Long term observation of de novo adipogenesis using novel bioabsorbable implants with larger size in a porcine model.

Introduction: The regeneration of adipose tissue in patients after breast cancer surgery would be desirable without the use of growth factors or cells to avoid potential recurrence and metastasis. We reported that prolate spheroidal-shaped poly-L-lactic acid (PLLA) mesh implants of approximately 18-mm polar diameter and 7.5-mm greatest equatorial diameter containing collagen sponge (CS) would be replaced by regenerated adipose tissue after implantation, thereby suggesting an innovative method for breast reconstruction. Our study aimed to evaluate the adipose tissue regeneration ability of implant aggregates in a porcine model. Methods: We prepared implant aggregates consisting of thirty PLLA mesh implants containing CS packed in a woven poly (glycolic acid) bag. The implant aggregates were inserted under the mammary glands in the porcine abdomen for a year. Single and double groups were classified by inserting either one or two implant aggregates on each side of the abdomen, respectively. Results: In both groups, the volume of the implant aggregates decreased over time, and the formation of adipose tissue peaked between 6 and 9 months. Histologically, the formation of adipose tissue was confirmed in the area that was in contact with native adipose tissue. Conclusions: Our implant aggregates could induce the autologous adipose tissue after long term implantation in vivo, without the use of any growth factor or cell treatment, presenting a potential novel method of breast reconstruction.

Effects of RGD-fused silk fibroin in a solution format on fibroblast proliferation and collagen production.

BACKGROUND: Collagen production in fibroblasts is important for skin tissue repair. Cell-adhesive Arg-Gly-Asp (RGD) peptides immobilized on scaffolds stimulate fibroblast collagen production, but RGD peptides in solution exhibit opposite effects. Transgenic silkworm technology enables the design of fusion positions for RGD peptides in silk fibroin molecules. The effect of RGD-fused silk fibroin in solution on fibroblast cell activity remains unclear. OBJECTIVE: To clarify the effects of RGD peptides fused to silk fibroin heavy (H)-chain or light (L)-chain on fibroblast proliferation and collagen production when RGD-fused silk fibroin proteins were added to the culture medium. METHODS: Silk fibers with RGD-fused H-chains (H-RGD) or L-chains (L-RGD) were degummed, dissolved, and dialyzed to prepare H-RGD or L-RGD aqueous solutions, respectively. These solutions were added to the fibroblast medium, and their proliferation and collagen production were quantified. RESULTS: Both L- and H-RGD stimulated fibroblast proliferation at a similar level, even in a solution format, but L-RGD promoted fibroblast collagen production significantly, indicating the synergistic effect of the native H-chain and RGD-fused L-chain. CONCLUSION: RGD-fused silk fibroin in solution stimulated fibroblast proliferation and collagen production, depending on the fusion position of the peptides.

Preliminary report of de novo adipogenesis using novel bioabsorbable implants and image evaluation using a porcine model.

Our bioabsorbable poly-L-lactic acid (PLLA) mesh implants containing collagen sponge are replaced with adipose tissue after implantation, and this is an innovative method for breast reconstruction. In this preliminary study, we investigated the formation of adipose tissue and evaluated the process via multimodal images in a porcine model using an implant aggregate to generate the larger adipose tissue. The implant aggregate consists of PLLA mesh implants containing collagen sponge and a poly-glycolic acid woven bag covering them. We inserted the implant aggregates under the porcine mammary glands. Magnetic resonance imaging (MRI), ultrasonography (USG), and 3-dimensional (3D) surface imaging and histological evaluations were performed to evaluate the formation of adipose tissue over time. The volume of the implant aggregate and the formed adipose tissue inside the implant aggregate could be evaluated over time via MRI. The space within the implant aggregate was not confirmed on USG due to the acoustic shadow of the PLLA threads. The change in volume was not confirmed precisely using 3D surface imaging. Histologically, the newly formed adipose tissue was confirmed on the skin side of the implant aggregate. This implant aggregate has the ability to regenerate adipose tissue, and MRI is an appropriate method for the evaluation of the volume of the implant aggregation and the formation of adipose tissue.

Enhanced ß2-microglobulin binding of a "navigator" molecule bearing a single-chain variable fragment antibody for artificial switching of metabolic processing pathways.

