Characterization of a PEG-DE cross-linked tubular silk scaffold

Abstract

The secondary structure and compliance of a novel small caliber (≤6 mm) silk fibroin (SF) tubular scaffold (SFTS) were investigated. Imitating the structure of natural vascular tissue, the SFTS consisted of a silk knit as the medium and a poly(ethylene glycol) diglycidyl ether (PEG-DE) cross-linked silk fibroin (SF) membrane as the intimal and adventitial layers, integrated to form a porous tissue. FTIR and XRD results showed that PEG-DE could induce SF molecules to form β-sheets during the cross-linking reaction process, resulting in improved crystallinity. As a result of the silk knit medium the SFTS had excellent mechanical properties. The intimal layer, which is in contact with a continuous flow of blood, must have adequate compliance. The results showed that the intimal layer of the SFTS had good stress-strain resistance when combined with the silk knit medium. When the SFTS was prepared with 6% SF, its axial breaking strength was >62 kPa and breaking elongation could reach about 33%; the circumferential breaking strength was >10 MPa and breaking elongation was >18%. The results of compression testing showed that the radial compression resilience of SFTS reached 94%, which was a significant improvement on commercial artificial blood vessels prepared from Dacron.

Keywords

Silk fibroin PEG-DE molecular conformation crystal structure stress-strain compression elasticity

Cardiovascular disease is a major threat to human health, especially since in most people the original function of vascular lesions cannot be restored, even after drug treatment. Many cases result in a life-threatening crisis, due to the limited availability of natural blood vessels derived from autografts or allografts. As a result there has been increasing research into the use of vascular prostheses for vascular tissue repair.¹ However, there is a scarcity of medical materials suitable for clinical use in vascular prostheses. Synthetic polymers such as Dacron,² expanded polytetrafluoroethylene (e-PTFE),³ and polyurethanes⁴ have been used to construct large and medium caliber artificial blood vessels for clinical application.⁵ However, these synthetic grafts are not suitable for clinical use in small caliber (≤6 mm) vascular grafts, and they can contribute to the development of thrombosis.⁶ Surface modification and treatment of these materials is required on account of their poor blood compatibility and histo-compatibility.^7–9

SF from silkworms such as Bombyx mori is a natural protein which has been used to create an extracellular matrix,¹⁰ as a drug-release carrier¹¹ and to form tissue engineering scaffolds.¹² Many investigators have shown that SF has good adhesive properties and is compatible with many types of cell, including fibroblasts,¹³ epithelial cells,¹⁴ bone marrow mesenchymal stem cells,¹⁵ and vascular endothelial cells.¹⁶ It has been reported that gel spinning,¹⁶ dipping,¹⁷ and electrospinning techniques^18–20 can be utilized to prepare tubular SF scaffolds for small caliber vascular grafts. However, most studies have focused only on exploring the preparation parameters and biocompatibility, while the mechanical properties were either unsatisfactory or were neglected. The smooth muscle which forms the middle layer of the blood vessel, the tunica medium, endows it with important biomechanical properties such as stress deformation responsiveness, pressurized burst strength, suture retention strength and compression elasticity, and these allow vasoconstriction, vasodilation, and continual hemokinesis.

Differences in material properties and consequent mismatch at the anastomoses have been shown to cause intimal hyperplasia and a reduction in the patency rate.²¹ As previously mentioned, the technique generally used to prepare Dacron vascular prostheses is based on knitted or braided fabric.²² Most other synthetic materials used to create prosthetic vessels, such as polyglycolide (PGA),²³ e-PTFE,²⁴ poly(trimethylene carbonate-co-l-lactide),²⁵ and methacrylates²⁶ are usually processed using spinning technology. They can be used successfully for large diameter artificial blood vessels, but are unsuitable for small caliber vessels due to their poor biocompatibility^7,9 and/or inappropriate mechanical characteristics.^27–28 Silk fibers and silk knit have been reported as being fabricated as small caliber vessels showing good tensile properties.^29–30

We have designed and prepared a bionic SFTS, in three layers to mimic the natural organizational structure of blood vessels. A degummed silk braided tube was used as the middle layer covered with a PEG-DE cross-linked porous regenerated SF. PEG-DE is a flexible chain polymer, and is commonly used to modify biological material. We have analyzed the molecular conformation and crystal structure of a PEG–DE cross-linked regenerated SF membrane, and studied the stress deformation responsiveness and compression elasticity of the regenerated SF membrane integrated with a silk braided tube.

