Date Published: July 06, 2017
Publisher: John Wiley and Sons Inc.
Author(s): Wanpeng Liu, Zhitao Zhou, Shaoqing Zhang, Zhifeng Shi, Justin Tabarini, Woonsoo Lee, Yeshun Zhang, S. N. Gilbert Corder, Xinxin Li, Fei Dong, Liang Cheng, Mengkun Liu, David L. Kaplan, Fiorenzo G. Omenetto, Guozheng Zhang, Ying Mao, Tiger H. Tao.
Precise patterning of biomaterials has widespread applications, including drug release, degradable implants, tissue engineering, and regenerative medicine. Patterning of protein‐based microstructures using UV‐photolithography has been demonstrated using protein as the resist material. The Achilles heel of existing protein‐based biophotoresists is the inevitable wide molecular weight distribution during the protein extraction/regeneration process, hindering their practical uses in the semiconductor industry where reliability and repeatability are paramount. A wafer‐scale high resolution patterning of bio‐microstructures using well‐defined silk fibroin light chain as the resist material is presented showing unprecedent performances. The lithographic and etching performance of silk fibroin light chain resists are evaluated systematically and the underlying mechanisms are thoroughly discussed. The micropatterned silk structures are tested as cellular substrates for the successful spatial guidance of fetal neural stems cells seeded on the patterned substrates. The enhanced patterning resolution, the improved etch resistance, and the inherent biocompatibility of such protein‐based photoresist provide new opportunities in fabricating large scale biocompatible functional microstructures.
Precise patterning of micro and nanostructures using polymer‐based biomaterials has widespread applications including drug release, degradable implants, tissue engineering, and regenerative medicine.1, 2, 3, 4 In this context, natural silk proteins obtained from cocoons of the silkworm Bombyx mori provide “green” alternatives to synthetic materials with advantages such as superior mechanical properties (strength and toughness), outstanding biocompatibility and biodegradability, and controllable water‐solubility and degradation rate.1, 5, 6, 7
Figure1 illustrates the material synthesis, functionalization, and photolithographic results of UV‐reactive silk l‐fibroin (UV–LC) resists. The B. mori silkworm cocoons were first cut into small pieces and degummed for 60 min to remove sericin using a previously reported process38 (Figure 1a,b). Formic acid was used to break the covalent disulfide bonds between H‐fibroin and l‐fibroin, and to separate silk fragments based on their different solubilities in formic acid without causing severe protein degradation.39, 40 The soluble fractions (i.e., l‐fibroin) were harvested and air‐dried (Figure 1c). The l‐fibroin was modified to be photoreactive by conjugating a photoreactive reagent of 2‐isocyanatoethyl methacrylate (IEM) to l‐fibroin’s side groups, yielding a photocrosslinkable UV–LC precursor (Figure 1d). The UV–LC precursor was then dissolved in 1,1,1,3,3,3‐hexafluoro‐2‐propanol (HFIP, Sigma Aldrich, St. Louis, MO). An organic photoinitiator of Irgacure 2959 (Sigma Aldrich, St. Louis, MO) was added 0.5% (w/v) into the UV–LC precursor solution to generate (and transfer) reactive species (free radicals in this case) when exposed to UV radiation (Figure 1e). The UV–LC resist solution (2%, w/v) was spin coated on a silicon or glass substrate to form a resist layer with a controllable thickness ranging from 50 nm to several micrometers which was then exposed through a photomask (Figure 1f). In this case, the UV–LC resist acted as a negative photoresist which can be crosslinked due to IEM in the presence of UV light (followed by the development step) to generate wafer‐scale micropatterns on silicon and glass substrates via standard UV photolithography (Figure 1g). UV–LC microstructures were tested as cellular substrates and for the spatial guidance of fetal neural stems cells which were seeded on micropatterned surfaces and incubated for 3 d. Cells tended to preferentially attach to the UV–LC protein patterns in comparison to the surrounding surface (i.e., silicon in this case) (Figure 1h, more details in Figure 5, also see Supporting Information). Note that the sensitivity of UV–LC resists can be readily tuned by regulating the IEM molecules conjugated into l‐fibroin. Additionally, the presence of unmodified amino acids can enable further functions (e.g., association with favorable cellular interactions and the production of multifunctional biomaterial architectures) via concurrent or subsequent modification strategies.41 In this study, the IEM molecules were intentionally designed to exceed the population of available amino acids conversion to fully occupy nearly all active group sites on the protein chains to better investigate the underlying mechanism of photo‐only‐induced formation of crosslinked silk micro/nanostructures.
In conclusion, we report on a precise protein P3 for wafer‐scale, high‐performance biopatterning using chemically modified well‐defined silk l‐fibroins as the photoresist material. The lithographic and etching performance of UV–LC and UV–Silk resists have been evaluated systematically and the underlying mechanisms have been thoroughly discussed. A general guidance on the synthesis and the use of silk l‐fibroin resist has been provided. The inherent biocompatibility and the enhanced patterning resolution along with the improved surface roughness and etching performance of such protein‐based resists offer new opportunities in fabricating large‐scale high‐precision biocompatible functional micro/nanostructures.
Synthesis and Purification of Light Chain Proteins: 60 min degummed silk fiber was weighed and dispersed in 98–100% formic acid at a range of concentrations (0.01–8%, w/v) for 30 min. The mixture was then centrifuged at 4000 rpm for half an hour to sediment the undissolved material. The supernatant was filtered using glass fiber filters to remove any remaining suspended particles/fibers. Then, the soluble fractions were left under a flow of air at room temperature to evaporate to constant weight. Note that the degumming conditions significantly affect the performance of UV–Silk resist but have much less effect on UV–LC resist.
The authors declare no conflict of interest.