Elsevier

Biosensors and Bioelectronics

Volume 117, 15 October 2018, Pages 530-536
Biosensors and Bioelectronics

A rapid readout for many single plasmonic nanoparticles using dark-field microscopy and digital color analysis

https://doi.org/10.1016/j.bios.2018.06.066Get rights and content

Highlights

  • New approach using a consumer-grade digital camera for single particle analysis.

  • Improved throughput for conducting single nanoparticles analysis (300x faster).

  • Detected 4.76 nM of IL-6 as a proof-of-concept for this analytical method.

Abstract

The integration of plasmonic nanoparticles into biosensors has the potential to increase the sensitivity and dynamic range of detection, through the use of single nanoparticle assays. The analysis of the localized surface plasmon resonance (LSPR) of plasmonic nanoparticles has allowed the limit of detection of biosensors to move towards single molecules. However, due to complex equipment or slow analysis times, these technologies have not been implemented for point-of-care detection. Herein, we demonstrate an advancement in LSPR analysis by presenting a technique, which utilizes an inexpensive CMOS-equipped digital camera and a dark-field microscope, that can analyse the λmax of over several thousand gold nanospheres in less than a second, without the use of a spectrometer. This improves the throughput of single particle spectral analysis by enabling more nanoparticles to be probed and in a much shorter time. This technique has been demonstrated through the detection of interleukin-6 through a core-satellite binding assay. We anticipate that this technique will aid in the development of high-throughput, multiplexed and point-of-care single nanoparticle biosensors.

Introduction

The trend in the development of next-generation biosensors is towards the detection of smaller and smaller amounts of materials, with the goal being the detection of a single molecule (Gooding and Gaus, 2016). The detection of single molecules has the potential to provide new information about biological systems by: i) uncovering the heterogeneity within samples, and ii) the potential to identify rare species or subtle changes in analyte concentration (Holzmeister et al., 2014). However, with the detection of single molecules comes the challenge of having collected sufficient single molecule events to be able to obtain quantitative information (Gooding and Gaus, 2016).

Owing to the characteristic of plasmonic nanoparticles (e.g. gold or silver nanoparticles), known as localized surface plasmon resonance (LSPR), the intense scattering of light by these nanoparticles can be directly imaged in the wide field of view of a dark-field microscope (Anker et al., 2008, Sriram et al., 2015). This widefield imaging of nanoparticles offers a unique opportunity to build a biosensor for single nanoparticle assays, capable of detecting and counting single molecules. Beuwer et al. (Beuwer et al., 2015) used a modified dark-field microscope to monitor the stochastic interactions of hundreds of single proteins on individual gold nanorods. Other approaches involve the use of single particle spectroscopy/spectrometry or laser imaging to obtain high resolution spectra from single nanoparticles, that monitor and analyse single nanoparticles in high spectroscopic detail (Chen et al., 2013, Taylor and Zijlstra, 2017). These methods are of great use in characterizing single nanoparticle interactions. However, for biosensing applications, such a high level of spectral detail is not as important as acquiring specific and relevant information from many single nanoparticles. Hence, an approach that is capable of rapid and high-throughput single nanoparticle analysis would be advantageous.

Recently, a new strategy to analyse and monitor the plasmonic properties of a nanoparticle known as colorimetric analysis has been developed. This approach analyses the color of scattered light from plasmonic nanoparticles images using dark-field microscopy. The wavelength of the scattered light is directly related to the LSPR of the nanoparticle and its size, shape and local environment. By analyzing the RGB (red, green and blue) of the light scattered by the nanoparticles, this technique has been used to identify the sizes of gold nanoparticles (Jing et al., 2012) and also detect single antibody-antigen binding events (Ungureanu et al., 2010, Verdoold et al., 2011). The technique was further modified to streamline the colorimetric detection, by measuring hue (a single color variable) to detect the formation of a DNA monolayer on gold nanoparticles (Zhou et al., 2016).

The recent development of single nanoparticle analysis using CCD detectors costing thousands of dollars for colorimetric analysis, has opened the possibility for high-throughput single nanoparticle assays to be conducted (Xie et al., 2017). However, in order to move towards achieving a portable device for point-of-care detection, it is important for the tools used in these approaches to be readily accessible and cost effective. Although consumer-grade CMOS cameras do not provide the same level of image quality as CCD detectors (Lustica, 2011), the technology has reached a point where image quality is sufficient to perform image processing tasks at the single nanoparticle level. The other main advantages that these CMOS cameras provide are significantly lower costs of production, the capability for fast read-out speeds and their wide availability in commercially available cameras and smartphones (Waltham, 2013). The last point is especially significant as this would allow for smartphones to be employed in biosensing, a step towards a point-of-care device.

In this study, we demonstrate the ability of a commercially available consumer-grade CMOS camera, when coupled with a dark-field microscope, to analyse the LSPR of single plasmonic nanoparticles. Dark-field images are analysed using an in-house developed MATLAB algorithm that can process thousands of nanoparticles within seconds, a signficant improvement on the longer times of previous spectroscopic techniques. Furthermore, through the calibration of the CMOS camera, we demonstrate that the λmax of up to 5000 nanoparticles can be obtained within seconds, without the use of a spectrometer. In order to demonstrate the ability of this technique to be employed in the detection of an analyte, interleukin-6 (IL-6) was chosen as the model analyte. IL-6 is known to be associated in the inflammatory and auto-immune processes of many diseases, such as Alzheimer's disease (Wang et al., 2015). Using a core-satellite binding assay, we demonstrate the potential of this technique to be employed in biosensing.