Kidney dysfunction increases the blood levels of ß2-microglobulin (ß2-m), triggering dialysis-related amyloidosis. Previously, we developed a navigator molecule, consisting of a fusion protein of the N-terminal domain of apolipoprotein E (ApoE NTD) and the α3 domain of the major histocompatibility complex class I (MHC α3), for switching the metabolic processing pathway of ß2-m from the kidneys to the liver. However, the ß2-m binding of ApoE NTD-MHC α3 was impaired in the blood. In the current study, we replaced the ß2-m binding part of the navigator protein (MHC α3) with an anti-ß2-m single-chain variable fragment (scFv) antibody. The resultant ApoE NTD-scFv exhibited better ß2-m binding than ApoE NTD-MHC α3 in buffer, and even in serum. Similar to ApoE NTD-MHC α3, in the mice model ApoE NTD-scFv bound to the liver cells' surfaces in vitro and accumulated mainly in the liver, when complexed with 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC). Both ApoE NTD-MHC α3 + DMPC and ApoE NTD-scFv + DMPC significantly switched the ß2-m accumulation in mice from the kidneys to the liver, but only the ApoE NTD-scFv + DMPC group showed a significantly higher ratio of ß2-m accumulation in the liver versus the kidneys, compared with the control group. These results suggest that the enhanced ß2-m binding activity of the navigator molecule increased the efficiency of switching the metabolic processing pathway of the etiologic factor.

Initial immune response to a FRET-based MMP sensor-immobilized silk fibroin hydrogel in vivo.

To investigate the initial immune response to biodegradable silk fibroin (SF) hydrogels in vivo, a Förster/fluorescence resonance energy transfer (FRET)-based sensor was developed to detect matrix metalloproteinase (MMP) activity (FRET-MMPS) and immobilized to SF hydrogel. FRET-MMPS immobilized to SF hydrogel in vitro displayed intra-molecular FRET more than inter-molecular FRET, and MMP activity was detected through a decrease in FRET signal intensity. Then, the SF hydrogel modified with FRET-MMPS was implanted into mice subcutaneously, and it was observed that the FRET signal intensity decreased significantly soon (< 3 h) after implantation. Although the intensity exhibited a sharp decrease toward 24 h post-implantation, histological evaluation proved that bulk-level hydrogel degradation, such as breakdown, was mainly caused by macrophages and foreign body giant cells on a timescale of weeks. These results indicated that, immediately upon implantation, active MMPs reached the SF hydrogel and began cleaving SF networks, which might result in the loosening of the networks and then enabled immune cells, such as macrophages, to start the bulk-level hydrogel degradation. The sensor clarified the initial immune response to SF hydrogels and will provide clues for designing the biodegradation behaviors of scaffolds for regenerative medicine. STATEMENT OF SIGNIFICANCE: Silk fibroin (SF) materials are degraded gradually by the immune response. Immune cells, such as macrophages, break down implanted SF materials on a timescale of weeks or months, but the initial (< 24 h) immune response to SF materials remains unclear. In this study, SF hydrogels modified with Förster/fluorescence resonance energy transfer (FRET)-based matrix metalloproteinase (MMP) sensors were implanted in mice and within 3 h post-implantation, the SF hydrogels were degraded by MMPs. Although this molecular-level biodegradation was not correlated with the hydrogel breakdown, the MMPs were likely to loosen the SF networks to enable immune cells to infiltrate and degrade the hydrogel. This is the first study to unveil the initial stage of immune response to biomaterials.

Endothelial cell adhesion and blood response to hemocompatible peptide 1 (HCP-1), REDV, and RGD peptide sequences with free N-terminal amino groups immobilized on a biomedical expanded polytetrafluorethylene surface.

