Advanced Silk Fibroin Biomaterials and Application to Small-diameter

Subscriber access provided by WEBSTER UNIV

Review

Advanced Silk Fibroin Biomaterials and Application to Small-diameter Silk Vascular Grafts Tetsuo Asakura, Takashi Tanaka, and Ryo Tanaka ACS Biomater. Sci. Eng., Just Accepted Manuscript • DOI: 10.1021/ acsbiomaterials.8b01482 • Publication Date (Web): 28 Feb 2019 Downloaded from http://pubs.acs.org on February 28, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Biomaterials Science & Engineering

Advanced Silk Fibroin Biomaterials and Application to Smalldiameter Silk Vascular Grafts

Tetsuo Asakura,* Takashi Tanaka and Ryo Tanaka a Department

of Biotechnology, Tokyo University of Agriculture and Technology, 2-24-16

Nakacho, Koganei Tokyo 184-8588, Japan; E-mail: [email protected]; Fax: +81-423-88-7025

Tel: +81-423-83-7733

Abstract Since the incidences of cardiovascular diseases have been on the rise in recent year, the need for small-diameter artificial vascular grafts is increasing globally. Although synthetic polymers, such as expanded polytetrafluoroethylene or poly(ethylene terephthalate) have been successfully used for artificial vascular grafts with ≥ 6 mm in diameter, they fail at smaller diameters (< 6 mm) due to thrombus formation and intimal hyperplasia. Thus, development of vascular grafts for small diameter vessel replacement that are < 6 mm in diameter remains a major clinical challenge. Silk fibroin (SF) from Bombyx mori silkworm is well-known as an excellent textile, and also has been used as suture material in surgery for more than 2,000 years. Many attempts to develop small-diameter SF vascular grafts with < 6 mm in diameter have been reported.

Here, research and development in small-diameter vascular grafts with

SF are reviewed as follows. 1. The heterogeneous structure of SF fiber (Silk II) including the packing arrangements and type II -turn structure of SF (Silk I*) before spinning. 2. SF modified by transgenic silkworm which is more suitable for vascular grafts. 3. Preparation of small-diameter SF vascular grafts. 4. Characterization of SF in the hydrated state including dynamics of water molecules by nuclear magnetic resonance and 5. Evaluation of the SF

1 ACS Paragon Plus Environment

ACS Biomaterials Science & Engineering 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 42

grafts by in vivo implantation experiment. According to the findings, SF is promising material for small-diameter vascular graft development.

Keywords

Bombyx mori silk fibroin / Small-diameter vascular graft / Solid-state NMR /

Transgenic silk fibroin


Page 3 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1. Introduction Bombyx mori (B. mori) silk fiber produced by silkworms has a long history of use in textiles.1 The silk fiber has also been used in sutures in the surgical field for more than 2,000 years due to its toughness and biocompatibility.2 In recent years, silks and silk-based materials including recombinant silks and transgenic silks have increasingly been used with promising prospects in biomedical applications.3-9

With regard to the biomedical applications, it is essential to

clarify the structural and dynamic properties of B. mori silk fibroin (SF) in the dry and hydrated states, in addition to its mechanical properties and biocompatibility. There have been major advances in the elucidation of the structure and dynamics of SF in the dry and hydrated states by advances in structural analytical technology,10 especially solid-state nuclear magnetic resonance (NMR).11,12 In the present review, the authors focus on summarizing recent developments in knowledge on the structure and dynamics of silks, and the application of silks and silk-based materials to biomaterials, particularly small-diameter vascular grafts. Since the incidences of cardiovascular diseases have been on the rise in recent years, the need for small-diameter artificial vascular grafts is increasing globally.13 Synthetic polymers, such as expanded polytetrafluoroethylene (ePTFE) and poly(ethylene terephthalate) (PET) are clinically applied in artificial vascular grafts with diameter > 6mm.14,15 However, such materials present numerous challenges, since occlusion due to thrombus formation and intimal hyperplasia occur in small-diameter vascular grafts with diameter < 6mm.

Such

challenges arise because of lack of vascular endothelial cells and mismatch with original native blood vessel.16-19 Therefore, in the recent decades, studies on small-diameter artificial blood vessels using scaffolding materials that promote remodeling of self-tissues have been actively conducted.20-22 However, small-diameter artificial blood vessel which satisfy the requirements is still not accomplished. To date, numerous investigations have been carried out in search of various materials that could replace malfunctioning or diseased cardiovascular tissues.23 Degradable synthetic materials such as polyglycolic acid, poly-L-lactic acid, 3 ACS Paragon Plus Environment


Page 4 of 42

polylactide-coglycolide, and polycaprolactone have also been used to create vascular grafts, and have yielded promising results.24 However, owing to thrombus formation and compliance mismatch, none of the materials are suitable for generating small-diameter vascular grafts to replace the original vessel in clinical setting. Vascular endothelial cells25 and bone marrow cells26 are prepared by seeding them in advance in artificial vascular grafts or using “biotubes”.27 However, in such artificial blood vessels, there are major challenges. For example, their production is time-consuming and they may not be preserved for long periods. Compared with synthetic polymers, natural biopolymers such as collagen, elastin, and SF, offer superior cytocompatibility and biocompatibility in the presence of embedded structural and functional molecules. Their biological advantages, low inflammatory and antigenic response make them an alternative strategy to prepare vascular grafts.28-30 SF is one of the most preferred natural materials due to its impressive biocompatibility, controllable biodegradability, minimal inflammatory reactions, suitable mechanical properties, and high processing ability to form several tailorable morphologies such as hydrogels, powders, films, fibers, tubes, and porous sponges. There are many reports about these merits and reviewed by Wang et al. 23 and Thurber et al.5 Thus, silks and silk-based materials are suitable candidate for small-diameter vascular grafts (< 6 mm). In the present review, the authors will focus on the following sections. 2. Recent developments in structures of SF at the molecular level, 3. Transgenic SF with more suitable properties for vascular grafts, 4. Preparation of small-diameter silk vascular grafts, 5. Characterization of SF grafts in the hydrated state and 6. Evaluation of small-diameter silk vascular grafts in vivo.