Experimental details

Preparation of SF solution and tubular silk knit

B. mori silk fibroin solution was prepared as previously described.³¹ In brief, B. mori raw silk was first degummed to remove sericin by boiling for 30 min in deionized water containing 0.06 wt% Na₂CO₃, then rinsing thoroughly in warm deionized water. This procedure was repeated three times. After drying, the silk fiber was stirred in the ternary solution CaCl₂·CH₃CH₂OH·H₂O (molar ratio = 1:2:8) at 70 ± 2℃ until completely dissolved. Finally, an aqueous solution of regenerated SF was obtained by dialyzing against distilled water for four days at 4℃, and then concentrating and filtering the mixed solution. The final concentration of the SF aqueous solution was approximately 6–10% (w/v). A silk fabric was braided by twisting 24 shares of degummed threads, which were twisted into 2 × 20/22 denier yarns, and 3 × 20/22 denier raw silk, on a braiding machine (Shanghai Hakao, China) at a speed of 160 revolutions/min, and used to obtain a silk-knitted tube.

Fabrication of SFTS

The tubular silk knit was mounted on a mold,³² sprayed with freshly-mixed solutions of SF (4%, 5%, and 6%) and PEG–DE (Sigma, SF:PEG–DE (w/w) = 1.0:0, 1.0:0.2, 1.0:0.5, 1.0:0.8, 1.0:1.0, and 1.0:1.2) from the internal to the external surface, then frozen (−20℃ and −40℃) and thawed, to obtain the SFTS. The scaffold was then immersed in deionized water at 4℃ for four days to remove residual reactants. After air-drying at room temperature, the SFTS was ready for further evaluation. The inner diameters of the core column, silk fabric, and polypropylene pipe were approximately 3, 4 and 5 mm, respectively.

Structural assay

Photomicrographs of the SFTS cross-section and internal surface were obtained by scanning electron microscopy (SEM, Hitachi S–570, Tokyo, Japan). The internal and external surfaces were carefully stripped from the SFTS and cut into micro-particles. After filtering through a 300-mesh sieve, samples were mounted in KBr pellets for FTIR analysis using a Nicolet Avatar–IR360 (Thermo Fisher Scientific, Waltham, MA). The crystal structure of all the SF membranes was determined using an X-ray diffractometer (XRD, Mercury CCD, Tokyo, Japan) at a tube voltage of 40 kV, tube current of 40 mA, and a scanning speed of 2°/min. The diffraction intensity curve was recorded by scanning the 2θ range of 5–45°. The crystallinity was calculated using Peakfit v4.12 software through separate peak-fitting calculation.

Mechanical properties

The tensile strength of the SFTS was designed in accordance with ISO7198, Cardiovascular implants–tubular vascular prostheses, and confirmed using a tensile tester (YG065, Laizhou electronic instrument Co Ltd, Laizhou, China). Circumferential tensile and axial tensile tests were performed at 100 mm/min of tensile speed using two different grips. Sample sizes were 60 mm for the axial tensile test and 10 mm for the circumferential tensile test. Axial tension of the tubular silk knit was measured as the control, but a circumferential tensile test could not be performed due to the yarns falling out of the knit.

The compression elasticity of SFTS was measured on a compression tester (LLY–06D, Laizhou Electronic Instrument Co Ltd). Wet samples 30 mm in length were measured using a 20 mm diameter load plate at 0.05 cN initial strain at a head speed of 100 mm/min. The compression distance was half of the SFTS external diameter. The compress and reply stagnation times were set at 5 s and 30 s, respectively. A commercial Dacron (YZB/1352–2008, 8 mm diameter, Shanghai Chest Medical Technology Co, Shanghai, China) vascular prosthesis was tested as control.

Results and discussion

Structure of the SFTS

The SFTS formed a good shape with none of the silk knit exposed. The medial layer of the SFTS exhibited a netted structure formed from the silk knit, which was spread out as a stiffener (Figure 1(a)). Morphology and pliability were satisfactory (Figure 1(b)), with wall thickness and inner diameter of the permeated SF of approximately 1.0 mm and 3.5 mm, respectively. SEM of an SFTS cross-section (Figure 1(c)) showed that the regenerated SF matrix permeated from the internal surface to the external surface. Interconnected pores are necessary for material exchange between the inside and outside wall, and for ingrowth of cells such as smooth muscle cells or fibroblasts. The internal surface of the SFTS incorporated uniformly distributed micropores (Figure 1(d)), the formation of which could be controlled by a number of factors, including SF concentration and cross-linker proportion.