Blood compatibility generally requires two contradictory characteristics: reduced protein/platelet adhesion and excellent endothelium-related cell affinity. To understand the effect of cell adhesion peptides on blood compatibility, the peptides REDV, RGD, and hemocompatible peptide-1 (HCP-1) were immobilized on an expanded polytetrafluorethylene (ePTFE) surface and evaluated in vitro, in situ, and in vivo. Since the terminal amino groups of functional peptides often have an important effect, a cysteine residue was added to the C terminal and used for immobilization to keep the terminal amino groups free. Maleimide groups were added to carboxylic groups of highly hydrophilic and biologically inert (bioinert) polymer chains grafted onto ePTFE and coupled with cysteine residues. In vitro tests revealed that free N-terminal HCP-1 and RGD-immobilized surfaces improved the adhesion and spread of human umbilical vein endothelial cells (HUVECs), while, unexpectedly, a free N-terminal adjacent to REDV suppressed cell affinity. In situ evaluation with a porcine closed-circuit system for 2 h showed that no platelets adhered to the modified ePTFE sutures due to the bioinert graft chain containing phosphorylcholine groups. Simultaneously, leukocyte-related and endothelium-related cells were observed on RGD-immobilized ePTFE sutures because RGD was recognized by broad types of cells. These cells were not observed on the HCP-1- and REDV-immobilized ePTFE sutures, which may be due to insufficient exposure time. HCP-1-modified ePTFE graft implantation in a porcine femorofemoral (FF) bypass model for 10 days showed that the thrombus layer was clearly mitigated by HCP-1 immobilization. This study suggests that the HCP-1-immobilized ePTFE surface has potential for long-term application by mitigating thrombus and supporting endothelial cell adhesion.

Artificial switching of the metabolic processing pathway of an etiologic factor, ß2-microglobulin, by a "navigator" molecule.

Metabolic pathways in the body are highly specific. Dysfunction of a metabolic pathway triggers the accumulation of its target substance. For example, kidney failure results in increased ß2-microglobulin blood levels, causing dialysis-related amyloidosis. Previously, we proposed a novel therapeutic concept, that is a removal of an etiologic factor of metabolic disease by artificial switching of its metabolic processing pathway, and tested this concept using in cultured cells. However, the feasibility of artificial metabolic switching in vivo remained unknown. Here, we show that a newly developed "navigator" molecule changes the metabolic processing pathway of ß2-microglobulin from the kidney to the liver in mouse. The artificial metabolic switching is achieved by the capture of the etiologic factor by the navigator, which then steers the etiologic factor to hepatic lysosomes via low-density lipoprotein receptors. These findings demonstrate that navigator-based artificial metabolic switching can be a therapeutic strategy for various diseases caused by metabolic disorders.

Adipose tissue regeneration in a 3D-printed poly(lactic acid) frame-supported space in the inguinal region of rats.

BACKGROUND: Adipose tissue engineering has been studied as an alternative to current options for breast reconstruction, such as lipofilling, flap reconstruction, and silicone implants. Previously, we demonstrated that a poly(L-lactic acid) mesh containing a collagen sponge, containing neither cells nor growth factors, could be filled with the regenerated adipose tissues when implanted in rodent models. However, the main factor contributing to adipogenesis remained unclear. OBJECTIVE: We aimed to clarify whether adipogenesis can be achieved by the space provided by the mesh or by the bioactivity of collagen. METHODS: A three-dimensional (3D) poly(lactic acid) (PLA) frame, which was stiff enough to maintain its shape, was fabricated by 3D printing. The frame with (PLA+ColI) or without (PLA only) a type I collagen hydrogel was implanted in the inguinal region of rats for up to 12 months. Adipose tissue regeneration in the PLA only and PLA+ColI groups was evaluated histologically. RESULTS: The 3D PLA frame maintained its structure for 12 months in vivo and oil red O (ORO)-positive adipose tissues were regenerated in the frame. No significant difference in the ORO-positive area was detected between the PLA only and PLA+ColI groups. CONCLUSION: The space supported by the frame was a key factor in adipogenesis in vivo.

Anti-platelet adhesion and in situ capture of circulating endothelial progenitor cells on ePTFE surface modified with poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC) and hemocompatible peptide 1 (HCP-1).