2. Recent development in structure of SF at the molecular level Over the past 20 years, the structures of SF in the aqueous solution and in the solid have been clarified and reviewed in detail elsewhere.5,6,12 Therefore, we summarize the structure briefly 4 ACS Paragon Plus Environment

Page 5 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


and emphasize recent development of the structural analysis of SF in the solid state.

A B.

mori cocoon is composed of a single fiber with 1300 −1500 m long and 10−20 μm in diameter. An individual fiber is composed of twisted bunches of fibrils, which presumably strengthen the fiber. The twin fiber is composed of two silk proteins: SF and silk sericin (SS). The SF molecule, which consists of a heavy (H) 390-kDa chain and a light (L) 26-kDa chain linked by a disulfide bond, as well as a glycoprotein named P25 (30-kDa),31−34 is secreted into the posterior silk gland as an aqueous solution. The H-chain, L-chain, and P25 are thought to be assembled into a high molecular mass elementary unit at a ratio of 6:6:1. Then, SF stored in the middle silk gland is spun out through the anterior silk gland and converted into silk fibers, being coated with SS as it emerges. The SS coating is often removed during processing, through a process known as degumming. The amino acid composition of the H chain of SF (in mol %) is dominated by four amino acids: Gly (46%), Ala (30%), Ser (12%), and Tyr (5.3%).35 The repetitive core of the primary structure is composed of alternate arrays of 12 repetitive and 11 amorphous domains.35−37 The sequence is roughly divided into four motifs: (i) a highly repetitive GAGAGS sequence which makes up the bulk of the crystalline regions and is found primary at the beginning of each subdomain; (ii) a less repetitive sequence containing aromatic and/or hydrophobic residues such as GAGAGY and GAGAGV, making up the semi-crystalline regions; (iii) sequences very similar to (i) except for the presence of an AAS motif and (iv) amorphous regions which separate the domains and contain negatively charged, polar, bulky hydrophobic and/or aromatic residues.25,36 Structural analyses of SF fibers have revealed that they are composed of small β-sheet crystalline units embedded in an amorphous matrix.38 Direct images of SF β-sheet crystals are rare although it is expected that fiber mechanical properties would largely depend on their sizes, aspect ratios, and size distributions. The size of β microcrystallites was estimated to be 2.6 nm (intermolecular hydrogen bond direction), 3.2 nm (sheet stacking direction), and 11.5 nm (chain direction) by X-ray diffraction by Xu et al.39 The structure of the SF fiber is commonly called 5 ACS Paragon Plus Environment


Page 6 of 42

Silk II. Silk II was one of the first protein structures to be studied using X-ray diffraction, culminating in the proposal by Marsh, Corey, and Pauling in 1955 of a regular antiparallel βsheet, based on an X-ray fiber diffraction study of native SF fiber.40 Later studies41-43 supported the general features of the antiparallel β-sheet model, but proposed that the structure was less regular than the original Marsh model. Takahashi et al.44 carried out a detailed study using superior data and made two significant modifications to the original model. However, their model was energetically unstable and more energetically favorable model was required. The composite structures of silks, comprise both crystalline and amorphous domains, which imply that X-ray diffraction provides only limited information on the structures. Other spectroscopic techniques have been applied to silks, but the most detailed picture illustrating the structure and dynamics of silks has emerged from NMR. Currently, solid-state NMR offers the most reliable Silk II structure.

Figure 1a. presents 13C cross polarization/

magic angle spinning (CP/MAS) NMR spectrum of Ala Cβ carbon of [3-13C] Ala native SF fiber with Silk II form, indicating that Silk II is a heterogeneous structure,45 which is inconsistent with that presented by the previous Marsh model40 and in agreement with the less regular structures pointed out by other researchers.41-44 The Ala C peak consists of 56% crystalline fraction and 44% non-crystalline fraction. Moreover, the crystalline fraction consists of 18% distorted β-turn, 25% β-sheet (A), and 13% β-sheet (B), and non-crystalline fraction consists of 22% in both distorted β-turn and distorted β-sheet. Thus, the Ala C peak can be deconvoluted into five peaks. Here the crystalline fraction was considered to be the precipitated fraction after chymotrypsin digestion of all SF.46 The distorted -turn means that the structure is essentially similar to the Silk I* structure, as will be mentioned below, except for a larger distribution of the backbone torsion angles of Silk I* structure and the broad peak has the same Ala C chemical shift as that of a random coil conformation.45 The two crystalline β-sheet-like peaks of the sequence, (AGSGAG)n labeled as A and B, respectively, 6 ACS Paragon Plus Environment

Page 7 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


occur in a ratio close to 2:1, which indicates the presence of two different packing structures in the -sheet crystalline regions as shown in Figure 1b. Details of the analysis of two kinds of the packing structures, A and B are provided elsewhere, and the two packing structures are similar to those of the Takahashi model, but more energetically stable.45 The five peaks in Ala C carbons with different fractions are also observed in the 13C CP/MAS NMR spectra of regenerated SF in several forms, including film, fiber, sponge and powder. The physical properties of the SF samples depend on the fractions of several conformations determined from the relative intensities of the five Ala C peaks47as well as long-range structure such as the amount and size of the -sheet crystallites, orientation of -sheet crystallites determined from X-ray diffraction analyses.6

