Research highlights

Hybrid materials for sensitive immunosensing

Changes in the concentration of biomolecules, from DNA and RNA to proteins and peptides and even small molecule transmitters, often serve as indicators of illness and disease. Consequently, researchers have eagerly sought to develop new MEMS and microfluidic technologies to detect ever decreasing quantities of these molecules. While sensitivity in these platforms has greatly improved over the years, detection of these biomarkers in ever lower quantities may be the key to early diagnosis of rapidly-progressing and devastating diseases such as syphilis and acquired immune deficiency syndrome (AIDS). Nanomaterials have emerged as new alternatives to increase sensitivity and further push the limits of detection in these analyses. Graphene, in particular, is an attractive material due to its high carrier mobility as well as highly tunable conductance.^1,2 The integration of graphene into fluidic devices, however, has been hindered by its structural unreliability and irregular active areas, leading to poor sensitivity and selectivity for molecules of interest.

To overcome these issues, Kwon et al.³ describe the use of graphene micropatterns and conducting polymer nanoparticles in a novel hybrid immunosensing platform. The micropatterns, generated through direct patterning on catalytic metal substrates, self-assembly, and photolithography, are more stable and can be more easily integrated into MEMS and microfluidic devices than conventional graphene sheets. The hybrid platform was fabricated stepwise, with the graphene sheet first transferred from the metal growth substrate to a flexible plastic substrate. Source and drain electrodes were then patterned on either side of the sheet to create a field effect transistor (FET). Uniform carboxylated polypyrrole nanoparticles were subsequently attached to the graphene sheet to create the hybrid nanomaterial and increase the surface area for analyte binding. Lastly, the HIV-2 antigen, gp36, was covalently attached to the nanoparticles as a recognition binding moiety to interrogate system performance.

The authors initially characterized the system with high resolution transmission electron microscopy (HR-TEM), Raman, and FTIR spectroscopy to confirm the thickness of the graphene sheet (single layer) and the covalent attachment of both the nanoparticles and the antigen to the graphene and nanoparticles, respectively. They then determined that 0.023 mg of antigen was covalently attached to the nanoparticles versus 0.018 mg attached to the sheet alone, proving that the nanoparticles did increase the binding capacity of the sensor. The hybrid system was subsequently combined with liquid ion gating to examine device performance. Drain to source currents (I_ds) increased linearly with increasing gate voltage, indicating p-type (hole-transport) behaviour. When compared to n-type behaviour, the device performance in the p-type regime was found to offer better stability during operation and greater sensitivity.

Moreover, the group interrogated the FET system with increasing concentrations of HIV-2 gp36 antibody to examine target analyte binding. Antibody exposure resulted in real-time, concentration-dependent current responses with a minimum detectable level of 1 pM (versus 10 pM with graphene studded with antigen alone). This low detection limit is likely due to the combination of the increased surface area of the nanoparticles and the highly integrated electrical conduit between the conducting nanoparticles and graphene, which was able to easily transduce the electric field created by the antibody binding to changes in the I_ds current. Additional interrogation of the FET-type graphene micropattern nano-biohybrid-based immunosensors (GMNS) also yielded rapid and sensitive detection of multiple antibody concentrations (Fig. 1).

Fig. 1 (a) Schematic representation of the graphene-based biosensor system with integrated microchannels. (b) Drain to source current response (I_ds) of the graphene-based sensor to different concentrations of the HIV-2 gp36 antibody. Figure adapted and reprinted with permission from Kwon et al.³ Copyright © 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

In their paper, Kwon et al. demonstrated the sensitivity and rapid response of hybrid FETs generated from a combination of micropatterned graphene and polymer nanoparticles. When combined with a fluidic platform, the hybrids displayed excellent mechanical bendability, durability, and stable sensing, while also affording efficient electrical pathways to recognize target analyte binding events. Due to the platform’s low cost, rapid analysis times, and high sensitivity, it has a great potential for use not only in clinical settings and point of care diagnostics, but also for real-time portable environmental sensing to precisely detect disease biomarkers, toxins, pathogens, and environmental contaminants.