Figure 1.

Morphology of SFTS showing (a) tubular silk knit [32], (b) morphology of the bent SFTS, (c) SEM of the SFTS cross-section and (d) SEM of the SFTS internal surface.³² SFTSs were prepared by 4% SF and SF:PEG–DE = 1.0:1.0 at −20℃, the bar of c is 1.0 mm and d is 500 µm.

FTIR analysis of the SFTS surface membranes

The molecular conformation of the regenerated SF prepared from the aqueous solution was mainly as a random coil (1250 cm⁻¹ amide III and 665 cm⁻¹ amide V) and an α-form (Silk I, 1660 cm⁻¹ amide I and 1522 cm⁻¹ amide II), as shown in Figure 2. The characteristic absorption bands created by β-sheets (Silk II) appeared in the PEG–DE cross-linked porous SFTS membranes (Figure 2(a)–(e)) and were the same as those observed in an ethanol-treated SFTS membrane (Figure 2(f)). The spectral curves showed that there was no marked change in the secondary structure of SFTSs produced from different mass ratios of SF and PEG–DE. The double epoxide groups of PEG–DE can chemically react with –NH₂, –COOH, or –OH, which are present in the silk fibroin molecular chain. During the cross-linking reaction, the characteristic peak of amide I shifted from 1660 cm⁻¹ to 1627 cm⁻¹, amide III from 1250 cm⁻¹ to 1229 cm⁻¹ and amide V from 665 cm⁻¹ to 700 cm⁻¹, compared with the non-ethanol-treated SFTS membrane (Figure 2(g)). This result indicated that PEG–DE induced a change in the molecular conformation of the SF during SFTS manufacture.

Figure 2.

FTIR spectra of PEG–DE-cross-linked SFTS membranes created from (a) SF:PEG–DE at 1.0:0.2, (b) SF:PEG–DE at 1.0:0.5, (c) SF:PEG–DE at 1.0:0.8, (d) SF:PEG–DE at 1.0:1.0, (e) SF:PEG–DE at 1.0:1.2, (f) ethanol-treated SFTS membrane, and (g) SFTS membrane without ethanol treatment.

XRD analysis of the SFTS surface membranes

Although the FTIR peak shifts were unremarkable, crystalline transition of SF occurred during the cross-linking process. In all the PEG–DE cross-linked SFTS membranes, significant characteristic peaks appeared at about 9.1°, 20.6° and 24.3° assigned to Silk II (Figure 3(a)–(e)), similar to ethanol-treated SF film (Figure 3(f)). The Silk II content and crystallinity of the SFTS membranes increased with increasing PEG–DE content when SF:PEG–DE was below 1.0:0.8. At SF:PEG–DE = 1.0:0.8, the Silk II content was 16.6% and the approximate crystallinity 26%, similar to the ethanol-treated SFTS membrane (Table 1).

Figure 3.

XRD spectra of PEG–DE-cross-linked SFTS membranes created from (a) SF:PEG–DE at 1.0:0.2, (b) SF:PEG–DE at 1.0:0.5, (c) SF:PEG–DE at 1.0:0.8, (d) SF:PEG–DE at 1.0:1.0, (e) SF:PEG–DE at 1.0:1.2, (f) ethanol-treated SFTS membrane, and (g) SFTS membrane without ethanol treatment.

Table 1.

Approximate crystallinity of SFTS membranes

Samples	SF:PEG-DE					Ethanol treatment	No ethanol treatment
Samples	1.0:0.2	1.0:0.5	1.0:0.8	1.0:1.0	1.0:1.2	Ethanol treatment	No ethanol treatment
SilkI (%)	11.6	8.5	9.4	8.3	10.4	11.4	12.0
SilkII (%)	12.2	14	16.6	15.1	13.9	18.9	0.3
Crystallinity (%)	23.8	22.5	26	23.4	24.3	30.3	12.3

Concerning the XRD analysis of SF from aqueous solution, conformations other than random coil, and a low crystallinity in the Silk I crystalline form have been considered (Figure 3(f) and Table 1). Part of the random coil or Silk I was transformed into Silk II after ethanol treatment. As shown in Figure 3, another significant characteristic peak at about 40.2°, assigned to Silk I, appeared in all blend membranes, and this was absent in the non-ethanol-treated SFTS membrane (Figure 3(g)). These results show that PEG–DE could induce increased translation from random coil, or Silk I to Silk II, and changes in random coil translation into Silk I, thus promoting SF crystal formation.

