Biomolecules Registration Using an Optoelectronic Biosensor Based on LSPR
Table of Contents:
Peculiarities of LSPR Technique Biosensing Applications
The phenomenon of localized surface plasmon resonance, as a collective oscillation of free electrons in nanosized structures of high-conductive metals [22, 23], has become widely attractive to researchers since the development ofvarious methods of production of nanostructures that made it possible to fabricate relatively large arrays of nanoparticles and nanostructured films [24, 25]. In recent decades, numerous important applications of localized surface plasmon resonance have been demonstrated. In 1995, Kreibig  showed that the optical density of the immobilized monolayer of colloidal gold nanoparticles is a parameter sensitive to the refractive index of the surrounding solvent. Several experimental works related to the influence of the parameters of nanoparticle systems (e.g., shape, size, and interparticle distance) on their light extinction properties and observed optical dichroism were published [26-32].
Several basic methods for fabrication of nanostructure arrays to beusedassensitive elements in LSPR biosensors were demonstrated.
Two of the most accepted methods—deposition of nanoparticles from a colloidal solution and nanosphere lithography—were widely used to study the fundamental properties of these systems [27, 33- 35]. Van Duyne et al.  and Nathan et al. [37, 38] showed that silver or gold nanostructures on the surface of mica or glass could be fabricated using nanosphere lithography with monolayer formation, which could be optically detected and allowing the registration of the biomolecular interactions. G. Chumanov et al.  showed the possibility of creating stable monolayers of silver nanoparticles obtained from colloidal solutions using a transition polymer layer, which, by partially fixing a nanoparticle in its volume, does not prevent the appearance of LSPR on a polymer-free surface. The arrays of silver nanostructures manufactured by this method exhibit extremely narrow peaks in the absorption spectra, which can be used to create highly sensitive biosensors. The main disadvantages of these technologies are their limitations connected with the shape of nanoparticles, monodispersity, and reproduction, which negatively affect the possibility of their widespread use.
The nanoimprint lithography (NIL) technology was used to make gold nanoparticle arrays (NPAs) on glass substrates. It allows the creation of uniformly oriented and homogeneous NPA with controlled sizes, shapes, and interparticle distances, which can be the basis for the development of sensors for the detection of biomolecules. The most prominent advantage of this method is the use of matrix templates for the fabrication of nanostructures, making NIL a highly efficient and inexpensive process that allows the creation of templates with relatively large linear sizes and a resolution of less than 10 nm [40, 41]. This method can represent the general basis for the rapid prototyping of different types of NPAs (shapes, sizes, compositions, and interparticle distances) and enables simultaneous production of more than a billion monodisperse nanoparticles directly on a variety of substrates in a consistent, single process that is fully reproducible. The use of one-dimensional diffraction grating for producing two-dimensional templates allows obtaining three-dimensional nanostructures of a given shape, which form an NPA with the desired optical properties.
Among many areas of possible use of highly ordered NPAs, application in biosensors looks like one of the most promising due to the ability to detect interactions between biomolecules in real time [42, 43]. In order to develop a sensor structure for the registration of biomolecules, the NIL method was used to produce arrays of gold nanostructures with parallelepiped shape obtained from onedimensional gratings (see Subsection 184.108.40.206 and Section 3.3.1). As a result, arrays of nanostructures with strictly defined dimensions (120 x 130 x 50 nm), shape, and interparticle distances were obtained.
Using this approach, a method for measuring biomolecular interactions in real time using a UV-visible range spectrophotometer  is presented below. The developed approach has minimal technological requirements, namely, the presence of UV-visible spectrophotometer and peristaltic pump. The ability to biomolecular recognition depends on the change in the extinction spectrum of the NPA as a function of biomolecular adsorption onto the gold surface. Although NPAs are a completely non-selective sensor platform, a high degree of selectivity to the analyte can be achieved by the specificity of the attached ligands and the passivation of the sensor surface to non-specific binding [45, 46].
Atomic force microscopy (AFM) images and a section of 120 nm x 130 nm x 50 nm gold nanostructures with a distance between the nanoparticles of 237 nm are shown in Fig. 4.5. It should be noted that the AFM technology does not allow for a correct estimation of the distance between the NPA due to the limitations concerning the geometry of the AFM tip.
Figure 4.5 AFM image and section of Au NPA. Reprinted from Ref.  under a Creative Commons Attribution-NoDerivatives 4.0 International License. Figure caption was adapted.
The nanostructured biosensors based on LSPR spectroscopy operate in the same way as their SPR analogues by transducing the refractive index changes on the surface of a high-conductive metal into a plasmonic response (a shift in wavelength on which an SPR is observed). Biosensors are real-time liquid sensors, so initial testing of the NPA sensitivity to the change in the refractive index of the surrounding liquid medium was performed. For this purpose, aqueous solutions of glycerin with different refractive indices were used. Measurements for registering changes in the refractive index were carried out in the flow mode, with the registration of real-time changes in the extinction of light. The simplified scheme of LSPR biosensor setup is shown in Fig. 4.6.
Figure 4.6 Simplified scheme of LSPR biosensor setup: 1 — halogen lamp; 2, 10 — mirrors; 3 — polarizer; 4 — gold nanoparticles array; S — flow cell; 6 — glass substrate; 7, 8 — sample holders; 9,11 — lenses; 12 — fiber-optic cable; 13 — spectrometer; 14 — computer. Reprinted from Ref.  under a Creative Commons Attribution-NoDerivatives 4.0 International License. Figure caption was adapted.