(a)

(b)

-sheet A

-sheet B

Figure 1. (a). Expanded Ala C peak in the 13C solid-state NMR spectrum of native SF fiber together with the assignment. The crystalline domain (56%) with the sequence, (AGSGAG)n 7 ACS Paragon Plus Environment


Page 8 of 42

painted in blue consists of 18% distorted -turn,25% -sheet A, and 13% -sheet B. The noncrystalline domain (44%) painted in orange consists of 22% distorted -turn and 22% distorted -sheet. (b). The two kinds of the packing structures of -sheet A and -sheet B are also shown. The arrows in the packing structure represent the chains with the direction from the C-terminus to N-terminus. On the other hand, Silk I is the solid state structure of excised SF stored in the middle silk glands, and dried under mild conditions without any external forces such as stretching, shear stress and so on.48,49 The Silk I consists of a mixture of Silk I* and random coil structures, and the essential Silk I* structure of (AGSGAG)n in SF has also been revealed through numerous spectroscopic methods in addition to molecular dynamics simulation, although mostly using solid-state NMR.50-52 The Silk I* structure viewed from three orthogonal directions, a, b and c, is shown in Figure 2a. This is a repeated type II β-turn structure with the torsion angles, (φ, ψ) = (−62°, 125°) for Ala residues and (φ, ψ) = (77°, 10°) for Gly residues.51

(a)

(b)

(c)


Page 9 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. (a). Silk I* structure of the crystalline domain, (AGSGAG)n of SF. (b). Formation of intra-molecular hydrogen bonds (green broken lines) along a SF chain. (c). Inter-molecular hydrogen bonds (green broken lines) formed between the two adjacent SF chains. The expanded pictures, Figure 2b and 2c, are provided to further reveal the structure. The Gly NH protons contributed to intra-molecular hydrogen bond formation (green broken lines) with parallel to the chain (Figure 2b), whereas Ala NH protons contributed to inter-molecular hydrogen bond formation (green broken lines) perpendicular to the chain direction (Figure 2c). Therefore, intra- and intermolecular hydrogen bonding appears along the Silk I* SF chain, alternately and stabilizes the Silk I* structure of self-assembled SF molecules.

In

addition, Ser residues stabilize the Silk I* structure through hydrogen bond formation between the side chain OH groups and the C=O groups of the backbone chain.53 The Silk I* structure can change to Silk II following subjection to weak external forces.54 SF fibers can be solubilized in high concentration salt solutions such as lithium bromide, lithium thiocyanate, or calcium chloride. After solubilization in such aggressive solvents, dialysis into water or buffers can be used to eliminate the salts or acids, although premature reprecipitation is a common problem. From SF dissolved in aqueous solution or SF dissolved in an organic solvent prepared originally from silk cocoons, several different forms of SF can be prepared easily as shown in Figure 3, which is one of the excellent SF properties when they are applied as biomaterials. Films of silkworm silk have been produced by air-drying aqueous solutions prepared from the concentrated salts, followed by dialysis.55-59 Fibroin sponges are prepared from fibroin solution by freeze-drying or adding salt or sugar particles as porogens.60,61 They provide a 3D porous scaffolding material with large interior surface area and interconnected pore spaces. Non-woven mats and fibers are usually fabricated from SF solutions by electrospinning techniques.62,63 As will be described below, different approaches and methodologies are being used to produce silks and silk-based materials with adaptable material forms targeting a wide range of applications. 2-8,64 9 ACS Paragon Plus Environment


Page 10 of 42

Regenerated Fiber

Film

Non-woven Fabric

Sponge

Tube

Figure 3. Various morphologies of SF, which were originally obtained from silk cocoons. From the aqueous solution and solution dissolved in an organic solvent of the regenerated SF, fiber, film, tube, sponge and non-woven fabric can be prepared and used for several kinds of biomaterials. 3. Transgenic SF with more suitable properties for vascular grafts SF is largely a promising material for small-diameter vascular grafts owing to the long history of application in the human body as sutures. As a biomaterial, SF exhibits impressive biocompatibility, controllable biodegradability, minimal inflammatory reactions, excellent mechanical properties, and high processing ability to form several tailorable morphologies. In addition, numerous researchers have attempted to enhance the desirable properties of SF to yield vascular grafts with excellent hemocompatibility and cytocompatibility using chemical (sulfated or heparinized) or physical (physical blends of SF with other materials) modification strategies. These are fascinating ways of tweaking SF properties and widening the potential range of application in the medical field. Murphy and Kaplan65 and Pritchard et al.66 have 10 ACS Paragon Plus Environment

Page 11 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


extensively reviewed such efforts. Over the past two decades, recombinant silk protein production has been optimized to the extent and it can be manufactured.8,67 In the present review, we describe transgenic (TG) silks that were used to introduce functional groups into SF for the development of small-diameter vascular grafts. In 2,000, a germ line transformation method for silkworms was developed using the piggyBac transposon.68,69 There are numerous advantages of using TG silkworms for the production of recombinant silk proteins. It is easy to manipulate large quantities of TG silkworms because the rearing system using mulberry leaves and an artificial diet has considerably improved. The adult moths are unable to fly, and the silkworms are unable to live in nature. In addition, the SF fiber can be used without the need for fiber processing of the regenerated SF. The SF proteins fused with a fibroin L-chain and green fluorescent protein (GFP) or a fibroin H-chain and GFP can be produced in the posterior silk gland and expelled from body in the cocoon.69–71 Therefore, another fusion protein could be expressed using this TG method, and the characteristics of SF could be modified by the introduction of a novel gene fused with the fibroin L-chain or H-chain gene. We fused some functional peptides into fibroin L-chain or H-chains. Notably, the SF with Arg–Gly–Asp (RGD) found in fibronectin exhibited enhanced cell adhesion properties.72 The SF with Tyr- Ile-Gly-Ser-Arg (YIGSR) derived from laminin B1 chain was fused to the H-chain and showed enhanced cell adhesion activity and greater cell migration compared with wild type (WT) SF.73 These results showed that fused fibroin has potential application in small diameter silk vascular grafts. Here, we focus on using two kinds of functional sequences, RGD and the vascular endothelial growth factor (VEGF), to improve adhesion to endothelial cells and antithrombogenic properties.74 VEGF promotes the growth of endothelial cells and endothelial cell rolling inhibition of platelet attachment.75 Fluorescent microscope images of human umbilical vein endothelial cell (HUVEC) adhesion observations for three SF samples are shown in Figure 4. Both VEGF SF and RGD SF exhibited significantly higher attachment and 11 ACS Paragon Plus Environment