We recently reported in vitro suppression of platelet adhesion on expanded polytetrafluoroethylene (ePTFE) by surface grafting of poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC). However, this may be inadequate for long-term hemocompatibility of blood-contacting biomaterials, and it has led us to develop a strategy of circulating mononuclear cell-capture. ePTFE was treated with argon (Ar) plasma, and grafted with 2-methacryloyloxyethyl phosphorylcholine (MPC) and methacrylic acid (MAA), by glycidyl methacrylate (GMA)-anchored graft polymerization. Next, it was immobilized with integrin α4ß1-positive circulating blood cell-specific peptides, i.e., the traditional arginine-glutamic acid-aspartic acid-valine (REDV), and our original hemocompatible peptide-1 (HCP-1). Both the surfaces retained the anti-platelet property just like the PMPC-grafted surface, and revealed considerable affinity to human umbilical vein endothelial cells (HUVEC), which is a well-known in vitro integrin α4ß1-positive model. Better HUVEC spreading and proliferation was also confirmed, in terms of the cell extension property. Since coagulation and endothelialization on the materials compete in the body, they cannot be properly evaluated separately, in vitro. They were assessed by using an in situ porcine closed-circuit system for 18 h in the present study. Our findings suggest that poly(MPC-co-MAA) is a great ePTFE surface modifier, exhibiting good hemocompatibility in association with REDV/HCP-1 immobilization, which suppresses anti-platelet adhesion and enhances circulating cell capture simultaneously.

De novo adipogenesis using a bioabsorbable implant without additional cells or growth factors.

Adipose tissue regeneration in breast cancer patients without additional growth factors or adipose-tissue-derived stromal cells is desirable because of the possibility of recurrence and metastasis. We report that a poly-L-lactic acid (PLLA) mesh implant containing a collagen sponge (CS) maintained the internal space in vivo for up to 12 months and substituted for adipose tissue. We developed a PLLA capsule that maintained the internal space longer than that of PLLA mesh and compared adipose tissue formation at 12 and 24 months after implantation between the PLLA mesh with CS implant and the PLLA capsule implant with or without CS in a rabbit model. After 12 months, all implants maintained the internal space, and the adipose tissue that formed in all implant groups was larger than that in the control group. At 24 months, PLLA mesh maintained the internal space just as well as that at 12 months, while the PLLA capsule collapsed and accumulated a large number of macrophages. The formed adipose tissue in the PLLA mesh group was maintained up to 24 months; however, those in two PLLA capsule groups decreased and showed no difference from the control group. In conclusion, the internal space of the PLLA mesh implant with CS was substituted for adipose tissue at 12 months and sustained the formed adipose tissue after 24 months. The PLLA mesh implant containing CS is a desirable bioabsorbable implant that can be replaced by autologous adipose tissue after implantation in vivo without using any growth factors or cells.

Biodegradation of injectable silk fibroin hydrogel prevents negative left ventricular remodeling after myocardial infarction.

In the present study, we investigated the optimal material properties of a tunable hydrogel for the treatment of myocardial infarction (MI) along with the therapeutic mechanism. We developed silk fibroin (SF) hydrogels with the same physical properties (e.g., stiffness) but different biodegradation rates via a peptide modification and evaluated the effect of hydrogel biodegradation on the prevention of negative left ventricular (LV) remodeling in a rat model of MI. LV enlargement was attenuated to a greater extent by injection of a slowly degrading unmodified SF hydrogel as compared to a rapidly degrading peptide-modified SF (SF + Pep) hydrogel for up to 12 weeks post-injection. This could not be explained by the mechanical stabilization of the LV wall by the injectant since the SF and SF + Pep hydrogels degraded completely within 4 weeks and the two groups showed no difference in LV wall thickness at 12 weeks. The SF group had dense randomly aligned collagen fibers whereas the SF + Pep group had sparse fibers surrounding the LV. These results suggest that randomly aligned fibrous tissues formed during/after biodegradation of a slowly degrading SF hydrogel may be more resistant to LV pressure and thus prevent LV enlargement.

Short-term evaluation of thromboresistance of a poly(ether ether ketone) (PEEK) mechanical heart valve with poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC)-grafted surface in a porcine aortic valve replacement model.