Stress–strain of the SFTS

The SFTSs possessed excellent tensile properties, as shown in Table 2. The axial tensile strength was 640∼760 kPa and the circumferential tensile strength was >12 MPa. While the axial tensile strength of SFTS was lower than that of the reported gel spinning silk tubers, this was possibly due to the difference in measurement method. The axial elongation was about 60% and the circumferential elongation was more than 300%, significantly higher than the gel spinning silk tubers and electrospun silk fibroin tubes.^18,33 The SF concentration had no obvious effect on the tensile properties; however, lower freezing temperature was beneficial to the increase in tensile strength, particularly the circumferential tensile strength.

Table 2.

Tensile properties of the SFTSs (SF:PEG–DE = 1.0:1.0)

SF concentration (%)		4		5		6
T (℃)		−20	−40	−20	−40	−20	−40
Axial	Tensile strength (MPa)	0.64 ± 0.03	0.67 ± 0.02	0.69 ± 0.04	0.76 ± 0.05	0.63 ± 0.03	0.76 ± 0.04
Axial	Elongation (%)	65.12 ± 6.24	54.67 ± 9.23	67.80 ± 6.80	60.75 ± 8.29	62.20 ± 8.56	60.35 ± 4.58
Radial	Tensile strength (MPa)	12.08 ± 0.57	18.67 ± 0.67	13.43 ± 0.56	18.60 ± 1.28	14.58 ± 0.43	21.55 ± 0.75
Radial	Elongation (%)	369 ± 15.65	369 ± 12.92	364 ± 9.92	354 ± 13.05	399 ± 10.03	365 ± 14.63

Similar to native vessels, tissue-engineered small diameter vascular grafts should have the ability to sense hemodynamic loading during blood flow and adapt through mechanical and biological responses.^34–35 On further analysis of the tensile curve, a set of segmented lines with different slopes (Figure 4(a)) can be seen in the circumferential load-elongation curve of the SFTS. The a→b segment in the load-elongation curve represents breakage of the 3-D porous regenerated SF membrane covering the tubular silk knit. The integrity of the SFTS was destroyed by breakage of the inner and outer layers. The b→c segment shows the continued stretching of the fabric coils. The tensile fracture process of the silk yarn is shown in segment c→d. In the SFTS axial load-elongation curve, two conspicuous straight line segments with different slopes were observed, and are seen in Figure 4(b). The first stage (a→b) in the load-elongation curve represents breakage of the regenerated SF membrane, accompanied by stretching of the fabric coils (a→b′). The b→c segment shows the fracture process of the silk yarn.

Figure 4.

Representative radial (a) and axial (b) load-elongation curves of the SFTSs. SFTSs were prepared by 5% SF and SF:PEG–DE = 1.0:1.0 at −20℃, abcd and abc were the tensile curves of SFTSs, ab′c was a tensile curve of tubular silk knit.

An artificial blood vessel must have an inner surface that is not easily ruptured, as this would be detrimental to endothelialization and prone to blood infiltration and coagulation. We focused on tensile strength at the a→b stage in the axial tensile and circumferential tensile test curves. Figure 5 shows that both axial tensile strength and elongation were similar for all the SFTS samples manufactured under different conditions, with a tensile strength of approximately 62–66 kPa and a breaking elongation in the range 25–33%. Figure 6 shows that the circumferential tensile strength of a SFTS membrane prepared from 6% SF was approximately 10–14.5 MPa and the breaking elongation approximately 18%. The circumferential tensile strength increased with increasing SF concentration, and also with decreasing freeze temperature. Tensile tests were also performed on scaffolds manufactured without silk fabric in the mold, but these failed since their poor mechanical properties meant that they could not withstand the experimental conditions required by the ISO standard. These results imply that the good tensile properties of the SFTS membrane resulted from the synergistic behavior of the braid. These tensile test results demonstrate that the SFTS would provide satisfactory compliance for continual hemokinesis.

Figure 5.

Axial tensile properties at the first stage (a→b) of the SFTSs (SF:PEG–DE = 1.0:1.0).

Figure 6.

Radial tensile properties at the first stage (a→b) of the SFTSs (SF:PEG–DE = 1.0:1.0).

