Title: Plasmonic Approach to Fluorescence Enhancement of Mesoporous Silica-Coated Gold Nanorods for Highly Sensitive Influenza A Virus Detection Using Lateral Flow Immunosensor
Journal: ACS Nano
IF: 17.1
Original link: https://doi.org/10.1021/acsnano.3c02651
Reporter: Chaoying Wang, Master of Grade 2022
Metal-enhanced fluorescence (MEF), which is a plasmonic effect in the vicinity of metallic nanoparticles, can be an effective strategy to improve the detection sensitivity of fluorescence-based LFIs.The key factors for obtaining a strong plasmonic effect include the distance and spectral overlap of the metal and fluorophore in the MEF system. In this study, MEF probes were designed based on core-shell nanostructures employing a gold nanorod core, mesoporous silica shell,and cyanine 5 fluorophore.To optimize the efficiency of MEF probes incorporated on the LFI platform (MEF-LFI), we experimentally and theoretically investigated the distance dependence of plasmonic coupling between cyanine 5 and gold nanorods by adjusting the shell thickness, resulting in significant fluorescence enhancement. The proposed MEF-LFI enabled highly sensitive detection of influenza A virus(IAV) nucleocapsid protein with a detection limit of 0.52 pg mL-1 within 20 min and showed high specificity and accuracy for determining IAV clinical samples. Overall, our findings demonstrate the potential of this method as an effective tool for molecular diagnosis under emergency conditions.
Antigen testing, which provides direct evidence of viral infection by detecting the presence of viral proteins in the early stages of infections, is an effective diagnostic tool. It provides benefits including significant cost advantage, simple interpretation with minimal training or infrastructure, and short turnaround time (≤20 min), unlike nucleic acid amplification tests, such as reverse transcription-quantitative polymerase chain reaction (RT-qPCR) and isothermal amplification. Lateral flow immunoassay (LFI), one of the most widely used analytical techniques for detecting target analytes, provides a rapid response and high specificity with easy operation. Hence, it is commonly used as an antigen test for viral analysis. However, the sensitivity of conventional LFI systems is inadequate, leading to false-negative results. Since slight changes in analyte concentrations cannot be simply distinguished from the color shades of gold nanoparticles in conventional LFI, there is a need to develop an effective analytical strategy to improve sensing performance.
Fluorescence is a widely used detection strategy in biomedical and biosensing applications. A core-shell structure, a single structure composed of two materials, offers many advantages with respect to MEF behavior. The shell controls the distance between fluorophores and metallic core nanoparticles, which permits fluorescence enhancement by the plasmonic effect, and prevents fluorescence quenching caused by proximity to the metal. The use of metallic nanostructures as the core favorably improves the optical properties of the fluorophore, resulting in an increased quantum yield with improved photostability. The inorganic silica shell provides excellent chemical and physical properties such as high stability, high dispersibility, and multifunctionality. Furthermore, mesoporous silica (mSiO2) shells, having a high surface area and periodically aligned pore structure, resolve the insufficient loading capacity issue of nonporous shells. The inclusion of a large number of fluorophores in the mesoporous structure has motivated numerous researchers to use mesoporous materials in devices involving the enhancement of fluorescence. However, these studies have limitations in obtaining maximally enhanced fluorescence emission due to the insufficient determination of optimum spectral overlap and the distance setting between fluorophores and metals. Consequently, advanced strategies are needed to effectively exploit the MEF phenomenon for detecting targets at low concentrations with enhanced fluorescence intensity in LFI systems.
In this study, we aimed to establish an effective MEF condition of metallic colloids and improve the sensing performance of the LFI system by using an optimized MEF condition. We studied the MEF phenomenon by incorporating a GNR core-mSiO2 shell (mSiO2@GNRs) to provide a strong plasmonic effect and a high fluorophore loading capacity. The MEF phenomenon was then optimized by investigating the effect of shell thickness on fluorescence behavior. To this end, we used both experimental and theoretical approaches to investigate the fluorescence behavior of cyanine 5 (Cy5), which has a high spectral overlap with GNRs (Cy5-mSiO2@GNRs). We utilized various mSiO2 shell thicknesses to elucidate how the distance between Cy5 and the GNR core affects the MEF enhancement factor (EFMEF). Experimental investigations demonstrated fluorescence enhancement behavior depending on the presence of GNR cores and the silica shell thickness. Theoretical results, wherein both Förster resonance energy transfer (FRET) and plasmon enhancement (PE) were considered, determined the specific distance between Cy5 and GNRs that provided the maximal enhancement of fluorescence intensity. After further optimization to improve the sensing performance, a MEF-based lateral flow immunosensor (MEF-LFI) used the optimized MEF probe to target the detection of IAV. Our system allowed for the highly sensitive and specific detection of IAV at low concentrations in clinical samples within 20 min. This platform can serve as a powerful POCT for rapidly screening suspected IAV and other viruses, especially during early stages of viral infections.