Page 12 of 42

growth activity in HUVECs than in WT SF. Therefore, cell adhesion activity in RGD and vascular endothelial growth activity in VEGF facilitated adhesion and growth of endothelial cells. Figure 5 shows platelet adhesion in the three SF examined by scanning electron microscopy (SEM) and lactate dehydrogenase activity. The attachment intensity of platelets was higher in RGD and WT SFs as observed in the SEM images, whereas VEGF SF had significantly lower levels of platelet attachment. Therefore, RGD SF had higher attachment of HUVECs (Figure 4C) and platelets (Figure 5C) than WT SF. Conversely, VEGF SF had higher attachment of HUVECs (Figure 4B) and lower attachment of platelets than WT SF (Figure 5B). Consequently, VEGF modified SF-based materials not only exhibited optimal patency but also showed optimal tissue infiltration and in turn new vessel formation. Therefore, the variant particularly promising for future applications in anti-thrombosis smalldiameter grafts. 8,74


Page 13 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. HUVEC adhesion observation by fluorescent microscopy. (A) WT, (B) VEGF-fused SF, and (C) RGD-fused SF. (D) Number of adherent HUVECs at 2 h after incubation and (E) HUVEC growth at 1, 3, and 5 days after cultivation.　TCP: Tissue Culture Plate. SF’s with VEGF and RGD showed higher numbers of the attached HUVECs than the WT SF. Adapted with permission from ref.74. Copyright 2015 Royal Society of Chemistry.

10.0μm

10.0μm

10.0μm

Figure 5. Canine platelet-rich plasma (PRP) evaluation of SEM images of (A) WT SF, (B) VEGF-fused SF, and (C) RGD-fused SF. (D) Platelet count by SEM evaluation and (E) lactate dehydrogenase (LDH) lysis assay. The VEGF SF showed significantly lower attachment of platelets compared with WT SF and RGD SF. Adapted with permission from ref.74. Copyright 2015 Royal Society of Chemistry.



Page 14 of 42

4. Preparation of small-diameter silk vascular grafts The ideal small-diameter vascular graft must possess several characteristics, including (a) mechanical strength, which includes physiological compliance, the capacity to withstand long-term hemodynamic stress without failure, and resistance to permanent creep that can lead to aneurysms; (b) biocompatibility, which is associated with a confluent, quiescent, and non-activated endothelium; (c) a suitable healing response that does not result in inflammation, hyperplasia, or fibrous capsule formation; and (d) ease of handling and suturability, which are crucial from a surgical perspective.76,77 It is challenging to accomplish these requirements completely. SF is one of the most valuable candidates as described above. SF can be made into films, nanoscale electrospun fibers, sponges, knitted or woven mats, and tubes, for use as matrices in vascular tissue engineering as shown in Figure 2. SF films can also be made into tubes using the sequential dipping method for vascular applications. Lovett et al.78 prepared SF microtubes with controllable pore sizes by dipping stainless steel wires into SF solution, with poly(ethylene oxide) as the porogen being used to control the porosity of microtubes. Such a microtube can withstand physiological pressures and allow protein diffusion and cell migration. However, the burst strengths of such silk tubes are significantly low, and most microtubes fail at higher than 100–140 mmHg (the approximate physiological pressures of coronary artery during systole) or 15–40 mmHg (capillary blood pressure), indicating the need for additional in vivo studies. Numerous researchers explored the electrospinning technique in the preparation of small-diameter vascular grafts from SF solutions.79–91 Electrospinning is the formation of fibers in the micrometer to nanometer range by electrically charging slowly extruded solutions.62,63,92 First, a droplet forms at the tip of the needle, and, if the parameters are set appropriately, the forces of the electrical charge overcome the tension of the droplet, and a jet is formed. Eventually, the jet undergoes whipping instabilities and is stretched into an ultrathin fiber in the ideal case. As shown in Figure 6, the most basic set-up 14 ACS Paragon Plus Environment

Page 15 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


for producing electrospun tubular scaffolds for vascular applications consists of a highvoltage power supply, a controlled syringe pump, and a rotating mandrel as the fiber collector, which is the key element that technically allows polymeric fibers to be arranged in a cylindrical construct.92

Figure 6. A series of the electrospinning apparatus to prepare small-diameter SF vascular grafts. The grafts can be obtained from the accumulation of SF fibers deposited on the rod. Adapted with permission from ref.92. Copyright 2015Wiley.

