromboresistance [10]. Therefore, imparting antiproliferative properties to this composite material appears promising for controlling postoperative neointimal hyperplasia at the site of vascular wall reconstruction.

SRL added to the composite polymer matrix did not substantially alter its ultrastructure. It resulted in a slight increase in the diameter of the polymer fibers compared with the SF-PU matrix and also reduced matrix porosity. These features predictably affected the physical and mechanical properties of the patches: the SF-PU-SRL 0.35 sample, which had the greatest fiber thickness and the lowest porosity, demonstrated the greater material strength and stiffness, whereas the SF-PU-SRL 0.525 sample, whose ultrastructure differed least from that of the SF-PU comparison matrix, had lower stiffness and, based on the overall elastic and deformation characteristics, more closely matched the native arteries.

The presence of SRL in the polymeric patches had no adversely effects on the hemocompatibility of the samples under study, which is critically important when developing materials intended for prolonged contact with blood. We previously demonstrated the absence of pronounced red blood cell hemolysis and platelet aggregation upon blood contact for monocomponent matrices fabricated by electrospinning from regenerated SF and thermoplastic PU [14, 16]. At the same time, according to the literature, SRL itself can enhance ADP-induced platelet aggregation at high concentrations (50 and 100 ng/ml) [17, 18]. In our study, the incubation of the samples, even those with the highest SRL content of 1.393 mg/cm² (SF-PU-SRL 0.7), did not result in such a high local drug concentration, and therefore had no effect on platelet aggregation. Similarly, in studies [19–21] devoted to the in vitro evaluation of the hemocompatibility of stents with SRL-loaded polymer coatings, the authors reported no increase in red blood cell hemolysis or platelet aggregation, which is consistent with our findings.

The biocompatibility of the developed polymeric patches warrants separate discussion. The need to impart antiproliferative properties to the patches required addressing two objectives: achieving optimal material biocompatibility, including the absence of cytotoxicity toward endothelial cells in order to preserve the potential for complete endothelialization; and ensuring the ability to inhibit the proliferation of cells cultured on the sample surface to prevent neointimal hyperplasia. Overall, the SF-PU surface did not exhibit high adhesive capacity, which was also characteristic of the SRL-containing polymeric patches. This may have been related to the specific microtopography of the matrices [22]. However, the biocompatibility study using noncontact co-culture of the patches with endothelial cells demonstrated no pronounced cytotoxicity and no significant differences between the matrices with different SRL contents.

The evaluation of the antiproliferative activity of the SRL-containing polymeric patches showed that complete inhibition of the proliferation of endothelial cells cultured on the surface was achieved with the SF-PU-SRL 0.525 matrix that makes it optimal in terms of functional activity while containing the minimum amount of the drug.

Conclusion

The present study conducted a comprehensive comparative assessment of the ultrastructure, physical and mechanical properties, hemocompatibility, and biocompatibility of the polymer patches fabricated by electrospinning a blend of SF, PU, and SRL at four concentrations: 0.175, 0.35, 0.525, and 0.7%.

The SF-PU-SRL 0.525 polymer patch appeared to be the most promising due to its highly porous surface structure, the closest match between its physical and mechanical properties and those of the native arteries, high hemocompatibility, and antiproliferative activity without cytotoxicity. This enables to proceed to evaluating its efficacy in preventing postoperative neointimal hyperplasia and restenosis in an arterial wall reconstruction model.

Study funding. The study was funded by a grant from the Russian Science Foundation No.25-25-00791, https://rscf.ru/project/25-25-00791/.

Conflict of interest. The authors declare no conflict of interest.