Reliable biosensors for low-cost protein sensing

Electrical biosensors are of great importance for various biomedical applications, such as protein sensing for early detection and diagnosis of numerous diseases including cancer.⁴ Biosensors are commonly fabricated from semiconducting materials such as silicon, AlGaN/GaN heterostructures or carbon nanotubes.⁵ Silicon-based biosensors, especially, are desirable due to their low cost, simple fabrication, and cytocompatible properties. However, their long term electrical drift, instability and measurement inaccuracy are drawbacks limiting their broad application. A majority of these problems arise from the diffusion of ions from biological fluids with high osmolarity into the gate oxides of Si-based biosensors. The physiological environment in which the biosensors operate contains Na⁺ and K⁺ mobile ions that might travel along the surface of active devices and penetrate into the devices. These mobile ions have the potential to induce variable drifts in the threshold voltage of the transistors and subsequently lead to inaccurate measurement of protein concentrations. Therefore, it is critical to reduce the mobility and permeability of the alkaline ions into the devices to enhance the accuracy of the biosensors and to develop long-term stable and low-cost silicon-based biosensors.

Ramesh et al. have recently presented a new sensor fabrication technology where they replaced the silicon dioxide layer with a high quality dielectric made of Al₂O₃.⁶ In this new configuration, the use of an Al₂O₃ film instead of the thermally grown SiO₂ allowed them to eradicate the issue of mobile ion interference in the biosensors during protein concentration measurements. Al₂O₃ was deposited on the surface of the silicon substrate using an atomic layer deposition process, with trimethyl aluminium and water as precursors. Photolithography was used to create several square electrodes with four different alternative topographies and configurations including floating gate, slots with and without holes and opaque electrodes to conduct a broad range of ion permeation experiments. The electrodes were patterned so that various concentrations of ions can permeate into the devices and allow the researchers to fully examine the effect of this approach in reducing the presence and mobility of alkaline ions. The long term stability of the devices was then studied by soaking some devices made with Al₂O₃ dielectric and others made with thermally grown SiO₂ insulating layers (as control) in simulated physiological environments for different periods of time. Parallel plate metal oxide semiconductor (MOS) capacitors were used to evaluate the effectiveness of the materials by measuring the ion content using capacitance-voltage (CV) measurements.

The in vivo physiological environment can be modelled in vitro by conducting the experiment in saline solution with 0.15 M concentration of Na⁺ and K⁺ ions. The authors quantified the permeation of mobile charges into the oxide dielectrics by using a triangular voltage sweep method. In this approach, the electrodes were stressed for 5 minutes with a 1 MV cm⁻¹ electric field at 250 °C to induce the movement of all the mobile ions to one side of the capacitor plate. Quasi-static capacitance–voltage measurements were then conducted to monitor the mobility of ions during the exposure to voltage. The mobilities of ions are known to change the polarity of the voltage and thus lead to the formation of a peak in the measured capacitance. The FET-based protein biosensor and the layout of the MOS test structure are schematically shown in Fig. 2.

Fig. 2 The schematic of the FET-based protein biosensor in which a sensing channel connects the source (S) to the drain (D) where the reference electrode (RE) is used to bias the transistor and the magnified view of the MOS capacitor configuration. Figure inspired by Ramesh et al.⁶

Triangular voltage sweep measurements on thermally grown SiO₂-based MOS devices indicated the presence of mobile ion peaks before and after soaking the electrodes in PBS. More importantly, the mobile ion peaks exhibited a linear increase corresponding to higher soaking times in PBS. For instance, the numerical integration of the peaks showed that the alkali mobile ions were enhanced by nearly ten-fold by increasing the soaking time from 30 to 90 minutes. This result indicated substantial penetration of alkali ions into the devices, which usually leads to electrical drift and dramatic fluctuations in the protein measurements. The use of Al₂O₃ based MOS devices, however, significantly reduced the mobility of ions.