Section snippets

Chemicals and materials

Sodium citrate tribasic dihydrate, gold(III) chloride trihydrate (HAuCl4·3H2O), (3-aminopropyl)triethoxysilane (APTES), N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide (EDC), N-hydroxysuccinimide (NHS) and streptavidin were purchased from Sigma-Aldrich (Sydney, Australia). Small glass coverslips were purchased from Thermo Fisher Scientific (Australia) (22 × 22 mm, 0.16 – 0.19 mm thickness). Large coverslips for use in the flow cell were purchased from Thermo Scientific (USA) (Gold Seal™, 35 ×

Results and discussion

The light scattered by plasmonic nanoparticles is observed as a colored point-spread-function (PSF), of which the color is dependent on the size, shape, material and the local environment of the nanoparticle. Using an inexpensive CMOS-sensor based digital camera, which is readily available in many consumer products (e.g. smartphones), a new colorimetric technique was developed to allow for the rapid and high-throughput analysis of single plasmonic nanoparticles, as shown in Scheme 1. Using this

Conclusion

In summary, we have developed a new digital analysis technique that enables readily-available CMOS digital cameras, to perform rapid and high-throughput analyses of single plasmonic nanoparticles. This technique is based on observing the change in color of scattered light from plasmonic nanoparticles as seen under a dark-field microscope. By calibrating hue, measured by a digital camera with a CMOS sensor, against wavelength, and by using an in-house developed MATLAB program, up to 5000

Acknowledgements

The authors acknowledge the generous financial support from the Australian Research Council Centre of Excellence in Convergent Bio-Nano Science and Technology (CE140100036), an ARC Australian Laureate Fellowship (FL150100060 to J.J.G.) and UNSW Vice-Chancellor's Research Fellowships (to S.R.C.V.). This research used the facilities at the Electron Microscope Unit at UNSW. This work was supported through access to the Australian National Fabrication Facility (ANFF) Design House software at the

References (22)

  • J.J. Gooding et al.

    Single-molecule sensors: challenges and opportunities for quantitative analysis

    Angew. Chem. - Int. Ed.

    (2016)
  • Cited by (39)

    • A sliver deposition signal-enhanced optical biomolecular detection device based on reduced graphene oxide

      2022, Talanta
      Citation Excerpt :

      Nowadays, many signal amplification methods had been used in high sensitivity biological detection [11]. Such as, rolling circle amplification reaction [12,13], molecular beacons [14,15], and functionalization nanoparticles [16,17]. Therefore, finding a more effective signal amplification strategy was vital important to realize high sensitivity biological detection.

    • Artificial intelligence-assisted enumeration of ultra-small viruses with dual dark-field plasmon resonance probes

      2022, Biosensors and Bioelectronics
      Citation Excerpt :

      Therefore, we selected a screening system characterized by hue, saturation, and value (HSV) in which the color of each individual particle is dictated by a single variable, hue (Zhou et al., 2016). After HSV screening (Markhali et al., 2020; Sriram et al., 2018), large scattering contaminants (for instance, dust and impurities in the glass slides) and small object errors (for instance, single bright pixels or noise) are removed from the analysis by bandpass filters, and target particles within the hue threshold are retained. However, heterodimer nanostructures associated with chemical composition, shape, or size exhibit distinct diversity in the localized surface plasmon resonance effect (Weller et al., 2016; Tian et al., 2018).

    • Plasmonic biosensors for food control

      2021, Trends in Food Science and Technology
      Citation Excerpt :

      Information on SPR and LSPR can further be found in several reviews (Abdulhalim, 2013; Esfahani Monfared, 2020; Roh et al., 2011; Shalabney & Abdulhalim, 2011; Jingwei; Sun et al., 2020; Szunerits, Shalabney, Boukherroub, & Abdulhalim, 2012). Plasmonic properties of a single nanoparticle can be analyzed and monitored by using dark-field microscopy (Anker et al., 2010; Sriram et al., 2018; Sriram, Zong, Vivekchand, & Gooding, 2015). The color analysis of scattered light from plasmonic nanoparticles, which is directly related to the LSPR of the nanoparticle, its size, shape, and environment, allows to detect single analyte-ligand binding events.

    • Single particle detection of protein molecules using dark-field microscopy to avoid signals from nonspecific adsorption

      2020, Biosensors and Bioelectronics
      Citation Excerpt :

      In contrast, when the correct protein analyte, 10 ng/mL IL-6, was present then a 0.12 hue shift was recorded. It can be seen from the single particle analysis approach (Sriram et al., 2018a) that the 0.12 hue shift (Fig. 4B) was due to many single particles shifting whilst with the IL-11 (Fig. 4D) very few of the individual particles showed a hue change. Note that the absence of shifts due to incorrect binding of the reporter nanoparticles to the core particles not causing a shift is hue is despite that fact that reporter nanoparticles do nonspecifically adsorb to the underlying surface.

    • Enhanced color sensitivity by coupling of surface plasmon and fabry-perot resonances for spectrometer-free and label-free biosensing

      2020, Sensors and Actuators, B: Chemical
      Citation Excerpt :

      Though simple and compact, this black/white computation method is suffered from limited optical information and nonlinear changes of light intensity. Meanwhile, colorimetric analysis based on dynamic color changes of nanoparticles enables the detection of antigen/antibody binding [14], DNA concentration [15,16], and toxic metal ions [17]. In general, such sensing methods based on nanostructures and nanoparticles require time-consuming functionalization of the surface for high sensitivity.

    View all citing articles on Scopus
    1

    These authors contributed equally to this work.

    View full text