1. Preparation of Cy5-mSiO2@GNRs
In Figure 1a, the synthesis of Cy5-mSiO2@GNRs involves synthesizing GNR cores, mSiO2 shell coating, and loading of the Cy5 fluorophores.36−38 In the first step, a silica shell was coated on hexadecyltrimethylammonium bromide (CTAB)-capped GNRs (CTAB@GNRs) synthesized by using the seed-mediated growth method. The pellet of theCTAB@GNRs was dispersed in CTAB,which served as a softtemplate to generate the mSiO2 shell. The silica-coated GNRssynthesized by a sol−gel reaction based on the hydrolysis of tetraethyl orthosilicate (TEOS) were washed to form mSiO2@GNRs. The mSiO2@GNRs surfaces were functionalized with a succinic anhydride group using 3-triethoxysilylpropyl succinic anhydride (TESPSA) to achieve antibody immobilization via covalent bonding between the amino group of the antibodies and activated carboxyl group of mSiO2@GNRs after the ring-opening reaction.39 Finally, Cy5-mSiO2@GNRs were prepared by loading Cy5 into the mesoporous structure of the mSiO2@GNRs. We used an electrostatic adsorption strategy for aminated Cy5 and carboxylated mSiO2@GNRs to load large amounts of fluorophores into the porous structures to obtain strong fluorescence emission.
2. Characterization of Cy5-mSiO2@GNRs.
we analyzed the elemental composition of mSiO2@GNRs using energy-dispersive spectroscopy (EDS)and X-ray photo electron spectroscopy (XPS). The EDS mapping images revealed a uniform and homogeneous Si and Au elemental distribution of the silica shell (Figure 1c, green) and GNR cores (Figure 1c, red). Furthermore, the EDS line-scanning analysis of mSiO2@GNRs showed the relative proportion of Au and Si for each position in the line scan direction with a yellow arrow in the EDS mapping image (Figure 1d). The net intensity of Si and Au elements by location showed that Au appeared mostly in the center and was rarely present at both sides, thereby indicating the successful synthesis of mSiO2@GNRs with gold cores and silica shells (Figure S3). The XPS analysis confirmed successful silica coating by the Si 2p3/2 signal of the mSiO2@GNRs at 104.78 keV (Figure 1e).We examined the presence ofCy5 in the Cy5-mSiO2@GNRs. We using thermogravimetric analysis (TGA), which showed a weight loss during the thermal decomposition of Cy5. The loading amount of Cy5 was estimated to be 0.346% by weight (Figure 1f).
Figure 1. Characterization of Cy5-mSiO2@GNRs. (a) Schematic representation of the synthesis process of Cy5-mSiO2@GNRs. (b) Transmission electron microscopy images ofbare GNRs (top) and mSiO2@GNRs with 10.3±1.1 nm shell thickness (bottom). Scale bar: 200 nm. (c) Energy-dispersive spectroscopy (EDS) mapping images of a mSiO2@GNR showing Si (green), Au (red), and overlapped Si and Au elements. (d) EDS line-scanning profile for Si (green) and Au (red) in a single mSiO2@GNR. The direction of the line scan is indicated with a yellow arrow of the EDS mapping images. (e) X-ray photoelectron spectra showing the elemental composition of the mSiO2@GNRs. (f) Thermogravimetric analysis of mSiO2@GNRs and Cy5-mSiO2@GNRs. (g) Fluorescence spectra (λex = 600 nm) depend on each synthesis step for Cy5-mSiO2@GNRs. (h) Zeta potential showing changes in the surface charge of GNRs at each step during the synthesis.