The results of implanting small-diameter silk grafts prepared by electrospinning in animals will be described in a later section. Electrospinning has several advantages. It requires low working volumes to produce large amounts of scaffold, and there is relatively fine control over what is produced. It is a promising method for fabricating nanofiber scaffolds that mimic the morphological properties of natural extracellular matrices. However, for the silk grafts with larger than 2mm diameter, the electrospun grafts are generally too weak and further studies are required to improve the mechanical strength of electrospun silk grafts.85 An alternative method for the direct use of silk fibers is the generation of a knitted or 15 ACS Paragon Plus Environment


Page 16 of 42

weaved silk structure to reinforce three-dimensional (3D) porous tissue engineering scaffolds. Double-raschel knitting has been used in the preparation of vascular grafts of commercially available polyester fibers. The preparation of small-diameter silk vascular grafts has been achieved using the same knitting technique.93-100 In the double-raschel knitting process, the physical or mechanical characteristics of the vascular graft and the sizes can be altered using a computer-controlled double-raschel knitting machine. Figure 7 presents an image of the double-raschel knitting and SEM pictures of the outer surface, inner surface, and a cross section of the double-raschel knitting. Merits of double-raschel knitting are integrated as that the silk threads do not become flat even if the threads are pressurized by a guide or needle and the silk threads exhibit appropriate elasticity. In addition, since there are numerous contact points in the fibers, adequate strength and protection from loosening at the edges during the implantation process could be attained. Friction at the contact points is also reduced, which prevents thread tears and/or thread separation during the manufacturing process.

(a)

(b)

Outer Surface

Inner Surface

Cross Section

Figure 7. (a) An image of SF graft fabrication using a double-raschel knitting machine. (b) Scanning electron microscopy (SEM) images of small-diameter SF vascular grafts before coating; outer surface (left), inner surface (middle), and cross section (right), respectively. Adapted with permission from ref.97. Copyright 2016 Royal Society of Chemistry. 16 ACS Paragon Plus Environment

Page 17 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(a)

Animal Implantation experiment

Rod was covered by The rod was a double-raschel immersed in knitted SF tube SF/PGDE(1:1) mixture solution

The test tube was allowed to stand overnight in a freezer (-20 °C)

After the rod was pulled out, the SF tube was dialyzed in water for 3 days

The SF tube was housed in the bag with water

The bag was sterilized in an autoclave (120 °C,20min)

(b)

Figure 8. (a) Preparation of small-diameter SF vascular SF graft coated with SF sponge: 1) a rod was covered by a double-raschel knitted SF tube; 2) the rod with the SF tube was inserted in a pipe with slightly wider diameter, and the space was filled with mixed SF/PGDE aqueous solution; 3) the graft was frozen at −20 °C for 24 h in a refrigerator; 4) the graft was allowed to stand in deionized water to remove PGDE; and, 5) the graft was placed in a bag and 6) sterilized in an autoclave at 120 °C for 20 min. (b) (A) Macroscopic images of the SF vascular grafts coated with SF sponge with 1.5 mm (left) and 3.5 mm (right) diameter; (B) SEM image of the cross section of a 1.5 mm graft, and (C) the expanded SEM image of the cross section. Adapted with permission from ref.93. Copyright 2014 Royal Society of Chemistry. Figure 8a summarizes the process of preparing an SF vascular graft coated with SF sponge using poly(ethylene) glycol diglycidyl ether (PGDE) as porogen to decrease the leakage of blood from the graft, to protect from the loosing of the fibers at the edges in the implantation process, and to strengthen the elastic character of the graft.93 Min et al.101 prepared porous tubular scaffolds from only SF aqueous solution with the addition of PGDE instead of poly(ethylene oxide).78 The scaffolds were flexible, elastic, and transparent in the wet state with pore sizes of 81–128 μm and porosity of 90–96% depending on the concentrations of SF and PGDE. The tubular SF scaffolds had satisfactory tensile and compression properties, 17 ACS Paragon Plus Environment


Page 18 of 42

particularly excellent deformation-recovery ability. No cytotoxicity to mouse L-929 fibroblasts was detected. In addition, leakage of blood from the graft decreased and the loosening of the fibers at the edges during the implantation process was prevented. Therefore, the double-raschel knitted SF graft was coated with an SF sponge prepared with PGDE as porogen.93-100 The graft exhibited adequate mechanical strength that could withstand hemodynamic forces. The results of animal implantation will be shown in a later section. Many other methods for preparing 3D porous tissue engineering scaffolds by SF30,60,102–111 have been reported.

5. Characterization of SF grafts in the hydrated state Vascular grafts are generally used in a hydrated state, and the mechanical properties of SF molecules change remarkably because of the effect of water.5,38 Therefore, it is critical to understand the structure and dynamics of SF grafts and the SF sponge coatings in the hydrated state during the development of SF vascular grafts. Numerous analytical methods including Raman,112 IR,112–116 dynamic scanning calorimetry,115-118 thermogravimetric analysis,117,118 X-ray diffraction,118 dynamic mechanical thermal analysis,119 and NMR120–137 among others have been used for the purpose. They have provided numerous insights into the structure and dynamics of SF molecules in the hydrated state. Among the analytical methods, the most detailed illustrations of the structures and dynamics of SF in the hydrated state have emerged from NMR, as well as those of the dry state. Particularly, the conformationdependent 13C chemical shift coupled with 13C-selective labeling in SF is useful for the determination of the fractions of several conformations using amino acid-specific manner, which has been used frequently in structural analyses of SF.11,12,45,50,126–133,135–139 Figure 9 shows the expanded region of [3-13C] Ala-, [3-13C] Ser-, and [3-13C] Tyr- peaks in the 13C dipolar decoupled magic angle spinning (13C DD/MAS) NMR spectra of [3-13C] Ser-, [3-13C] Tyr-, and [3-13C] Ala-labeled native SF fibers in the dry and hydrated states.126,128,130 The 18 ACS Paragon Plus Environment

Page 19 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


peak deconvolutions also determine the fractions of several conformations for the three 13C labeled residues and the fractions are summarized in Table 1.　Here, Ser residue is present

