Silicon biointerfaces for all scales

Design principles for the development of silicon biointerfaces enable the non-genetic, light-controlled modulation of intracellular Ca2+ dynamics, and of cellular excitability in vitro, in tissue slices and in mouse brains.

Intercellular and inter-tissue communication within organisms is mediated by signals that are biochemical (small molecules and proteins) and biophysical (mechanical and electrical)1,2. To harness these factors and modulate cellular-signalling processes, optical technologies such as optogenetics3 (the genetic insertion of light-sensitive proteins into cells) can be employed to translate an optical input into a biological response, in particular because light can be manipulated with exquisite spatial and temporal precision. However, these techniques often require genetic manipulation of the target cell or tissue to facilitate precise cellular activation, thereby limiting their use in vivo. Owing to the well-characterized photothermal, optoelectronic and photoacoustic properties of silicon4,5,6, biointerfaces based on this material offer a more amenable alternative for tapping into intracellular and extracellular signalling pathways alike. They can also be used to manipulate genetically unperturbed cells with high spatial and temporal resolutions7. Although Si-based devices can translate an optical stimulus into a targeted cellular response8, the relationship between the physiochemical phenomena occurring at the material/tissue interface and the size and shape of the Si structures remain undefined. This lack of knowledge has curtailed the application of devices based on Si that take full advantage of the material’s physical and chemical properties. Reporting in Nature Biomedical Engineering, Bozhi Tian and colleagues describe a set of quantitative design principles for the development of a range of Si-based biointerfaces with sizes ranging from nanometres to centimetres9. The work provides a foundation for developing Si-based systems aimed at manipulating biological pathways at the organelle, cellular and organ levels.

Tian and co-authors’ systematic approach focuses on two key design considerations: the structure and mechanics of the materials that facilitate close contact between device and target, and the material properties that efficiently translate optical stimulation to the desired cellular response. The authors quantitatively analysed a palette of design parameters, including the physical dimensions, metal doping and surface chemistry of the biointerfaces. They then correlated these properties to the processes occurring at the device/tissue interface, including induced currents, mechanical forces and localized heating. These efforts allowed them to propose general design principles for Si biointerfaces for the electrical manipulation of intracellular, intercellular and extracellular signals at the organelle, cellular and organ levels, respectively. Their findings were then implemented in Si-based devices, ranging from nanowires to dopant-modified microscopic membranes to macroscopic meshes (Fig. 1).

Fig. 1: Si-based biointerfaces enable optical control of cellular signalling at multiple scales.

a, The photothermal effect of i-type (intrinsic) coaxial Si nanowires causes the release of heat on illumination, which can be used to modulate intracellular and intercellular signalling at the single-cell or organelle levels. b, Silicon diode membranes (p-type–intrinsic–n-type) can be used to generate a current on illumination, for the intercellular and extracellular manipulation of cell or tissue cultures, such as brain-slice preparations. c, Silicon meshes on soft polymer substrates can conform to larger tissues (for example, they can be wrapped around organs, such as the brain) and thereby increase the contact area at the tissue/device interface. Optical stimulation of the mesh can trigger a response in nearby tissue.

Silicon nanowires, the smallest of the structures fabricated by Tian and co-authors, have been previously used for the intracellular recording and modulation of electrical signals in mammalian cells10. To translate these nanomaterials into an optically controllable biointerface, the authors synthesized coaxial Si nanowires by depositing nanocrystalline Si shells onto the nanowire backbone. The nanowires were then applied to a mixed culture of dorsal root ganglia containing sensory neurons and Schwann cells, and were preferentially internalized into the Schwann cells through phagocytosis. Illumination of the nanowires with an amber laser (592 nm) evoked a local temperature elevation via the photothermal effect that induced Ca2+ release from the cells’ internal stores. The authors also demonstrate that the nanowires can act as a sensor of the interactions between transport proteins and microtubules, and that active transport of the nanowires correlated with local Ca2+ dynamics. The photoacoustic properties of the nanowires were also exploited for the mechanical manipulation of the cytoskeleton: when human umbilical vein endothelial cells were cultured in the presence of Si nanowires, illumination at 592 nm led to the disruption of the microtubule networks.

Tian and colleagues also demonstrate the application of planar Si photodiodes for the remote modulation of cell-membrane potentials in ex vivo cultures and tissue slices. The researchers fabricated Si diode junctions (p-type–intrinsic–n-type) that, on illumination with 530-nm light, enabled the efficient separation of photogenerated carriers and their conduction towards the interface between the electrolyte and the biological target. The resulting capacitive current was sufficient to trigger a cellular response in cultured dorsal root ganglia cells and in ex vivo mouse brain slices, from which electrical signals were recorded in response to pulses of green light. Notably, patterned illumination of the Si interfaces afforded high spatiotemporal control over the electrical response, as only cells near the illuminated areas were affected.