3. Examination of Metal-Enhanced Fluorescence
A suitable nearby metallic GNR can modify the rate at which the fluorophore, Cy5, emits photons by increasing its intrinsic radiative decay rate (Figure 2a). In our system, the absorbance spectrum of GNRs (maximumλ = 628 nm) was found to almost overlap with the excitation of Cy5 (maximumλ = 639 nm), leading to an increase in the excitation and emission rate of Cy5 (Figure 2b). Additionally, lifetime measurement of the synthesized Cy5-mSiO2@GNRs and free Cy5 was performed by using a time-resolved fluorescence technique (Figure 2c). The fluorescence lifetime of Cy5 was shortened from 0.665 to 0.366 ns after loading into the mSiO2@GNRs, when the distance between Cy5 and the GNRs was 10.3 nm. Further experimental and theoretical studies of the MEF phenomenon were conducted. First, a synthetic strategy for etching the GNR cores at Cy5-mSiO2@GNRs was established to experimentally investigate the influence of the presence of metal cores on EFMEF (Figure 2d). The FRET efficiency (EFRET) gradually decreased asthe distance from the GNRsurface increased and then decreased rapidly as the distance further increased (Förster radius, R0 =5.935 nm) (Figure 2f, black line). The PE for incident light λ =640 nm decreased exponentially as the distance from the GNRsurface increased (Figure 2f, red line). To obtain the experimental EFMEF dependent on the distance between Cy5 and GNRs, we synthesized mSiO2@GNRs (Figure 2e, top) and their hollow counterparts mSiO2@GNRs (Figure 2e, bottom) with various mSiO2 shell thicknesses. After confirmation using an inductively coupled plasma-atomic emission spectrometer, the Si concentration among mSiO2@GNRs and hollow mSiO2@GNRs with the same shell thickness was adjusted to be similar, and we loaded Cy5 into both the mSiO2@GNRs and hollow mSiO2@GNRs, having equal shell thickness. Finally, we found that the EFMEF largely increased with an increase in shell thickness. The EFMEF under various mSiO2 shell thickness conditions reached a maximum at 10.3 nm of shell thickness and was approximately 5.68 in the solution phase and 7.85 on the nitrocellulose (NC) membrane of the LFI test strip. We compared the theoretical and experimental results, and the general trend ofthe two resultsappeared to be similar (Figure 2g).
Figure 2. Examination of metal-enhanced fluorescence (MEF). (a) Schematic representation of the Jablonski diagram without (top) and with (bottom) the metallic gold nanorod (GNR) demonstrating MEF using a radiative decay rate mechanism. Ε: rate of excitation without GNR. Εm:metal-enhanced excitation in the presence of GNR. Γm: radiative rate in the presence of GNR. km: nonradiative decay rate in the presence of GNR. (b) Absorbance and fluorescence spectra of GNR(absorbance, black line), excitation of Cy5 (blue line), and emission of Cy5 (red line). (c) Time-resolved fluorescence measurements for the lifetime decay of free Cy5 (black line) and Cy5-mSiO2@GNRs (red line). (d) Schematic representation of the investigation of the experimental MEF enhancement factor (EFMEF) with the preparation ofCy5-mSiO2@GNRs (left) and Cy5-hollow mSiO2@GNRs (right) by an etching process. The experimental EFMEF dependence on the shell thickness was determined by the fluorescence intensity ratio (I/Io) of Cy5-mSiO2@GNRs to Cy5-hollow mSiO2@GNRs in solution phase and on a NC membrane. (e) Transmission electron microscopy images showing individual mSiO2@GNR (top) and hollow mSiO2@GNR (bottom) with varying shell thickness. Scale bar: 20 nm. (f) Theoretical FRET efficiency (EFRET, black line) and plasmon enhancement (EPE, red line) affected by the interaction with a GNR under an incident light wavelength of 640 nm. A combination of EFRET and EPE determined the theoretical EFMEF: EFMEF= (1 − EFRET) × EPE. (g) Comparison of experimental and theoretical results for evaluating the EFMEF of Cy5 fluorescence by GNR cores. The values of the experimental EFMEF were obtained from three independent experiments.
4. Evaluation of IAV Detection System
The testing procedure for the MEF-LFI starts by applying 100 μL of an IAV sample solution onto the sample pad of the test strip (Figure 3a). The target analytes specifically bind to the detection antibody-immobilized Cy5-mSiO2@GNRconjugates, which are predried on the conjugate pad, thereby forming analyte-GNR conjugates. The conjugates are released from the conjugate pad and flow through the NC membrane driven by capillary forces and are captured by the IAV NP capture antibody at the test region on the NC membrane, thus forming a sandwich immunocomplex. The excess conjugates were reacted with anti-mouse IgG immobilized in the control region. A positive result was demonstrated by the appearance ofturquoise color in the test and control zones. The appearance of color only in the control zone indicated a negative result. The fluorescence signal was measured 20 min after adding the sample to the sample pad by irradiation from a red-light source and selective capture of Cy5 emission light from a 695 (±55) filter built into a charge-coupled device (CCD) camera-based imaging system. The resulting fluorescence intensity showed highly sensitive IAV NP detection at concentrations as low as 1 pg mL−1 (marked with an asterisk) with a wide dynamic linear range (0.001−100 ng mL−1) (Figure 3b).
5. Selectivity and Stability
Selectivity was evaluated with NPs from different viruses, including Middle East respiratory syndrome coronavirus (MERS-CoV), severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and influenza B (IBV). These virus NPs were prepared at 10 ng mL−1. The fluorescence and colorimetric intensity of IAV NP at 1 ng mL−1 were much higher than that of the other three viruses. Moreover, the intensity of a mixture containing the IAV NP and NPs of the other three viruses showed negligible interference with the IAV result (Figures 3d). These results demonstrate that the proposed system exhibited excellent selectivity for IAV NP. Furthermore, prepared test strips, which were stored at 4, 25, and 37 °C for 1 month, were used for investigating fluorescent conjugate stability by detecting 1 ng mL−1 of IAV NP. The intensity showed stable results, with coefficients of variation of 3.1% (4 °C), 3.6% (25 °C), and 8.2% (37 °C), respectively, indicating good temperature stability of the MEF-LFI platform (Figure 3e).