Figure 9. Observed and simulated 13C DD/MAS NMR spectra of Ala C, Ser C and Tyr C carbons of [3-13C] Ser-, [3-13C] Tyr-, and [3-13C] Ala-labeled native SF fiber in the (a) dry and (b) hydrated states. In both Ala C and Ser C peaks, the symbols are as follows. ① βsheet B, ② β-sheet A, ③ random coil, and ④ hydrated random coil. In the Tyr C region, the peaks at 42.7 ppm are assigned to natural abundant Gly Cα peaks. The symbols are ③ random coil, ④ hydrated random coil, and ⑤ β-sheet. Adapted with permission from ref.126. Copyright 2015 American Chemical Society. Table1. Fractions (%) of different conformations of Ala, Ser, and Tyr residues determined from the 13C DD/MAS NMR spectra of [3-13C] Ser-, [3-13C] Tyr-, and [3-13C] Ala-labeled native SF fibers in the dry and hydrated states. 126

Ala

β-sheet B

β-sheet A

random coil

Dry Hydrated

19.0 18.5

47.0 46.5

34.0 26.4

Ser

β-sheet B

β-sheet A

random coil

Dry Hydrated

17.0 17.4

33.0 32.7

50.0 39.1

Tyr

β-sheet B

random coil

Dry Hydrated

41.0 41.4

59.0 52.3

Hydrated random coil

ratio(%;H/(r+H))

8.6

25


ratio(%;H/(r+H))

10.8

22


ratio(%;H/(r+H))

6.3

11



Page 20 of 42

In the hydrated state of SF, Both the hydrated random coil, H and non-hydrated random coil, r was observed. The ratio (%;H/(r + H)) was calculated and listed in the last column. predominantly in the crystalline domains of SF, Tyr residue is present in the non-crystalline domains, and Ala residue is present in both domains.35–37 In the spectra of the dried SF sample, the Ala C and Ser C peaks could be resolved into three peaks, i.e., -sheet A, sheet B, and random coil (painted by yellow), respectively. Two kinds of the packing structures of -sheet structure have already been presented in Figure 1.45 The Tyr C peak was resolved into two peaks, -sheet and random coil, although the lower field peak is overlapped with a natural-abundance Gly C peak.126 In the hydrated state, the fractions of β-sheets of the residues in the native fiber did not change significantly compared with those in the dry state (Table 1). Conversely, all the random coil peaks of the three 13C-labeled Ala C, Ser C and Tyr C carbons, and naturalabundance Gly C carbon were resolved into two peaks, i.e., broad and sharp peaks (painted light blue). The presence of the latter sharp peak implies fast motion of the residues through the penetration of water molecules into relatively disordered random coil regions of SF and increase in the chain motion due to breakage of the hydrogen bonding. The fractions of random coil conformations with fast motion in the total random coil fraction of the hydrated fiber were 25%, 22%, and 11% for Ala, Ser, and Tyr residues, respectively. Therefore, Tyr residues tend to hydrate less among the residues. The effect of water on the structure and dynamics of regenerated SF was also studied using these NMR methods.130 The structures and dynamics of SF sponges prepared using three kinds of porogens, i.e., PGDE, glycerin (Glyc), and poly(ethylene glycol) (PEG) were studied in the hydrated states.129 Softer and highly flexible SF sponges are required for the development of biomaterials that are biodegradable and exhibit stiffness that match sponges and soft tissues in water. Such sponges can be used as coating materials on SF vascular grafts. There were no 20 ACS Paragon Plus Environment

Page 21 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


significant differences in the 13C refocused insensitive nuclei enhanced by polarization transfer (r-INEPT) spectra among the three kinds of sponges in the hydrated state for exclusive detection of mobile components, but there were significant differences among the 13C

CP/MAS NMR spectra for exclusive detection of immobile components. Therefore, the

selection of porogen for the preparation of SF sponges is also critical in controlling degradation behavior140–145 and softness. In order to study the structure and dynamics of the SF sponge coatings on the double-raschel knitted SF tubes, [3-13C] Ser-, [3-13C] Tyr-, and [313C]

Ala-labeled SF sponges were coated on non-labeled double-raschel knitted SF tubes and

the 13C DD/MAS NMR spectra observed.100 The conformations of the coated SF sponges were essentially random coil with fast motion in the hydrated state. In general, water in the silk–water system can be divided into three categories: free water, freezing bound water, and nonfreezing bound water.38,112−117,121-123 Recently, more distinct components of the freezing bound water were identified from 2H solution NMR relaxation property measurements for water in SF fibers.127,128 Figure 10 shows two-dimensional



Page 22 of 42

Figure 10. 2D 2H T1-T2 correlation map of 2H2O in a 2H2O-SF fiber system at 25 °C. The reservoirs A, B, C, and D were observed, indicating four distinct components in the relaxation times for water molecules. The dotted line shown is a guide to the eye where T1 = T2. Adapted with permission from ref.127. Copyright 2017 Elsevier. 2H

T1-T2 correlation map of 2H2O in a 2H2O-SF fiber system at 25 °C. T1 denotes spin–lattice

relaxation time, and T2 represents spin–spin relaxation time. On the basis of an inverse Laplace transform algorithm, four distinct components in the relaxation times for water in SF fiber could be identified, including A: bulk water outside the fiber, B: water molecules trapped weakly on the surface of the fiber, C: bound water molecules located in the inner surface of the fiber, and D: bound water molecules located in the inner part of the fiber. T2 is more sensitive to exchange and dynamic processes in the NMR relaxation times than T1. The dominant reservoir A arises from bulk water outside the fibers exhibiting highly isotropic fast motion as T1 ≈ T2. In contrast, T2 values were significantly less than T1 values for reservoirs, B–D. The remaining three water reservoirs, B–D, may be assigned to exchangeable water molecules interacting with the SF fiber in three different degrees of binding interactions. The 1H

spin density, 1H T2, and diffusion coefficient of mobile water molecules in a small-

diameter SF vascular graft–water system were also studied using the 1H MRI method.146

6. Evaluation of small-diameter silk vascular grafts in vivo Evaluation of small-diameter silk vascular grafts prepared in Section 4 should be finally carried out in animal implantation experiments. In vivo studies of silk vascular grafts reported previously are listed in Table 2.