Compression elasticity of the SFTS

During blood circulation, vasoconstriction and vasodilation occur at a frequency corresponding to the heartbeat, and blood vessels therefore need excellent compression elasticity and flexibility. As shown in Figure 7, the compression elasticity of the SFTS showed satisfactory rebound performance with an elastic recovery rate commonly as high as 90%, significantly better than the commercial Dacron vascular prosthesis, which had only about 60% elastic recovery rate. In addition, the plastic deformation rate of the present SFTS was significantly less than the control (Table 3). These data show not only that the medial silk knit had elastic properties, but also that the regenerated porous structural SF membrane (no corresponding layer in the control) enhanced elasticity. The results of our compression elasticity tests demonstrate that an SFTS would provide satisfactory vasoconstriction and vasodilation.

Figure 7.

Elastic recovery rate of the SFTS (SF:PEG–DE = 1.0:1.0).

Table 3.

Comparison of elasticity of the SFTSs (SF:PEG–DE = 1.0:1.0) with Dacron

Samples	SFTSs	Dacron
Elastic recovery rate (%)	79.31~95. 57^a	59.93 ± 6.90
Plastic deformation rate (%)	2.63~7.34^a	10.67 ± 1.43

^aData for SFTSs were from all samples, regardless of conditions of manufacture

Conclusions

We have studied the structure and the the tensile and compressive characteristics of a PEG–DE cross-linked bionic silk fibroin tubular scaffold in order to evaluate its potential for use in small diameter vascular grafts. FTIR and XRD tests showed that PEG–DE could cross-link with SF and induce SF molecules to form β-sheets during the gel reaction process, resulting in improved crystallinity of the inner and outer membranes. With increased SF concentration, the SFTS showed good performance in tensile and elastic properties. The most suitable SF concentration during the fabrication process was 6%. The uniform micropore structure formed by freezing could enhance the elastic properties of the SFTS. The study provided a new method for the preparation of small caliber artificial blood vessels, and would also expand the application of silk fiber or fabric in the medical field.

Footnotes

Funding

This work was supported by the National Natural Science Foundation of China (No. 51173125), the Natural Science Foundation of Jiangsu Province, China (No. BK2012633), the College Natural Science Research Project of Jiangsu Province, China (No. 12KJA43004), the Science and Technology Plan Foundation of Suzhou, China (No. SYG201001 and ZXS2012002) and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

References

Kakisis

Liapis

Breuer

. Artificial blood vessel: The Holy Grail of peripheral vascular surgery. J Vasc Surg 2005; 41: 349–354.

Chandy

Das

Wilson

. Use of plasma glow for surface-engineering biomolecules to enhance blood compatibility of Dacron and PTFE vascular prosthesis. Biomaterials 2000; 21: 699–712.

Deutsch

Meinhart

Fischlein

. Clinical autologous in vitro endothelialization of infrainguinal ePTFE grafts in 100 patients: A 9-year experience. Surgery 1999; 126: 847–855.

Hsu

Chuang

Chen

. Endothelial cell attachment to the gamma irradiated small diameter polyurethane vascular grafts. Biomed Mater Eng 2006; 16: 397–404.

Kannan

Salacinski

Butler

. Current status of prosthetic bypass grafts: a review. J Biomed Mater Res B: Appl Biomater 2005; 74: 570–581.

Pereira

Albers

Romiti

. Meta-analysis of femoropopliteal bypass grafts for lower extremity arterial insufficiency. J Vasc Surg 2006; 44: 510–517.

Hoshi

Lith

Jen

. The blood and vascular cell compatibility of heparin-modified ePTFE vascular grafts. Biomaterials 2013; 34: 30–41.

Matsuo

Kanasugi

Ohgoe

. Biocompatibility of a –C:H film coating for synthetic vascular grafts. Diamond Relat Mater 2009; 18: 1031–1034.

Zhu

Zhang

Shen

. Blood compatibility of chitosan/heparin complex surface modified ePTFE vascular graft. Appl Surf Sci 2005; 241: 485–492.

10.

Inouye

Kurokawa

Nishikawa

. Use of Bombyx mori silk fibroin as a substratum for cultivation of animal cells. J Biochem Biophys Methods 1998; 37: 159–164.

11.

Bessa

Balmayor

Azevedo

. Silk fibroin microparticles as carriers for delivery of human recombinant BMPs. Physical characterization and drug release. J Tissue Eng Regen Med 2010; 4: 349–355.

12.