Improved thromboresistance of mechanical valves is desired to decrease the risk of thromboembolism and thrombosis and reduce the dosage of anticoagulation with a vitamin K antagonist (e.g., warfarin). For several mechanical valves, design-related features are responsible for their improved thromboresistance. However, it remains unclear whether material-related features provide a practical level of thromboresistance to mechanical valves. Here, we studied the effect of a bileaflet valve made of poly(ether ether ketone) (PEEK) with a poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC)-grafted surface (PEEK-g-PMPC). PMPC is a well-known thromboresistant polymeric material. A short-term (<26 h) porcine aortic valve replacement model using neither an anticoagulant nor an antiplatelet agent showed that the PEEK-g-PMPC valve opened and closed normally with an allowable transvalvular gradient. Unlike an untreated PEEK valve, no thrombus formed on the PEEK-g-PMPC valves on gross anatomy examination in addition to the absence of traveled thrombi in the kidney and lung tissues. Material (PEEK-g-PMPC)-related thromboresistance appeared to decrease the risk of thromboembolism and thrombosis for patients with mechanical valves. However, thromboresistance of the PEEK-g-PMPC valve requires improvement because fibrous fouling was still observed on the leaflet. © 2019 Wiley Periodicals, Inc. J Biomed Mater Res Part A: 107A: 1052-1063, 2019.

A surface graft polymerization process on chemically stable medical ePTFE for suppressing platelet adhesion and activation.

An effective surface grafting method for chemically inert and elaborately porous medical expanded-polytetrafluoroethylene (ePTFE) was developed. Although surface graft polymerization onto basic polymeric biomaterials has been widely studied, successful modification of the ePTFE surface has been lacking due to its high chemical resistance. Herein, we succeeded in surface graft polymerization onto ePTFE through glycidyl methacrylate (GMA) as a bridge linkage following argon (Ar) plasma treatment. The epoxy group of GMA was expected to react with the peroxide groups produced on ePTFE by Ar plasma exposure, and its methacrylic groups can copolymerize with various monomers. In the present study, we selected 2-methacryloyloxyethyl phosphorylcholine (MPC) as a model monomer and the blood compatibility of modified ePTFE was evaluated. Two sequences of surface grafting were compared. In a two-step graft polymerization, GMA was first immobilized onto Ar plasma treated ePTFE, and then MPC was polymerized. In a one-step graft copolymerization, MPC and GMA were mixed and copolymerized simultaneously onto Ar plasma treated ePTFE, resulting in a poly(MPC-co-GMA) (PMG) graft surface. The roughness of the node-and-fibril structure of ePTFE was reduced by the uniform polymer layer, and the modified ePTFE had a good hydrophilic nature even after being stored in an aqueous environment for 30 days. The indispensable GMA in graft polymerization improved the surface grafting on ePTFE. The one-step and two-step graft polymerization methods could decrease the number of adhered platelets, and almost inhibit platelet activation. We concluded that graft polymerization with the GMA linker provides a novel strategy to modify the chemically inert ePTFE surfaces for functionalizing as new medical devices.

Vascular induction and cell infiltration into peptide-modified bioactive silk fibroin hydrogels.

In hydrogel-based soft tissue engineering, vascular induction into a hydrogel as well as long-term volume retention is essential to maintain tissue shape and function without causing necrosis in the deeper part of the hydrogel. A silk fibroin (SF) hydrogel shows a sufficiently high mechanical strength to maintain its shape during implantation for a month, but it has not been well evaluated whether it has vascular-inducing bioactivity to achieve its replacement by vascularized tissues. Here, we produced a vascular-inducing peptide (VIP) containing an endothelial cell (EC)-adhesive REDV and vascular endothelial growth factor-mimicking QK peptides to modify the SF hydrogel. In vitro experiments showed that the modification of the SF hydrogel with VIP changed only biological properties of the hydrogel due to the bioactivity of VIP. Subcutaneous implantation of SF hydrogels in rats revealed isotropic EC migration into the hydrogels, which was followed by infiltration of macrophages and fibroblasts. Since these macrophages and fibroblasts appeared to degrade the SF network and to produce collagen, respectively, SF hydrogels were replaced gradually by regenerated tissues. VIP accelerated cell infiltration and doubled the formation of blood vessels in the regenerated tissue. These results suggest the potential of the VIP-modified SF hydrogel as a material for soft tissue engineering applications.

Early tissue formation on whole-area osteochondral defect of rabbit patella by covering with fibroin sponge.