Page 23 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Since numerous researchers have studied small-diameter silk vascular grafts prepared using the electrospinning technique, we will begin with the most recent results of electrospun silk grafts in vivo as examples. Filipe et al.147 assessed in vivo performance of electrospun silk grafts in rat model abdominal aortic replacements. Having confirmed the appropriate mechanical and biological compatibility of the silk scaffolds, small-diameter silk grafts were implanted into the abdominal aorta of rats. Control animals were implanted with commercial ePTFE with similar diameters. Silk scaffolds were evaluated in a rat aortic interposition grafting model at 3, 6, 12, and 24 weeks. Graft survival was 95% for the silk grafts, with only one death within 24 h of the implantation. However, ePTFE failure was higher at 27% with graft failures related to deaths occurring postoperatively. The SEM images of the luminal sides of the grafts revealed that silk grafts had complete cell coverage as early as 3 weeks with no underlying silk visible. Higher magnification further revealed rough cobblestone cell morphology at 3 weeks, changing to a more elongated morphology by 6 weeks. Conversely, the SEM images of the ePTFE grafts revealed bare surfaces, with the underlying material morphology clearly visible at 3 and 6 weeks. At later time points of 12 and 24 weeks, focal 23 ACS Paragon Plus Environment


Page 24 of 42

points of cell adhesion were observed; however, large areas were not covered with ePTFE. In addition, silk graft cross sections showed endothelial cells present as early as 3 weeks, with complete coverage of anastomotic regions by 6 weeks and full endothelialization by 12 weeks. Next, the evaluation of small-diameter SF grafts prepared using native silk fibers instead of electrospun silk fibers is carried out in animal implantation experiments. Enomoto et al. evaluated the potential of such SF grafts prepared using native silk fibers to generate vascular prostheses for small arteries.148 Twenty-seven silk grafts with 1.5 mm inner diameters were prepared using a combination of braiding and winding of SF fibers two or more times and coating them with SF. The grafts were implanted into the abdominal aorta of male rats by end-to-end anastomosis. Ten ePTFE-based grafts were used as the control. In vivo patency, cellularization, and degradation of arterial circulation in rats were evaluated for up to 72 weeks (Figure 11). Four of the 10 ePTFE grafts were occluded after 4 weeks, and 1 was occluded at 12 weeks. In contrast, only 3 of 27 SF grafts became occluded. Overall 1 year patency of SF grafts was 85%, which was significantly higher than that of ePTFE grafts (85% vs. 30%, p < 0.01). Sirius red staining was used to evaluate biodegradability and extracellular matrix deposition in the SF grafts. Changes in the polarization microscopic images and the SF and collagen contents were examined after implantation of the small-diameter SF grafts over time (Figure 12). The SF contents decreased gradually until 48 weeks after implantation (33 %, p < 0.05 at 48 weeks vs. 2 weeks). Conversely, collagen content increased significantly at 12 and 48 weeks. A histological analysis did not reveal any aneurysmal dilatation. The graft diameters remained unchanged up to 1 year after implantation. Macroscopic observations of the implanted SF grafts at 1 year confirmed smooth luminal


Page 25 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Time(weeks)

Figure 11. Kaplan–Meier analysis shows the graft patency of 27 SF and 10 ePTFE grafts implanted into rat aortas at 2, 4, 8, 12, 24, 48, and 72 weeks. Adapted with permission from ref.148. Copyright 2010 Elsevier.

Figure 12. Polarization microscopic images of SF grafts after sirius red staining. The SF contents (white) gradually decreased, whereas collagen (red) content increased after implantation (* p < 0.05 vs. 2 weeks). The error bars indicate standard error of the mean. Adapted with permission from ref.148. Copyright 2010 Elsevier. 25 ACS Paragon Plus Environment


Page 26 of 42

surfaces with no signs of thrombosis or aneurysmal dilatation as shown in Figure 13A. Histologic analysis of the SF grafts showed the formation of an endothelial layer and a medialike smooth muscle layer (Figure 13B). Vasa vasorum had also formed in the adventitia. Anti-CD68 immunostaining revealed substantial infiltration of macrophages and phagocytic phenomena around the SF remnants. The findings indicate that organization of a vessel-like structure could occur within 1 year when SF grafts are used as scaffolds. In addition, the results suggest that SF could be promising material for the development of small-diameter vascular grafts.