Vepari

Kaplan

. Silk as a biomaterial. Prog Polym Sci 2007; 32: 991–1007.

13.

Minoura

Aiba

Higuchi

. Attachment and growth of fibroblast cells on silk fibroin. Biochem Bioph Res Co 1995; 208: 511–516.

14.

Bray

George

Ainscough

. Human corneal epithelial equivalents constructed on Bombyx mori silk fibroin membranes. Biomaterials 2011; 32: 5086–5091.

15.

Liu

Fan

Wang

. The interaction between a combined knitted silk scaffold and microporous silk sponge with human mesenchymal stem cells for ligament tissue engineering. Biomaterials 2008; 29: 662–674.

16.

Lovett

Eng

Kluge

. Tubular silk scaffolds for small diameter vascular grafts. Organogenesis 2010; 6: 217–224.

17.

Lovett

Cannizzaro

Daheron

. Silk fibroin microtubes for blood vessel engineering. Biomaterials 2007; 28: 5271–5279.

18.

Marelli

Alessandrino

Farè

. Compliant electrospun silk fibroin tubes for small vessel bypass grafting. Acta Biomater 2010; 6: 4019–4026.

19.

Zhou

Cao

. A novel three-dimensional tubular scaffold prepared from silk fibroin by electrospinning. Int J Biol Macromol 2009; 45: 504–510.

20.

Zhang

Wang

Keshav

. Dynamic culture conditions to generate silk-based tissue-engineered vascular grafts. Biomaterials 2009; 30: 3213–3223.

21.

Abbott

Megerman

Hasson

. Effect of compliance mismatch on vascular graft patency. J Vasc Surg 1987; 5: 376–382.

22.

Quiñones-Baldrich

Moore

Ziomek

. Development of a “leak-proof” knitted Dacron vascular prosthesis. J Vasc Surg 1986; 3: 895–903.

23.

Cao

Liu

Wei

. In vitro tendon engineering with avian tenocytes and polyglycolic acids: a preliminary report. Tissue Eng 2006; 12: 1369–1377.

24.

Deutsch

Meinhart

Zilla

. Long-term experience in autologous in vitro endothelialization of infrainguinal ePTFE grafts. J Vasc Surg 2009; 49: 352–362.

25.

Dargaville

Vaquette

Peng

. Cross-linked poly(trimethylene carbonate-co-l-lactide) as a biodegradable, elastomeric scaffold for vascular engineering applications. Biomacromolecules 2011; 12: 3856–3869.

26.

Baudis

Heller

Liska

. (Meth)acrylate-based photoelastomers as tailored biomaterials for artificial vascular grafts. J Polym Sci Part A: Polym Chem 2009; 47: 2664–2676.

27.

Baguneid

Seifalian

Salacinski

. Tissue engineering of blood vessels. Brit J Surg 2006; 93: 282–290.

28.

Tai

NRM

Giudiceandrea

Salacinski

. In vivo femoropopliteal arterial wall compliance in subjects with and without lower limb vascular disease. J Vasc Surg 1999; 30: 936–945.

29.

Enomoto

Sumi

Kajimoto

. Long-term patency of small-diameter vascular graft made from fibroin, a silk-based biodegradable material. J Vasc Surg 2010; 51: 155–164.

30.

Nakazawa

Sato

Takahashi

. Development of small-diameter vascular grafts based on silk fibroin fibers from Bombyx mori for vascular regeneration. J Biomat Sci 2011; 22: 195–206.

31.

Wang

Wei

. Preliminary biocompatibility evaluation of regenerated Antheraea yamamai Silk Fibroin in vitro. J Wuhan Univ Technol: Mater Sci 2011; 26: 1044–1048.

32.

Liu

Huang

Yang

. Construction and analysis of morphological characteristics of a novel fabric-based silk fibroin vascular prosthesis. In: TBIS 2011 Int Sym on Textile Bioengineering and Informatics 2011, pp. 151–154.

33.

Lovett

Cannizzaro

Vunjak-Novakovic

. Gel spinning of silk tubes for tissue engineering. Biomaterials 2008; 29: 4650–4657.

34.

Hoenig

Campbell

Rolfe

. Tissue engineered blood vessels: alternative to autologous grafts? Arterioscl Throm Vasc Biol 2005; 25: 1128–1134.

35.

Patel

Fine

Sandig

. Elastin biosynthesis: the missing link in tissue-engineered blood vessels. Cardiovasc Res 2006; 71: 40–49.