The results obtained for calibrated glycerin solutions clearly show that the optical extinction of the NPA structures is sensitive to the change in the refractive index of the surrounding liquid medium (Fig. 4.7). A near-linear dependence of extinction on the right slope of the extinction peak (A = 660 nm) on the refractive index was observed, which is related to the LSPR wavelength shift (Fig. 4.7, inset). From the point of view of maximizing the LSPR response, it was found for most cases that it is more efficient to use the wavelengths of the right side of the LSPR peak.
Figure 4.7 The real-time kinetic dependence of light extinction by NPA when the refractive index of a liquid medium changes. Inset: dependence of extinction on refractive index. Reprinted from Ref.  under a Creative Commons Attribution-NoDerivatives 4.0 International License. Figure caption and inset were adapted.
Study of a Biomolecular Antigen-Antibody Reaction and Molecular Recognition Using LSPR Biosensors
To evaluate the kinetic response of the LSPR biosensor to the adsorption of real biological materials, NPAs were used to detect a biospecific reaction between the protein BSA (molecular weight
= 67 kDa) and anti-BSA immunoglobulin (molecular weight = 150 kDa). Experimental results are shown in Fig. 4.8, which shows the real-time kinetic dependence of light extinction (sensogram) for a specific reaction between BSA and anti-BSA, dissolved in a PBS buffer solution, with different concentrations of the anti-BSA solution. Figure 4.9 shows respective changes in the light extinction spectra.
Figure 4.8 Real-time kinetic dependence of light extinction by nanorectangle NPA for the BSA-anti-BSA specific reaction with different concentrations of anti-BSA solution.
Figure 4.9 (a) Light extinction spectra for nanorectangle NPA in air, in a buffer
solution and for BSA-anti-BSA specific reaction with different concentrations of the anti-BSA solution, (b) Zoomed in portion of the spectra detailing the extinction shifts.
As it can be seen from Figs. 4.8 and 4.9, a clear response to the specific BSA-anti-BSA reaction occurs when an anti-BSA solution is injected at a concentration of 80 ng/mL. In addition, a high degree of reaction specificity was observed during the experiment after washing the IgG layer with a 0.1 M HC1 solution when about 90% of the antibodies were eliminated, while the BSA layer completely remained on the surface of the sensor. As it can be seen from the sensogram, the developed technique allows the regeneration of a sensitive element.
4.5 Comparative Study of SPR and LSPR
Comparative Study of SPR and LSPR Techniques for Small Molecule Detection
To recognize the ability of SPR sensors for small molecule detection,
SPR experiment with thiourea (molecular weight = 76 Da) was performed using the saline-sodium citrate (lxSSC) buffer as the base solution. The results of experiment (Fig. 4.10) reveal low recognition limit of about 0.8 pM, whereas the expressed SPR response starts
Figure 4.10 Real-time kinetic dependence of SPR angle for detection of different concentrations of thiourea (molecular weight = 76 Da).
from the thiourea concentration of 4 цМ. Due to the presence of the thio group in the structure of this compound, the removal of molecules was not revealed after washing with buffer solution due to the strong chemical binding with gold surface. Moreover, no saturation of SPR response was observed even for the highest concentration of 0.5 mM used in the experiment, which indicates the high binding capacity for the surface of SPR slides.
Figure 4.11 Real-time kinetic dependence of light extinction of random NPA for detection of different concentration of thiourea (molecular weight = 76 Da). Inset: AFM image of a random array of gold nanostructures on a glass substrate fabricated by thermal annealing.
It should be noted that LSPR biosensor may be more suitable for the registration of biomolecules with small molecular weight due to more developed surface of gold compared with conventional SPR sensors and ability to fill the volume between nanostructures with small biomolecules that can easily penetrate and adsorb on their side faces, where distribution of local electric field is still strong. The experiment with thiourea detection was repeated under the similar conditions (Fig. 4.11) using a random array of gold nanostructures on a glass substrate (Fig. 4.11, inset) fabricated by thermal annealing (see Section 3.3.1). The sensogram of experiment expressed noticeable difference in the behavior of LSPR response in comparison with SPR. The LSPR sensor clearly recognized the low limit concentration of thiourea molecules (0.8 pM); however, the saturation of response was observed with higher concentration of molecules.
The result of the experiment proves the above assumption and shows that LSPR biosensors can register the nanomolar concentration of molecules with small molecular weight even using random NPA arrays. The obtained results suggest the promising use of random NPA for the construction of LSPR biosensors, primarily due to the simpler manufacturing technology compared with NIL. At the same time, it is necessary to keep in mind the possible deviation of parameters between the samples of random NPAs due to the existing dispersion of nanostructures in size and shape as well as the lower long-time stability of parameters due to the absence of adhesive chromium layer (see Section 3.3.1).
Specific studies described in this chapter demonstrate the potential of the SPR method as a basis for building high-sensitive SPR and LSPR biosensors capable of real-time registration of biomolecular interactions. Both sensor platforms can register biospecific interactions and exhibit concentration-dependent response, which can be calibrated for obtaining various molecular absorption parameters, for example, biomolecular layer mass.
In the first study, the experimental approach for the 3D quantification of the surface concentration of biological molecules from SPR data was proposed. SPR measurements of biospecific reaction between IgG and anti-IgG molecules on the Au surface were performed and the surface concentration of biomolecular layer was estimated. This approach may be useful to determine or estimate the orientation of biomolecules relative to the surface of SPR sensors for a more comprehensive understanding and quantitative estimation of biomolecular layers.
The second study presented developed LSPR biosensors based on the approach to fabricate noble metal nanoparticle arrays using nanoimprint lithography and thermal annealing. The capability of ordered and random Au nanostructure arrays for real-time detection of adsorption and interaction of biomolecules with different molecular weight was demonstrated. The performed experiments showed promising prospects of the LSPR method for the exploitation of NPA structures as the basis for biomolecular detection and monitoring of specific biomolecular interactions.