All the MOS devices which had a 100 nm layer of Al₂O₃ showed no response to the alkali ion penetration, even after soaking the devices in PBS for 24 hours. This result was consistent for all four configurations of the devices which used Al₂O₃ as the dielectric layer. In addition, the authors found that the sensitivity of measurements improved by reducing the thickness of the Al₂O₃ layer from 100 nm to 10 nm without compromising the positive impact of this layer on reducing the movement of alkali ions. The results of triangular voltage sweep measurements confirmed that no mobile ion response was observed in devices with a thinner Al₂O₃ layer even after soaking them in PBS for 24 hours.

In summary, Ramesh et al. demonstrated that silicon-based devices made with Al₂O₃ insulators can overcome the issues of long term electrical drift and enhance the accuracy of measurements compared to conventional silicon protein biosensors. This modification in fabrication of the electrodes reduces the penetration and thus mobility of alkali ions from high osmolarity biological buffers. In addition, the sensitivity of the biosensors can be further optimized by decreasing the thickness of the Al₂O₃ layer down to 10 nm. These silicon-based transistors made with Al₂O₃ dielectrics are, therefore, deemed to be potential candidates for the fabrication of reliable and low-cost biosensors for in vivo protein detection.

Insect muscle bioactuator operating in air

Microactuators provide a driving force for applications on the microscale and have numerous biomedical applications such as miniature cardiac pumps,⁷ muscular thin films,⁸ and self-assembled microrobots.⁹ Experimental designs for these microactuators include both artificially and organically powered models. The latter have several advantageous characteristics such as energy efficiency, microscale size and the ability for self-repair.¹⁰ Typically rat cardiomyocytes have been used in these devices, whereas the use of mammalian cells require demanding cell culture conditions: the pH and the temperature of the environment need to be maintained similar to physiological conditions and the medium must be replaced every few days. Morishima and his colleagues have therefore created a microbioactuator that can be operated in medium as well as in air, and is powered by insect muscle tissue.¹¹ They chose an insect dorsal vessel (DV) tissue due to its ability to be cultured in a wide range of culture conditions. The microactuator can effectively function at temperatures from 5 to 40 °C and can operate continuously for 90 days when given excess medium.¹² This air-operating bioactuator (AOB) had a coating of L-paraffin which allowed it to overcome the issues arising from the evaporation of the medium and the drying of biological components.

In their recent paper, Akiyama et al.¹¹ reported a new bioacutator fabricated using standard lithography techniques to create separate molds for the tweezers and the capsule which were then filled with poly(dimethylsiloxane) (PDMS) and baked. The tweezers extended out of the capsule via a slit at one end (Fig. 3). By taking the capillary force into consideration, the tweezer slit was designed to be 4 mm by 2 mm so that even if the AOB was upside down in atmosphere, the medium would remain confined inside due to the surface tension forces. The capsule held 40 μL of medium which, based on the previous studies, would be sufficient to keep the DV tissue contracting for over 1.2 days. Optimal microtweezer thickness and the hinge width were determined to be 450 μm and 200 μm, respectively, using a finite element simulation software. Microtweezers with thinner and narrower hinges are more easily deformed by DV tissue contractions, but lead to decreased tweezer elasticity and lack an ability to reopen. DV muscle tissue was excised from larvae of the inch worm Thysanoplusia intermixta and immersed in the medium. The tweezer notches were hydrophilized, coated with Cell Tak (a cell and tissue adhesive for organic and inorganic surfaces) and then the DV tissue was wrapped around one arm and attached to the other at the appropriate tension.

