Nano- to Microscale Porous Silicon as a Cell Interface for Bone-Tissue Engineering

An ideal material for orthopedic tissue engineering should be biocompatible, biodegradable, osteoconductive, osteoinductive, mechanically stable, and widely available. Porous silicon (PSi), a silicon-based material, fulfills these criteria. It is biocompatible and biodegradable, and supports hydroxyapatite (HA) nucleation. The micro-/nanoarchitecture of PSi may regulate cell behavior. The surface chemistry of PSi is flexible so that the interfacial properties between this material and living cells can be tailored easily by chemical modifications. Here, we report that PSi can support and promote primary osteoblast growth, protein-matrix synthesis, and mineralization. We also show that the osteoconductivity of PSi and other cellular responses can be controlled by altering the micro-/nanoarchitecture of the porous interface. With this material, we are closer to a functional biomaterial with both osteoconductivity and drug-delivery functions. Recently, tissue-engineering strategies using engineered biomaterials that support and promote bone-tissue growth have been proposed for reconstructive surgeries. The goal of a tissue-engineering approach is to repair and regenerate damaged human tissue with biomaterial-based devices. The approach requires functional cells derived from the target tissue, a matrix supporting those cells, bioactive molecules regulating cellular behavior, and the integration of this composite in the damaged tissue. Si, a semiconductor material, has the potential to achieve all the properties required for a tissue-engineering strategy. The physical and chemical properties of Si are widely known because of its wide use in the microelectronic industry. Moreover, sophisticated microfabrication techniques allow precise structures to be formed on Si substrates, some of which have been proposed for medical care. 16] The recent discovery of the biocompatibility, biodegradability, and bioactivity of PSi has opened the door for implantable applications of this Si-based material. After implantation of Si-based bioactive glass into rabbit bone, the elevation in Si concentration was only found at the implant site and not in other organs, and the implanted Si was efficiently excreted by urine. Furthermore, the large surface-to-volume ratio and the chemical flexibility of PSi makes it attractive for immobilizing bioactive molecules for drug-delivery purposes. These findings suggest that PSi can be a candidate for orthopaedic tissue engineering. Our investigations on the osteoconductivity of PSi were carried out using nanoscale (< 15 nm, NanPSi), mesoscale (ca. 50 nm, MesPSi), and macroscale (ca. 1 lm, MacPSi) pores in vitro. The PSi samples were produced by electrochemical etching of p-type Si wafers in HF-based electrolytes. The various pore configurations were achieved by changing the Si substrate, the electrolyte content, or the current density (see Supporting Information for detailed experimental information). As shown in Figure 1, MacPSi had pores with openings close to 1 lm; MesPSi had pores with pore openings around 50 nm; and NanPSi had a spongy porous structure with pore sizes under 15 nm. Unlike polished Si wafers, which do not degrade in cell media, PSi can be degraded in such a solution. Preliminary experiments showed that freshly-etched MesPSi degraded faster than MacPSi in cell media (see Supporting Information). The observation indicates that PSi, rather than Si, has potential in vivo degradation, and that MacPSi may be the most favorable candidate for bone-tissue engineering in terms of both biodegradability and stability. To protect PSi from gradual oxidation and degradation, a chemical oxidation in hydrogen peroxide was carried out after etching to form a thin oxide layer on the surface. Primary rat calvaria cells (osteoblasts) or rat osteosarcoma cells (ROS 17/2.8) were seeded onto PSi substrates for from 1 h to 5 weeks and the substrates and cells were assayed both qualitatively and quantitatively. Standard cell culture in 24-well polystyrene culture plates was used as a control. The adhesion of osteoblasts to PSi surfaces was quantified by direct counting of the attached cells. The viability of the attached cells was determined by an adenosine triphosphate (ATP)-based cell-viability assay. In adhesion studies (0.5– 4 h), PSi chips bound slightly fewer osteoblasts than the tissue-culture plate, but the difference was not statistically sigC O M M U N IC A IO N