Large osteochondral defects have been difficult to repair via tissue engineering treatments due to the lack of a sufficient number of source cells for repairing the defect and to the severe mechanical stresses affecting the replacement tissue. In the present study, whole-area osteochondral defects of rabbit patella were covered and wrapped with a fibroin sponge containing chondrocytes, with or without Green Fluorescent Protein (GFP) transgenic marking, on the surface facing the osteochondral defect. Five of eight osteochondral defects that were covered with the chondrocyte-seeded fibroin sponges showed hyaline cartilage-like repair containing no fibroin fragments at 6 weeks after surgery. The repaired tissue showed a layer formation, which showed intensive safranin-O and toluidine blue staining, and which showed positive type II collagen immunostaining. The average surface coverage of the repaired cartilage was 53%. On average, 48% of the cells in the repaired tissue were derived from GFP transgenic chondrocytes, which had been seeded in the fibroin sponge. The fibroin-sponge covering had the potential to allow the early repair of large osteochondral defects. © 2016 Wiley Periodicals, Inc. J Biomed Mater Res Part B: Appl Biomater, 104B: 1474-1482, 2016.

Development of a FRET-based recombinant tension sensor to visualize cell-material interactions.

A Förster/fluorescence resonance energy transfer (FRET)-based molecular tension sensor was originally reported by the fusion of intracellular molecules, which has contributed to the elucidation of cell mechanotransduction. However, it is still unclear whether recombinant tension sensors can detect forces in the extracellular environment. Here, we developed a recombinant FRET-based tension sensor (rFRET-TS) and immobilized it to a glass surface. Fibroblasts seeded onto the surface likely bound to an RGDS peptide fused to one terminal of the rFRET-TS, and intra-molecular FRET was dominantly observed on the sensor-immobilized surface compared to inter-molecular FRET. Time-lapse FRET imaging showed that the rFRET-TS enabled the real-time visualization of forces between cells and material surfaces that stemmed from focal adhesion formation and actin cytoskeletal reorganization. This sensor is expected to be useful for clarifying cell-scaffold mechanical interactions by its insertion between protein molecules of the scaffold, which will provide clues for the control of cell behavior in/on scaffolds.

Silk fibroin sponges with cell growth-promoting activity induced by genetically fused basic fibroblast growth factor.

Transgenic silkworm technology has enabled the biological properties of silk fibroin protein to be altered by fusion to recombinant bioactive proteins. However, few studies have reported the fabrication of genetically modified fibroin proteins into three-dimensional spongy structures to serve as scaffolds for tissue engineering. We generated a transgenic silkworm strain that produces fibroin fused to basic fibroblast growth factor (bFGF) and processed the fibroin into a spongy structure using a simple freeze/thaw method. NIH3T3 mouse embryonic fibroblasts grown on bFGF-fused fibroin sponges proliferated and spread out well, showing half the population doubling time of cells cultured on wild-type fibroin sponges. Furthermore, the number of primary rabbit articular chondrocytes growing on bFGF-fused fibroin sponges was around five-times higher than that of the wild-type control at 3-days post cell-seeding. As the physical properties of wild-type and bFGF-fused fibroin sponges were almost identical, it is suggested that bFGF fused to fibroin retained its biological activity, even after the bFGF-fused fibroin was fabricated into the spongy structure. The bFGF-fused fibroin sponge has the potential for widespread application in the field of tissue engineering, and the method of fabricating this structure could be applicable to other recombinant bioactive fibroin proteins.

Recombinant production and film properties of full-length hornet silk proteins.

Full-length versions of the four main components of silk cocoons of Vespa simillima hornets, Vssilk1-4, were produced as recombinant proteins in Escherichia coli. In shake flasks, the recombinant Vssilk proteins yielded 160-330mg recombinant proteinl(-1). Films generated from solutions of single Vssilk proteins had a secondary structure similar to that of films generated from native hornet silk. The films made from individual recombinant hornet silk proteins had similar or enhanced mechanical performance compared with films generated from native hornet silk, possibly reflecting the homogeneity of the recombinant proteins. The pH-dependent changes in zeta (ζ) potential of each Vssilk film were measured, and isoelectric points (pI) of Vssilk1-4 were determined as 8.9, 9.1, 5.0 and 4.2, respectively. The pI of native hornet silk, a combination of the four Vssilk proteins, was 4.7, a value similar to that of Bombyx mori silkworm silk. Films generated from Vssilk1 and 2 had net positive charge under physiological conditions and showed significantly higher cell adhesion activity. It is proposed that recombinant hornet silk is a valuable new material with potential for cell culture applications.