Figure 3. Detection of IAV using MEF-LFI. (a) Schematic representation of the sandwich immunoassay on a LFI test strip. (b) Fluorescence images showing the detection of serially diluted IAV NP. Fluorescence intensity of the test lines shown in images corresponding to antigen concentrations of 0.001−100 ng mL−1, showing sensitive antigen detection at concentrations as low as 1 pg mL−1. Error bars represent the standard deviation for three independent experiments. (c) Fluorescence images showing the detection of the serially diluted IAVH1N1 culture fluid. Fluorescence intensity of the test lines shown in images corresponding to cultured viral sample concentrations of 12.8−40000 plaque-forming units (pfu) mL−1, showing a limit of detection of 1.85 pfu mL−1 with a coefficient of determination of 0.9962. (d) Evaluation of selectivity for IAV (1 ng mL−1) and five other viruses (interferences; 10 ng mL−1). (e) Stability ofthe MEF-LFI for IAVNP detection evaluated by measuring the fluorescence intensity using the prepared and stored test strips at various temperature conditions (4, 25, and 37 °C) for 1 month. All asterisks indicate the lowest concentrations of samples distinguishable by the fluorescence measurement. Error bars represent the standard deviation for three independent experiments.
6. Clinical Validation.
For the clinical validation study, we used 23 patient samples as a discovery cohort (Figure 4a). The test line intensity obtained from IAV patient samples was significantly higher than that of negative samples (unpaired two-tailed Student’s t-test) (Figure 4b). Additional analysis using the receiver operating characteristic (ROC) curve with an area under the curve (AUC) of 1 indicated the high specificity and accuracy of the MEF-LFI for determining IAV clinical samples (Figure 4c). From the ROC curve analysis, the fluorescence intensity corresponding to maximum Youden’s index (Youden’s index = sensitivity + specificity −1) was taken as a cutoff value (fluorescence intensity = 138127). The cutoff value was applied to a separated validation set (29 patients with IAV and 29controls), which maintained high diagnostic power (1-β > 0.8) at a confidence level of95% (α= 0.05) (Figures 4d,e). The large dispersion of fluorescence intensity in positive sampleswas attributed to the various viral loads associated with the concentration of IAV NP，showing strong correlation with a Pearson’s coefficient (r) of 0.949 (regression equation: y = 1.48x + 12.71) (Figure 4f).
Figure 4. Profiling of clinical samples with the MEF-LFI platform was used for the determination of IAV. (a) Fluorescence intensity level for 23 patient samples (11 patients with IAV and 12 control patients with IBV). (b) Classification of clinical samples into positive (+) and negative (−) samples, resulting in a significant difference (*P < 0.05, unpaired two-tailed Student’s t-test). (c) ROC curve and AUC analysis. The fluorescence cut off value was determined from the ROC curve analysis of the discovery set, which corresponds to the maximum Youden’s index; fluorescence intensity =138127. (d)−(f) Validation cohort analysis. (d) Waterfall distribution of fluorescence intensity for all IAV clinical samples tested in the validation cohort (n = 58). The cutoff (dashed line) value was applied. (e) Significantly higher fluorescence signal was present in patient samples with IAV than non-IAV patient samples (***P < 0.001, unpaired two-tailed Student’s t-test). (f) Evaluation of analytical concordance between the proposed MEF-LFI and RT-qPCR. The results were positively correlated (Pearson’s coefficient, r = 0.949).
In this study, we developed an MEF-LFI platform for highly sensitive detection of IAV using strong fluorescent Cy5-mSiO2@GNRs. To optimize the morphology condition of Cy5-mSiO2@GNRs to maximize the EFMEF, we experimentally and theoretically investigated the fluorescence behavior variation contingent on the thickness ofthe mSiO2 shell. Strong plasmonic effects with high spectral overlap between Cy5 and GNRs, and a 10.3 nm optimized shell thickness, showed enhanced fluorescence intensity. This proposed MEF-LFI platform, which is based on the optimized condition providing maximum EFMEF, has practical advantages: (1) the enhanced fluorescence signal detected considerably low concentration of IAV (1.85 pfu mL−1 for IAV culture fluids); (2) the MEF-LFI showed no cross-reactivity with MERS-CoV, SARS-CoV-2, and IBV; and (3) it showed high accuracy (>99%) of clinical patient samples for IAV diagnosis. Taken together, our findings demonstrate the potential of this method as an effective tool in challenging fast-paced environments, including health emergency conditions.