Scale bar, 100 m (upper and lower panels)

Figure 13. A: Macroscopic images of implanted SF graft (G) at 1 year showing no aneurysm formation. The arrows indicate anastomotic sites. The fibroin grafts appeared to be integrated into the surrounding tissue and showed no signs of thrombosis, stenosis, or mechanical failure. Ao, aorta; IVC, inferior vena cava. B: Histological analysis of the SF graft at 1 year showed a confluent endothelial layer and smooth muscle cell layers, as determined by 26 ACS Paragon Plus Environment

Page 27 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


immunostaining against CD31 and -smooth muscle actin (-SMA), respectively. Immunostaining for CD68 revealed substantial macrophage infiltration in the adventitia. Formation of vasa vasorum was noted (arrow heads). Adapted with permission from ref.148. Copyright 2010 Elsevier. Further studies to improve the mechanical properties and handling in the implantation of the SF grafts were conducted using a combination of development and in vivo experiments on the small-diameter vascular graft. Yagi et al.93 prepared a small-diameter double-raschel knitted SF graft with 1.5 mm in diameter and 10 mm in length, which was coated with SF sponge and PGDE used as porogen. It demonstrated adequate mechanical strength and protected the ladder from the end in the implantation process. Such a coating also prevents blood leakage from the graft and allows elasticity of the graft. At 8 weeks after implantation into the abdominal aorta of rats, early formation of thrombosis could be avoided, and smooth muscle cell migration and endothelialization in the inner layer of the graft inner were observed. Aytemiz et al.94 prepared small-diameter double-raschel knitted SF grafts with 3 mm diameters and 3 cm lengths, coated them with SF sponge, used PGDE as porogen, and implanted them into the carotid artery of a dog. Of the five implanted grafts, four of them were exhibited patency over 4 weeks and one graft maintained patency over 1 year. Commercial GelsoftTM grafts (Terumo Corp., Japan) are prepared with double-raschel knitted PET grafts coated with cross-linked gelatin. Therefore, combining the graft bases (PET or SF fibers) and the coating materials (SF sponges or cross-linked gelatin) could yield four types of grafts (SF/SF. SF/gelatin, PET/SF and PET/gelatin, shown as “base/coating” material, respectively). Fukayama et al.96 prepared the four types of small-diameter doubleraschel knitted vascular grafts with 1.5 mm diameter and 10 mm long and compared the biological properties after implantation in rat abdominal aortae. Two weeks after implantation, there were no significant differences among the types of grafts in biological properties evaluated through histopathologic examination. However, after 3 months, tissue infiltration based on area inside the graft walls was approximately 2.5 times higher in SF/SF 27 ACS Paragon Plus Environment


Page 28 of 42

grafts than in PET/gelatin grafts. Almost 100% endothelialization was achieved in SF/SF grafts, whereas only 50% was achieved in PET/gelatin. Therefore, SF demonstrated promising results as base and coating materials for small-diameter vascular prostheses. A similar conclusion was obtained from an in vitro evaluation of the grafts using human blood.149 The use of transgenic SF could be valuable in the further improvement of adhesion and anti-thrombogenic properties, as well as remodeling abilities in the development of smalldiameter vascular grafts. Small-diameter vascular grafts prepared with SF fibers incorporating two kinds of functional sequences including RGD and VEGF were transplanted into the abdominal aorta of rats.74 As mentioned previously, VEGF and RGD SF exhibited superior attachment of endothelial cells than in WT SF, whereas VEGF SF exhibited less platelet adhesion than in RGD and WT SF. The VEGF SF graft exhibited early endothelialization, good patency, and native tissue organization 3 months after implantation in rat abdominal aorta. Fukayama et al.99 developed a novel in vivo evaluation system for small-diameter SF grafts in rats to evaluate the endothelialization at the central region of the grafts appropriately. They prepared small-diameter double-raschel knitted SF grafts 3 cm long rather than 1 cm long and implanted them into abdominal aortae in order to eliminate the effects of endothelialization in the anastomosis sites. Endothelialization in the center of vascular grafts is thought to occur by the migration of endothelial cells from the anastomosis site or by endothelial progenitor cells from the blood.150–152 The WT SF and VEGF SF grafts were compared using the novel in vivo evaluation system. Remarkably higher endothelialization was observed in the VEGF SF graft than in the WT SF graft 3 months after implantation, i.e., 80% in the VEGF SF graft and 40% in the WT SF graft. Tanaka et al.100 examined differences in the porogen used to prepare SF sponges for coating SF grafts. They prepared smalldiameter double-raschel knitted SF grafts coated with three kinds of SF sponges used as porogens, including PEG, PGDE, and Glyc. Similar double-raschel knitted SF vascular grafts 28 ACS Paragon Plus Environment

Page 29 of 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


coated with SF sponge using ethanol (EtOH-SF graft) for insolubilization were also prepared for a comparison. The SF grafts were prepared to avoid dryness during the coating process, and the grafts were maintained in the hydrated state until implantation into the abdominal aorta of the rats. The grafts coated with the three kinds of SF sponges, with PEG, PGDE, and Glyc as porogens, exhibited higher tissue infiltration rates than the EtOH-SF graft and were superior in the formation of smooth muscle cells and vascular endothelial cell remodeling. In particular, endothelialization at the central part of the 3 cm grafts was complete 3 months after implantation for the PGDE- and Glyc-SF grafts (Figure 14).

Figure 14. Histological longitudinal images of CD31 staining of the central parts of the smalldiameter double-raschel knitted SF grafts with 1.5 mm diameter and 3 cm length 3 months after implantation (a) EtOH-SF, (b) PEG-SF, (c) PGDE-SF, and (d) Glyc-SF grafts. Arrows indicate vascular endothelial cells adhering to inner surfaces of the grafts. EtOH: ethanol; Glyc: glycerin; PEG: poly(ethylene glycol); PGDE: poly(ethylene glycol diglycidyl ether); SF: silk fibroin. Adapted with permission from ref.100. Copyright 2018 SAGE. Conclusion and Future Prospects Currently, the need for small-diameter artificial vascular grafts is increasing globally. 29 ACS Paragon Plus Environment


Page 30 of 42

Synthetic polymers such as ePTFE and PET are clinically applied for artificial vascular grafts with diameter ≥ 6 mm, but the materials pose challenges for vascular grafts with diameter

Advanced Silk Fibroin Biomaterials and Application to Small-diameter

Recommend Documents