Optical Detection of CoV-SARS-2 Viral Proteins to Sub-Picomolar Concentrations

Abstract

The emergence of a new strain of coronavirus in late 2019, SARS-CoV-2, led to a global pandemic in 2020. This may have been preventable if large scale, rapid diagnosis of active cases had been possible, and this has highlighted the need for more effective and efficient ways of detecting and managing viral infections. In this work, we investigate three different optical techniques for quantifying the binding of recombinant SARS-CoV-2 spike protein to surface-immobilized oligonucleotide aptamers. Biolayer interferometry is a relatively cheap, robust, and rapid method that only requires very small sample volumes. However, its detection limit of 250 nM means that it is not sensitive enough to detect antigen proteins at physiologically relevant levels (sub-pM). Surface plasmon resonance is a more sensitive technique but requires larger sample volumes, takes longer, requires more expensive instrumentation, and only reduces the detection limit to 5 nM. Surface-enhanced Raman spectroscopy is far more sensitive, enabling detection of spike protein to sub-picomolar concentrations. Control experiments performed using scrambled aptamers and using bovine serum albumin as an analyte show that this apta-sensing approach is both sensitive and selective, with no appreciable response observed for any controls. Overall, these proof-of-principle results demonstrate that SERS-based aptasensors hold great promise for development into rapid, point-of-use antigen detection systems, enabling mass testing without any need for reagents or laboratory expertise and equipment.

Figure 1

Langmuir isotherm for the adsorption of biotinylated aptamer 1C,5′ to a streptavidin-coated BLI tip. Each data point represents the maximum BLI shift observed once the system had reached equilibrium, and the fitted isotherm is characterized by the parameters K_A = 0.384 μM and R_max = 1.845 nm.

Figure 2

BLI binding curves for the adsorption of biotinylated aptamers to a streptavidin-coated surface at a concentration of 5 μM (30–330 s), which were then exposed to a 500 nM solution of spike protein (390–690 s). For details of the aptamer sequences, see the main text.

Figure 3

BLI binding curves for the adsorption of biotinylated aptamers to a streptavidin-coated surface at a concentration of 5 μM (30–330 s), followed by 500 nM spike protein solutions with and without BSA (390–690 s).

Figure 4

Equilibrium BLI shifts as a function of concentration for spike protein and its receptor-binding domain (RBD) binding to biotinylated 1C,5′ aptamer immobilized on streptavidin-coated BLI tips, fitted to the Hill equation (eq 2) using the parameters reported in Table 1.

Figure 5

SPR sensorgrams for spike protein binding to thiolated 1C,5′ aptamer immobilized on a bare-gold SPR chip. Colored lines show averaged response curves based upon the raw data shown in gray, which are fitted to a first-order kinetic model. BSA control experiments were carried out at a concentration of 30 nM.

Figure 6

Average extrapolated equilibrium SPR responses at 10, 25, 50, 100, and 150 nM spike protein concentrations. Error bars represent the 95% confidence interval of the mean, obtained by fitting five replicates independently but keeping rate constants for lower concentration samples fixed at the values obtained at 100 nM. The fitted Hill model is characterized by the parameters R_eq, max = 570 nM, EC₅₀ = 174 nM and n = 1.

Figure 7

SPR sensorgrams for 2 and 5 nM spike protein binding to thiolated 1C,5′ aptamer immobilized on a bare-gold SPR chip. Colored lines show averaged response curves based upon the raw data shown in grey. BSA control experiments were carried out at a concentration of 30 nM.

Figure 8

SERS spectra of aptamer (A) with and without spike protein (P) immobilized on silver nanoparticles and deposited on an omniphobic surface at a series of different concentrations. In all cases, a 1:1 stoichiometric ratio of aptamer to protein was used.

Figure 9

Principal component loadings that describe the majority of the variability between sets of SERS spectra for surface-immobilized aptamers with and without spike protein.

Figure 10

SERS spectra of silver nanoparticles treated with 500 nM aptamer (A) in Tris buffer, 500 nM spike protein (P) in Tris buffer, and Tris buffer alone (Tris) and deposited on an omniphobic surface.

Figure 11

Differences in SERS intensities at 2872 cm^–1 (crosses) and 2912 cm^–1 (plus sign) between silver nanoparticles treated with aptamer alone (A) at a given concentration vs those treated with aptamer plus protein (A + P) at the same concentration. Detection limits are obtained by analysis of baseline scatter in the 3650–4000 cm^–1 region. Dotted horizontal lines indicate 99, 99.9, and 99.99% one-sided confidence intervals.