Fig. 3 (a) Design of the AOB. Insect DV muscle tissue was attached at the notches and the microtweezers were loaded into the capsule. (b) The left side shows displacement from the starting position due to contraction. The right side shows von Mises stress during contraction and indicates that it is concentrated around the hinges just below the DV tissue and notches. (c) Photos of the AOB in medium (left) and in air (right). (d) The top graph shows the displacement of the tweezer arms during contraction in medium and in air. The bottom graph shows the resting gap between the tweezer arms in medium and in air. (e) The AOB working in air at room temperature placed on a finger. Figure reprinted with permission from the Royal Society of Chemistry from Akiyama et al.¹¹

Akiyama et al. first tested the bioactuator's performance submerged in medium. The tweezer gap was reduced from ~900 μm to ~600 μm with a frequency of 1.1 Hz. Although the force produced by the contracting DV tissue (19 μN) was sufficient to deform the tweezers, it was only 20% of the DV tissue’s contractile force. The researchers proposed that attaching the DV tissue between the tweezer notches at higher tension would increase the device efficiency. Tests in a dry environment showed the reduction of the tweezer gap down to ~200 μm at 0.89 Hz during contraction. Surface tension forces attracting the two tweezer arms towards each other were responsible for the increased contraction distance and were strong enough to overcome the weaker contraction force from the DV muscles due to the smaller resting length between the arms. The AOB remained functional for 40 minutes until a significant portion of the medium evaporated, but regained functionality as more medium was added. In order to increase the longevity of the device the researchers coated the capsule with a parylene film, a biocompatible, low vapor permeability polymer, and poured L-paraffin, a liquid hydrocarbon with low cytotoxicity and a lower surface tension than water, onto the medium–atmosphere interface. This procedure extended the AOB’s lifetime to 5.2 days: the evaporation of medium, although significantly reduced, caused the meniscus of the paraffin–atmosphere interface to decrease until paraffin covered the muscle tissue and prevented the exchange of cellular gas, nutrient and waste.

This is the first demonstration of a bioactuator that is functional in air and at room temperature, which opens up a new set of applications and enables novel research areas. By precisely controlling the contractions of the actuator via new optogenetic procedures using light-activated ion channels, tasks such as manipulating microbes, oocytes and single cells both in the medium and air are now possible. Due to its robust design and its ability to operate in a variety of environmental conditions, the hybrid bioactuator will find numerous applications in the field of autonomous micro and nano robotics.

References

1. L. H. Hess, M. Seifert and J. A. Garrido, Proc. IEEE, 2013, 101, 1780–1792.
2. A. Walcarius, S. D. Minteer, J. Wang, Y. Lin and A. Merkoci, J. Mater. Chem. B, 2013, 1, 4878–4908.
3. O. S. Kwon, S. H. Lee, S. J. Park, J. H. An, H. S. Song, T. Kim, J. H. Oh, J. Bae, H. Yoon, T. H. Park and J. Jang, Adv. Mater., 2013, 25, 4177–4185.
4. G. Zheng, F. Patolsky, Y. Cui, W. U. Wang and C. M. Lieber, Nat. Biotechnol.., 2005, 23, 1294–1301.
5. J. Wang, Electroanalysis, 2005, 17, 7–14.
6. A. Ramesh, F. Ren, P. R. Berger, P. Casal, A. Theiss, S. Gupta and S. C. Lee, Electron. Lett., 2013, 49, 450–451.
7. J. Kim, J. Park, S. Yang, J. Baek, B. Kim, S. H. Lee, E. S. Yoon, K. Chun and S. Park, Lab Chip, 2007, 7, 1504–1508.
8. A. W. Feinberg, A. Feigel, S. S. Shevkoplyas, S. Sheehy, G. M. Whitesides and K. K. Parker, Science, 2007, 317, 1366–1370.
9. J. Xi, J. J. Schmidt and C. D. Montemagno, Nat. Mater., 2005, 4, 180–184.
10. Y. Tanaka, K. Sato, T. Shimizu, M. Yamato, T. Okano and T. Kitamori, Biosens. Bioelectron., 2007, 23, 449–458.
11. Y. Akiyama, T. Sakuma, K. Funakoshi, T. Hoshino, K. Iwabuchi and K. Morishima, Lab Chip, 2013, 13, 4870–4880.
12. Y. Akiyama, K. Iwabuchi, Y. Furukawa and K. Morishima, Fabrication and evaluation of temperature-tolerant bioactuator driven by insect heart cells, San Diego, USA, 2008.

---