Translating local potentials to more complex tissues in vivo requires the delivery of substantially larger currents from the Si interface to the target. To enhance both the capacitive and Faradaic currents at the Si biointerfaces, Tian and colleagues systematically decorated the Si diode junctions with metal nanoparticles. Gold, silver and platinum were tested, with gold proving the most effective at increasing the current at the interface. Silicon devices prepared as meshes on flexible porous polydimethylsiloxane membranes, which can conform to organ surfaces for improved electrical contact, could trigger light-evoked neural responses in the upper and middle layers of the mouse sensorimotor cortex. Moreover, illumination of the device near the forelimb region of the primary motor cortex triggered contralateral forelimb movements in anesthetized mice, which suggests that these interfaces can be used for the optical control of behaviour in wild-type rodents.

By providing a quantitative blueprint for the design of optically controllable Si biointerfaces that do not rely on transgenes, the principles outlined by Tian and colleagues will undoubtedly empower the design of new classes of devices for the optical manipulation of cell signalling across basic biophysics studies as well as tissue-engineering and neuroscience applications. Limitations associated with the operating principles and materials properties of Si devices, however, invite further investigation. For example, future designs could benefit from incorporating the cell-type specificity and optical-multiplexing capabilities offered by optogenetics. New surface chemistries or design motifs enabling more complex stimulation patterns to be generated at the biointerface will be instrumental in achieving the precision necessary to study complex physiological processes. Importantly, as for other techniques that rely on optical stimulation to control biological processes, Si biointerfaces are currently confined to studies in vitro or to shallow locations in vivo due to the scattering and absorption of visible light by tissue. This is often circumvented through the direct implantation of light sources within the organ of interest. Although effective, this approach is invasive and often accompanied by tissue damage around the device, by the elicited foreign-body response, and by the formation of insulating fibrotic tissue around the implants, resulting from chemical and mechanical mismatches at the tissue/device interface. Soft materials may serve as flexible waveguides or as substrates for microscale light-emitting devices with enhanced long-term biocompatibility11,12. Alternatively, multiphoton-based approaches commonly applied for deep-tissue imaging may enable the application of Si devices within more complex settings.

By illuminating the basic design principles required for fabricating functional Si-based biointerfaces, Tian and colleagues provide a roadmap for tailoring the physics and chemistry of Si to a specified biological application. This will aid in the development of new devices for the biophysical study of cell–cell communication, and may accelerate the development of future therapies for disorders characterized by disrupted signalling processes.


  1. 1.

    Jaalouk, D. E. & Lammerding, J. Nat. Rev. Mol. Cell Biol. 10, 63–73 (2009).

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  2. 2.

    Cogan, S. F. Annu. Rev. Biomed. Eng. 10, 275–309 (2008).

    Article  PubMed  CAS  Google Scholar 

  3. 3.

    Deisseroth, K. Nat. Methods 8, 26–29 (2011).

    Article  PubMed  CAS  Google Scholar 

  4. 4.

    Cao, L. et al. Nano Lett. 8, 601–605 (2008).

    Article  PubMed  CAS  Google Scholar 

  5. 5.

    Walter, M. G. et al. Chem. Rev. 110, 6446–6473 (2010).

    Article  PubMed  CAS  Google Scholar 

  6. 6.

    Chen, H. X. & Diebold, G. Science 270, 963–966 (1995).

    Article  CAS  Google Scholar 

  7. 7.

    Zhang, A. Q. & Lieber, C. M. Chem. Rev. 116, 215–257 (2016).

    Article  PubMed  CAS  Google Scholar 

  8. 8.

    Kotov, N. A. et al. Adv. Mater. 21, 3970–4004 (2009).

    Article  CAS  Google Scholar 

  9. 9.

    Jiang, Y. et al. Nat. Biomed. Eng. (2018).

    Article  Google Scholar 

  10. 10.

    Tian, B. A. et al. Science 329, 830–834 (2010).

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  11. 11.

    Park, S. et al. Nat. Neurosci. 20, 612–619 (2017).

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  12. 12.

    Park, S. I. et al. Nat. Biotechnol. 33, 1280–1286 (2015).

    Article  PubMed  PubMed Central  CAS  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Polina Anikeeva.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Park, S., Frank, J.A. & Anikeeva, P. Silicon biointerfaces for all scales. Nat Biomed Eng 2, 471–472 (2018).

Download citation

Further reading


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing