- Journal List
- HHS Author Manuscripts
- PMC8852303
As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice
Biosens Bioelectron. Author manuscript; available in PMC 2023 Mar 15.
Published in final edited form as:
Biosens Bioelectron. 2022 Mar 15; 200: 113918.
Published online 2021 Dec 25. doi:10.1016/j.bios.2021.113918
PMCID: PMC8852303
NIHMSID: NIHMS1768142
PMID: 34990957
Zheyu Wang,1 Qingjun Zhou,1 Anushree Seth,1 Samhitha Kolla,2 Jingyi Luan,3 Qisheng Jiang,3 Priya Rathi,1 Prashant Gupta,1 Jeremiah J. Morrissey,5,6 Rajesh R. Naik,4,* and Srikanth Singamaneni1,6,*
Author information Copyright and License information PMC Disclaimer
The publisher's final edited version of this article is available at Biosens Bioelectron
Associated Data
- Supplementary Materials
Abstract
Novel methods that enable facile, ultrasensitive and multiplexed detection of low molecular weight organic compounds such as metabolites, drugs, additives, and organic pollutants are valuable in biomedical research, clinical diagnosis, food safety and environmental monitoring. Here, we demonstrate a simple, rapid, and ultrasensitive method for detection and quantification of small molecules by implementing a competitive immunoassay with an ultrabright fluorescent nanolabel, plasmonic fluor. Plasmonic-fluor is comprised of a polymer-coated gold nanorod and bovine serum albumin conjugated with molecular fluorophores and biotin. The synthesis steps and fluorescence emission of plasmonic-fluor was characterized by UV-vis spectroscopy, transmission electron microscopy, and fluorescence microscopy. Plasmon-enhanced competitive assay can be completed within 20 minutes and exhibited more than 30-fold lower limit-of-detection for cortisol compared to conventional competitive ELISA. The plasmon-enhanced competitive immunoassay when implemented as partition-free digital assay enabled further improvement in sensitivity. Further, spatially multiplexed plasmon-enhanced competitive assay enabled the simultaneous detection of two analytes (cortisol and fluorescein). This simple, rapid, and ultrasensitive method can be broadly employed for multiplexed detection of various small molecules in research, in-field and clinical settings.
Keywords: competitive assay, small molecule detection, Plasmonic enhanced fluorescence, multiplexed detection
1. Introduction:
Detection and quantification of low molecular weight compounds, including biomolecules such as peptides, drugs, hormones and antibiotics, as well as organic molecules such as mycotoxins, pesticides and veterinary drugs, is critical in human health, industrial and environmental monitoring.(Dietrich et al. 2010; Li et al. 2017) Simple, sensitive and specific methods enabling rapid quantification of target small molecules will significantly benefit environmental surveillance, safety assurance of agricultural products, biomedical research, and clinical diagnosis.
Considering that the target analytes are often dispersed in complex chemical mixtures,(Glass et al. 2006) detection and quantification methods often rely on the multiple purification steps and time and resource-intensive analytical methods.(Wang et al. 2013) For example, separation and purification of target analytes through chromatography and analysis using optical spectroscopy or mass spectrometry have been well established and widely used for trace analysis.(de Lima et al. 2016; Lee et al. 2017; Lin et al. 2015) However, time-consuming and tedious sample preparation process as well as the requirement of bulky and expensive instruments preclude the application of such methods in resource-limited settings and in-field use. Therefore, there is a dire need for ultrasensitive sensing methods that can be employed in resource-limited and point-of-care settings.
Immunoassays have received increasing attention for the detection of small molecules, owing to their simplicity, specificity, and quantitative nature.(Anfossi et al. 2015; Fu et al. 2017; Watanabe et al. 2013) In contrast to conventional sandwich format, immunoassays for the detection of small molecules rely on the competitive binding between reference and sample analytes to the corresponding antibodies. Unfortunately, in many cases, the weak signal generated by conventional reporters (e.g., HRP, Au nanoparticles, molecular fluorophores, quantum dots) limits of the sensitivity of these assays and precludes the detection of low abundant small molecule analytes.(Chen et al. 2008; Espina et al. 2004; Luan et al. 2018; Wu et al. 2021) To improve the assay sensitivity, extensive efforts have been dedicated to enhancing the intensity of colorimetric or fluorescence signals, including the integration with immuno-polymerase chain reaction (PCR).(Cheng et al. 2012; Du et al. 2016; Li et al. 2021; Liu et al. 2014; Peteu et al. 2014) Although detection sensitivity is enhanced, these approaches are complex, time-consuming and labor-intensive. Digital assays, involving the counting of single analyte molecules instead of average signal intensity, have also been employed to improve the sensitivity.(Wang et al. 2018; Zhang and Noji 2017) However, due to the weak signals generated by conventional colorimetric substrates or fluorophores, digital assays necessitate partitioning the samples into femto-liter to nano-liter volume microwells or droplets, which inevitably complicates the assay and read-out procedure and increases cost-per-sample.(Rissin et al. 2010)
In this work, we demonstrate a plasmonically-enhanced competitive fluoroimmunoassay for detection and quantification of small molecules. In contrast to conventional approach, we employed plasmonic-fluor as an ultrabright and highly specific fluorescent nanolabel to improve the sensitivity of the competitive assay. Plasmonic-fluor exhibits more than 6000-fold brighter fluorescence signal compared to conventional fluorophores and has been demonstrated to improve the sensitivity of various bioassays by more than two orders of magnitude, including fluorescence linked immunosorbent assay (FLISA), multiplexed bead-based fluoroimmunoassays and flow cytometry.(Luan et al. 2020b) In this report, for the first time, we demonstrate the application of plasmonic fluor as an ultrabright nanolabel in competitive immunoassay with 10-fold improvement in LOD compared to enzymatic labels. We also demonstrate a spatially-multiplexed competitive assay for simultaneous detection of multiple target analytes. To the best of our knowledge, this is also the first demonstration of a digital immunoassay harnessing plasmon-enhanced fluorescence, which further improves sensitivity compared to the analog read-out. (Figure 1)
Figure 1.
Schematic illustration depicting plasmonic-fluor enhanced (A) singleplex and (B) multiplex competitive assays. Design and characterization of plasmonic-fluor: (C) Representative schematic illustration of plasmonic-fluor as an ultrabright nanolabel. (D) Normalized visible–NIR extinction spectra of plasmonic-fluor 800. Inset: TEM images of gold nanorods for plasmonic-fluor 800. (E) Fluorescence images and corresponding intensity of streptavidin-CW800 before and after applying plasmonic-fluor 800. Data are mean ± s.d.
2. Experimental section:
2.1. Synthesis of plasmonic fluor:
Plasmonic-fluors were synthesized following a previously reported procedure.(Luan et al. 2020a)
Synthesis of AuNRs:
For plasmonic-fluor 800 and plasmonic-fluor 650, two types of AuNRs, whose LSPR wavelength are around 760 nm and 630 nm, were prepared through a seed-mediated method, respectively.(Lee and El-Sayed 2005) Briefly, to first prepare the seed solution, 250 μl of 10 mM HAuCl4 (Sigma-Aldrich, 520918), and 600 μl of 10 mM ice-cold NaBH4 solution (Sigma-Aldrich, 71321) were sequentially added into 9.75 ml of 100 mM hexadecyltrimethylammonium bromide (CTAB) (Sigma-Aldrich, H5882), under vigorous stirring at room temperature. The color change from yellow to brown indicates the formation of seed. To prepare gold nanorods-760, 0.5 ml of 10 mM AgNO3 (Sigma-Aldrich, 204390), 2 ml of 10 mM HAuCl4, 0.22 ml of 0.1 M ascorbic acid (Sigma-Aldrich, A92902) and 0.9 ml of 1 M HCl (1 M, 0.9 ml) (Sigma-Aldrich, H9892) were sequentially added in to 38 ml of 100 mM CTAB, as the growth solution. Gold nanorods-630 were prepared in a similar approach, except that the volume of AgNO3 was reduced to 0.3 ml. Subsequently, 50-fold diluted seed solution was added into growth solution and allowed to react overnight in dark. AuNRs were collected through centrifugation at 6000 rpm to remove the supernatant and redispersed in nanopure water for further use.
Conjugation of Biotin and dye onto BSA:
Bovine serum albumin (BSA) was sequentially conjugated with biotin and dye via EDC/NHS chemistry. Firstly, 2 mg pf NHS-PEG4-Biotin (Thermo Scientific, A39259) was added into 2.2 ml of 5 mg/ml BSA (Sigma-Aldrich, A7030) in 1X PBS. After one hour, BSA-biotin conjugation was purified through a desalting column (Thermo Scientific, 89892, 7000 MWCO). To prepare BSA-Biotin-Cy7.5 conjugation, 100 μl of 1 M potassium phosphate dibasic solution (K2HPO4, Sigma Aldrich, P3786) was added into 1 ml BSA-Biotin solution to raise the pH above 9. Subsequently, 25 μl of 4 mg/ml NHS-Cy7.5 was added to the mixture, followed by two-hour incubation at room temperature. BSA-Biotin-Cy5 were conjugated with the same approach except using 25 μl of 4 mg/ml NHS-Cy5. Conjugation BSA-biotin-dye was purified through a desalting column which has been pre-equilibrated with nanopure water.
Synthesis of plasmonic-fluor:
To prepare both types of plasmonic fluor, AuNRs were first coated with a thin organic layer to prevent the fluorescence quench. Briefly, 5 μl of (3-mercaptopropyl)trimethoxysilane (MPTMS) (Sigma-Aldrich, 175617) was added into 5 ml of AuNRs solution (extinction 2), followed by one hour incubation at 24 °C. MPTMS modified AuNRs were collected through centrifugation at 6000 rpm for 10 minutes and redispersed in 1 mM CTAB solution. 2 μl of APTMS (Sigma-Aldrich, 281778) and 2 μl of TMPS (Sigma-Aldrich, 662275) were subsequently added into the MPTMS coated AuNRs solution to form the polymer layer. Finally, AuNR-polymer were collected through three centrifugations at 6000 rpm for 10 minutes and concentrated into a final volume of 10 μl.
Next, to coat BSA-Biotin-dye conjugate around AuNR-polymer, AuNR-polymer was first added into a solution comprised of 1 μl of 20 mg/ml citric acid (Alfa Aesar, 36664) and 100 μl 4 mg/ml BSA-biotin-dye solution followed by 20 minutes sonication in dark. Coated nanostructures were further collected by centrifugation at 4000 rpm for 5 minutes and incubated with 500 μl of 0.4 mg/ml BSA-Biotin-CW800 at pH 10 nanopure water for at least 3 days in 4 °C. The nanostructures were washed by 4-time centrifugation at 6000 rpm for 10 minutes using pH 10 water before use.
Fluorescence enhancement with plasmonic fluor:
The schematic illustration of this assay procedure was demonstrated in Supplementary Figure 1. Specifically, 100 μl of 50 ng/ml BSA-biotin was first incubated within 96 well plate for 10 minutes. The plate was washed by 1X PBS with 0.05% Tween-20 (Sigma Aldrich, P2287) (PBST) and blocked with 3% BSA 1X PBS solution for one hour. Streptavidin-Cy5 or streptavidin-CW800 (1 μg/ml) was subsequently added and incubated for 10 minutes, followed by PBST washing. The plate was further incubated with corresponding plasmonic fluor in 1% BSA solution. After washing, fluorescence signal before and after incubation of plasmonic fluor were recorded using LICOR CLx fluorescence imager for plasmonic fluor 800 with following parameters: channel, laser power: L2, resolution: 21 μm, height: 4 mm; Azure fluorescence scanner and for plasmonic fluor 650: laser power: 0.5, resolution: 21 μm, channel: 650, Focus: 4 mm.
2.2. Conjugation of BSA-FITC:
Fluorescein isothiocyanate isomer I (FITC) was conjugated onto BSA through the reaction between isothiocyanate groups and amine groups. Specifically, 100 μl of 1 mg/ml FITC (Sigma Aldrich, F7250) were added into 2 ml of 2 mg/ml BSA in sodium carbonate buffer (0.1 M, pH ~9) and the mixture was incubated overnight at 4 °C in dark. To stop the reaction, NH4Cl (Sigma Aldrich, 254134) was added to the final concentration of 50 mM. After incubation for 2 hours at 4°C, BSA-FITC was purified using a PD-10 column packed with Sephadex G-25 resin (GE Healthcare Life Sciences, 17–0851-01) pre-equilibrated with 1X PBS. To validate the conjugation and estimate number of dye molecules per protein, absorbance spectra of BSA-FITC conjugation were obtained by UV-Vis spectrometer. Based on the absorbance of FITC at 495 nm and BSA at 280 nm, the ratio of FITC/BSA was calculated to be 0.9.
2.3. Biotinylation of FITC antibody:
The conjugation procedure was conducted according to the instruction of product. Briefly, 6.7 μl of 0.2 mM NHS-PEG4-biotin (Thermo Fisher, A39259) was added into 20 μl of 0.5 mg/ml anti-FITC antibody (Thermo Fisher, 701078) in PBS buffer and incubated for 40 min. The biotinylated anti-FITC antibody was purified through a desalting column (Thermo Fisher, 89882) pre-equilibrated with PBS by centrifugation at 2800 rpm for 120 seconds.
2.4. Conjugation of antibody towards plasmonic-fluor:
The biotinylated antibody was conjugated onto plasmonic fluor through the interactions between biotin and streptavidin. Briefly, plasmonic-fluor was incubated with excessive amount of streptavidin for 30 minutes at room temperature and the conjugated nanostructures were collected and washed by three times centrifugation at 6000 rpm for 10 minutes. Subsequently, highly concentrated plasmonic-fluor was added into biotinylated antibody with four times higher molar ratio. The mixture solution was incubated for one hour before the collection of antibodies coated plasmonic-fluor through centrifugation. (Supplementary Figure S5)
2.5. Competitive ELISA for cortisol and fluorescein:
For cortisol competitive ELISA, 96 well-plate was first incubated with 100 μl BSA-cortisol (Fitzgerald Industries, 80–1060) in PBS at 4 °C overnight. After three times washing with PBST, the plate was blocked with 300 μl 3% BSA 1X PBS for 1 hour at room temperature, followed by washing. Subsequently, 100 μl mixture solution comprised of biotinylated cortisol antibody (Gentex, XM210) and serial diluted standard samples (Sigma Aldrich, C-106) in reagent diluent (1% BSA 1X PBS) was added into different wells and incubated for one hour. The plate was washed and incubated with streptavidin-horseradish peroxidase (HRP) (R&D Systems, 893975) for 20 minutes at room temperature. After three times washing with PBST, 100 μl substrate solution (1:1 mixture of color reagent A (H2O2) and color reagent B (tetramethylbenzidine)) (R&D Systems, DY999) was incubated and the colorimetric reaction was stopped by 50 μl 2N Sulfuric acid (R&D Systems, DY994). Optical density of each well was analysed using microtiter plate reader set at 450 nm. For fluorescein competitive assay, 96 well-plate was first incubated with 100 μl BSA-FITC in PBS for 15 minutes in room temperature. Following assay procedures are similar to cortisol ELISA, except the use of biotinylated anti-FITC antibody.
2.6. Competitive FLISA, p-FLISA and quantum dot-FLISA for cortisol and fluorescein:
Cortisol and fluorescein competitive FLISA was implemented adopting a similar approach as the ELISA, except that streptavidin-HRP was replaced with 20 ng/ml streptavidin-CW800 (LICOR, 926–32230) incubated for 20 minutes. The plate was then washed three times with PBST. In case of p-FLISA, plasmonic-fluor 800 was subsequently incubated for 20 minutes, followed by washing. Plate was imaged by LICOR Odyssey CLx fluorescence imager with the following parameter: laser power: L2, resolution: 21 μm, channel: 800, Focus: 4 mm. In case of quantum dot-FLISA, streptavidin-quantum dot 650 (Thermo Fisher, Q10123MP) was used as nanolabel and incubated for 20 minutes subsequently after biotinylated antibody, followed by washing with PBST. Plate was imaged by Azure fluorescence imager with the following parameter: laser power: 0.5, resolution: 21 μm, channel: 650, Focus: 4 mm.
2.7. Digital competitive p-FLISA for fluorescein:
The digital format of p-FLISA was implemented using the similar procedures as mentioned above. After the incubation of plasmonic-fluor 650, the images of each well were captured by an optical microscope (Nikon Eclipse TsR2). Number of plasmonic-fluors within the same area were counted through a custom-built algorithm for further analysis.
2.8. Spatially multiplexed dot blot p-FLISA for cortisol and fluorescein:
BSA-cortisol and BSA-FITC in 10% glycerol 1X PBS (1 μ/blot) were blotted in duplicates within one well on the microtiter plate and incubated for overnight or 15 minutes, respectively. After three times washing with PBST, the plate was blocked by 3% BSA 1X PBS for one hour at room temperature. The plate was washed and incubated with 100 μl mixture solution comprising two types of biotinylated antibody and two types of serial diluted standard samples in reagent diluent (1% BSA 1X PBS) for one hour. The concentration of cortisol and fluorescein are diluted reversely through each of the sample to avoid cross reactivity. The following procedures were implemented in a similar approach as mentioned above.
2.9. Material characterization:
TEM images were obtained using a JEOL JEM-2100F field emission instrument. To prepare the TEM sample, a drop of aqueous solution was dried on a hydrophilic carbon-coated grid. SEM images were obtained using a FEI Nova 2300 field-emission SEM at an accelerate voltage of 10 KV. The extinction spectra of plasmonic nanoparticles were obtained using Shimadzu UV-1800 spectrophotometer.
3. Results and discussion:
Competitive assay involves the detection and quantification of small molecule analytes by harnessing the competitive binding of the biorecognition element to reference vs. sample analytes. In conventional ELISA, the antibodies (i.e. biorecognition elements) are labelled with an enzyme reporter. In contrast to sandwich immunoassay, in a competitive assay, the amount of colored product generated by the enzymatic reaction is inversely proportional to the concentration of the analyte in the sample. Although this assay is widely employed, this method is not suitable for sensitive and multiplexed detection due to (i) weak signal and low signal-to-noise ratio associated with conventional labels and enzymatic reporters; and (ii) the soluble nature of the colored products precluding the possible multiplexed detection and quantification. To overcome these challenges, we employed a fluorescence-linked immunosorbent assay (FLISA) that relies on plasmonic-fluor as an ultrabright and multi-color fluorescent label, which enables ultrasensitive and multiplexed competitive assay (Figure 1). Plasmonic-fluor is comprised of a plasmonic nanostructure (as fluorescence enhancer), a light emitter (e.g., any given molecular fluorophore), spacer layer (siloxane organic layer), and a universal biological recognition element (e.g., biotin) (Figure 1C). The molecular fluorophores (light emitter) and biotin are covalently conjugated to bovine serum albumin (BSA), which serves as (i) a scaffold for assembling these functional components; (ii) a stabilizing agent, preventing the aggregation of the nanoconstructs; and (iii) anti-fouling agent, minimizing non-specific binding of the plasmonic-fluor to blocked (typically using BSA as a blocking agent) sensor surface, which is extremely important to achieve high specificity and signal-to-background ratio.
Herein, we employed the gold nanorod (AuNR) as the plasmonic nanoantenna owing to the facile tunability of the longitudinal LSPR wavelength as well as the large enhancement of electromagnetic field at the endcaps. Plasmonic-fluor-800 were synthesized by coating BSA-biotin-Cy7.5 conjugates on AuNR with longitudinal LSPR wavelength of 760 nm (Figure 1D). Binding of the plasmonic-flour-800 to streptavidin-CW800 on microtiter plates resulted in nearly 1400-fold enhancement of ensemble fluorescence intensity (Figure 1E and Figure S1).
To investigate the applicability of plasmonic-fluor as an ultrabright nanolabel in competitive assay, we first employ fluorescein as a model analyte. Among various types of competitive assays, a commonly used format involves the immobilization of the reference analytes (BSA-fluorescein) at the bottom of microtiter well, incubation of sample (containing fluorescein), exposure to antibody (biotinylated anti-fluorescein antibody), and labelling the bound antibodies with streptavidin-enzyme or streptavidin-fluorophore conjugates. In contrast to this conventional format, competitive plasmonic-fluor linked immunosorbent assay (p-FLISA) involves the use of plasmonic-fluor as the fluorescent label (Figure 1). To determine the sensitivity and limit of detection (LOD, defined as mean - 3σ of the blank) of competitive FLISA and p-FLISA, serial dilutions of fluorescein with known concentration (10−2 to 104 ng/ml) were used as standards. As expected, fluorescence signal intensity obtained after applying plasmonic-fluor monotonically decreased with an increase in the fluorescein concentration (Figure 2A). The LOD of the fluorescein competitive p-FLISA was found to be 0.34 ng/ml. On the other hand, fluorescence signals obtained with conventional FLISA were too low (close to the background fluorescence intensity) to reveal a dose-dependent curve (Figure 2B). To compare with the performance of other fluorescent nanolabels, we employed commercially available streptavidin-quantum dot conjugates as fluorescence reporters. The fluorescence intensity obtained with quantum dots is at a similar level for all concentrations of the analyte and did not reveal a dose-dependent curve, possibly due to their non-specific binding on the microtiter plate (Figure 2C).
Figure 2.
Fluorescein dose dependent fluorescence intensity and fluorescence mapping at various analyte concentrations by p-FLISA (A), FLISA (B) and quantum dot-FLISA (C). (D) Cortisol dose-dependent fluorescence intensity obtained by 20-minute p-FLISA and the fluorescence intensity maps at various cortisol concentrations. (E) Cortisol dose-dependent optical density obtained by 20-minute conventional ELISA and optical image at various cortisol concentrations. (F) The recovery rate of cortisol spiked in artificial saliva measured by p-FLISA and ELISA accomplished in 20 minutes.
Next, we evaluated the feasibility of plasmonic-fluor enhanced competitive assay for the detection of cortisol, which is a key biomarker in assessing adrenal, pituitary, and hypothalamic function.(de Weerth et al. 2003; Hawley et al. 2016) Following a procedure similar to that described above, the LOD of cortisol competitive p-FLISA was found to be 0.65 pg/ml, far lower than the physiological cortisol level (ng/ml or higher) in various biofluids (Figure S2A).(Arafah et al. 2007; fa*gerlund 1967) Moreover, owing to the superior brightness of plasmonic-fluor, cortisol can be quantified using a 20-minute competitive p-FLISA where the incubation time of each step in the assay is limited to 5 minutes. The LOD of this fast assay was found to be ~2 pg/ml (Figure 2D). In contrast, conventional competitive ELISA completed in 20 minutes exhibited extremely weak colorimetric signal, undetectable with naked eye. The LOD of competitive ELISA was found to be 71.4 pg/ml, which is more than 30-fold higher compared to that of p-FLISA under the identical assay conditions (Figure 2E). To evaluate the performance of this rapid competitive p-FLISA in analyzing complex biological matrices, artificial saliva spiked with different known concentrations of cortisol were employed as samples. Rapid competitive p-FLISA was able to reveal the concentration of all seven cortisol spiked samples and demonstrated more than 90% recovery rate (Figure 2F, Figure S2B). However, the rapid competitive ELISA was able to detect cortisol in only 6 out of the 7 samples and exhibited merely 50% recovery rate (Figure 2F, Figure S2C). These results demonstrated that the competitive p-FLISA could be used for rapid and sensitive detection of small molecules in complex biofluids.
To further improve the sensitivity of competitive p-FLISA, we optimized the assay parameters by reducing the amount of reference analytes on the plate. A recent study has demonstrated that the detection limit of competitive assay can be improved by employing antibodies with higher affinity and reducing the number of reference analytes immobilized on the microtiter plate.(Wang et al. 2018) However, reducing the density of reference analytes inevitably results in reduced binding sites as well as the number of labels on the plate, which in turn compromises the overall signal intensity. Therefore, we hypothesize that ultrabright plasmonic-fluor can be used as a promising signal reporter in competitive assays, with detectable signal even with low density of reference analytes. Here, to understand the effect of the density of the reference analytes on the plate, we have employed FITC as a model analyte and immobilized different concentrations of BSA-FITC at the bottom of the plate. The LOD of p-FLISA with 100 ng/ml and 5 ng/ml BSA-FITC was found to be 0.51 ng/ml and 0.30 ng/ml, respectively, indicating an improved sensitivity with lower amount of reference analytes (Supplementary Figure S3A and S3B). Conversely, competitive ELISA exhibited extremely low colorimetric signal with 5 ng/ml BSA-FITC with large standard deviations (Figure S3C and S3D). The LOD of competitive ELISA with 100 ng/ml BSA-FITC was found to be 3.25 ng/ml, 6-fold higher than that of p-FLISA.
To further improve the sensitivity, we analyzed the signals obtained from competitive p-FLISA in a digital format. Compared to the conventional analog immunoassays, digital assays have been demonstrated to exhibit higher sensitivity and lower LOD.(Cohen et al. 2020; Rissin et al. 2010; Rissin and Walt 2006) Although highly sensitive, existing digital assays require partitioning the samples into femtoliter volume before analysis, which inevitably increases the complexity and cost of the assay. Owing to their high brightness, individual plasmonic-fluors can be visualized and counted in an image acquired using standard epi-fluorescence microscope. Therefore, this unique feature of plasmonic-fluors enables digital analysis in a conventional immunoassay procedure without partition to further improve the sensitivity. As expected, the fluorescence images of the microtiter plates revealed individual nanostructures uniformly distributed on the surface in a dose-dependent manner (Figure S4A). A standard curve was obtained by counting (using a custom-designed image processing algorithm) the number of plasmonic-fluors in six images collected from each well corresponding to known concentration of analyte. The six images from each well represent nearly 50% of the total area of the microtiter well bottom (Detailed calculation can be found in supporting information). The LOD of p-FLISA with 5 ng/ml BSA-FITC as reference analyte was calculated to be 29 pg/ml, which is more than 10-fold lower than that of p-FLISA in analog format (Figure S4B). These results emphasize the importance of plasmonic-fluor as an ultrabright label in the improvement of competitive assay sensitivity. Improvements in the monodispersity of plasmonic-fluors and fluorescence signal hom*ogeneity from individual plasmonic-fluors will further enhance the performance of the digital assay.
Multiplex immunoassays, which enable the simultaneous detection of multiple analytes, significantly lower the amount of sample and time required for comprehensive analysis.(Dincer et al. 2017; Tabakman et al. 2011) In case of metabolites, detecting multiple analytes simultaneously from a small sample volume can improve specificity of diagnosis and, in some cases, ratio of two analytes might be more informative than single analyte.(Wang and Lei 2018) Here, to test the applicability of plasmonic-fluor as a nanolabel in a spatially-multiplexed competitive assay, we employ fluorescein and cortisol as two model analytes. Conventional singleplex assay involves the immobilization of reference analytes at the bottom of the well and incubation of a mixture of sample containing the analyte and specific antibody. In contrast, in spatially multiplexed assay, solutions comprised of different reference analytes were blotted in different locations of the same well, followed by incubation with a mixture of sample containing multiple analytes and multiple antibodies corresponding to the analytes (Figure 3A). To compare the LOD of competitive p-FLISA and FLISA in the multiplexed format, solutions comprised of serially diluted fluorescein and cortisol with known concentrations were used as standards. Specifically, two analytes were mixed in an inverse manner of their dilution sequence. Fluorescence signals obtained from conventional fluorophores were too weak (close to the background noise) and did not exhibit a discernable dose-dependence for either of the analytes (Figure 3B). Conversely, fluorescence signals obtained after applying plasmonic-fluor were inversely proportional to the concentration of both analytes and revealed the low crosstalk between the blots, indicating low cross-reactivity in the multiplexed assay. The LOD of fluorescein and cortisol was found to be 14 ng/ml and 21 pg/ml, respectively (Figure 3C).
Figure 3.
(A) Schematic illustration depicting the process of cortisol (green) and fluorescein (yellow) competitive p-FLISA in a spatial multiplex manner. Cortisol (top) and fluorescein (bottom) dose dependent fluorescence intensity and fluorescence mapping by multiplexed competitive FLISA (B) and p-FLISA (C).
The sensitivity of existing easy-to-use methods such as enzyme-linked immunosorbent assays (ELISA) and lateral flow assays (LFA) is not sufficient to detect low abundant small molecules. On the other hand, ultrasensitive detection techniques such as mass spectrometry are complex, expensive and involve bulky instruments that preclude their routine and in-field use. By harnessing antibody printing capabilities, improved monodispersity of plasmonic fluors and advances in miniaturized read-out devices(Xu 2021), the plasmonic fluor enabled competitive assay with its spatial-multiplexing capability can enable the simultaneous detection and quantification of a large number of analytes in various biofluids.
4. Conclusions
In summary, harnessing plasmonic-fluor as an ultrabright fluorescence nanolabel, we have demonstrated a multiplexed and ultrasensitive competitive immunoassay for the detection and quantification of small molecules. Under optimized assay conditions, the LOD of competitive p-FLISA is nearly 30-fold lower compared to ELISA. In comparison to plasmonic-fluors, conventional fluorophores and quantum dots exhibited fluorescence signal comparable to that of the background. The competitive p-FLISA can be completed within 20 minutes with higher than 90% recovery rate of cortisol spiked in saliva. More importantly, competitive p-FLISA is amenable to multiplexed detection of analytes by spatially confining the intense fluorescence signal from the plasmonic-fluors to antibody spots deposited at the bottom of microtiter well. This simple, fast, and highly sensitive approach will be attractive for the multiplexed detection of other small molecules in broad field, including biomedical research, clinical diagnosis, environmental surveillance and food inspection. In the future, this proof-of-concept demonstration will need to be rigorously validated with clinical samples. Integration of the competitive assay with simple and field-deployable fluorescence read-out devices will make it attractive for point-of-care and in-field applications.
Supplementary Material
Supporting Information
Click here to view.(1.7M, docx)
Acknowledgements
We acknowledge support from National Institutes of Health (R21EB030171). The authors also thank Nano Research and Environmental Facility (NREF) and Institute of Materials Science and Engineering (IMSE) at Washington University for providing access to electron microscopy facilities and Prof. Nathaniel Huebsch for providing access to the fluorescence microscope.
Competing interests
The authors declare the following competing financial interest(s): J.L., J.J.M., and S.S. are inventors on provisional patent related to this technology and the technology has been licensed by the Office of Technology Management at Washington University in St. Louis to Auragent Bioscience LLC, which is developing plasmonic-fluor products. J.J.M., and S.S. are co-founders/shareholders of Auragent Bioscience LLC. These potential conflicts of interest have been disclosed and are being managed by Washington University in St. Louis.
Reference:
- Anfossi L, Di Nardo F, Giovannoli C, Passini C, Baggiani C, 2015. Food Control57, 115–121. [Google Scholar]
- Arafah BM, Nishiyama FJ, Tlaygeh H, Hejal R, 2007. The Journal of Clinical Endocrinology & Metabolism92(8), 2965–2971. [PubMed] [Google Scholar]
- Chen Z, Tabakman SM, Goodwin AP, Kattah MG, Daranciang D, Wang X, Zhang G, Li X, Liu Z, Utz PJ, 2008. Nature biotechnology26(11), 1285–1292. [PubMed] [Google Scholar]
- Cheng S, Shi F, Jiang X, Wang L, Chen W, Zhu C, 2012. Analytical chemistry84(5), 2129–2132. [PubMed] [Google Scholar]
- Cohen L, Cui N, Cai Y, Garden PM, Li X, Weitz DA, Walt DR, 2020. ACS nano14(8), 9491–9501. [PubMed] [Google Scholar]
- de Lima MM, Vieira AC, Martins I, Boralli VB, Borges KB, Figueiredo EC, 2016. Food chemistry197, 7–13. [PubMed] [Google Scholar]
- de Weerth C, Graat G, Buitelaar JK, Thijssen JH, 2003. Clinical chemistry49(4), 658–660. [PubMed] [Google Scholar]
- Dietrich JA, McKee AE, Keasling JD, 2010. Annual review of biochemistry79, 563–590. [PubMed] [Google Scholar]
- Dincer C, Bruch R, Kling A, Dittrich PS, Urban GA, 2017. Trends in biotechnology35(8), 728–742. [PMC free article] [PubMed] [Google Scholar]
- Du P, Jin M, Chen G, Zhang C, Jiang Z, Zhang Y, Zou P, She Y, Jin F, Shao H, 2016. Scientific reports6(1), 1–8. [PMC free article] [PubMed] [Google Scholar]
- Espina V, Woodhouse EC, Wulfkuhle J, Asmussen HD, Petricoin III EF, Liotta LA, 2004. Journal of immunological methods290(1–2), 121–133. [PubMed] [Google Scholar]
- fa*gerlund U, 1967. General and comparative endocrinology8(2), 197–207. [PubMed] [Google Scholar]
- Fu X, Chen L, Choo J, 2017. Analytical chemistry89(1), 124–137. [PubMed] [Google Scholar]
- Glass TR, Ohmura N, Morita K, Sasaki K, Saiki H, Takagi Y, Kataoka C, Ando A, 2006. Analytical chemistry78(20), 7240–7247. [PubMed] [Google Scholar]
- Hawley JM, Owen LJ, Lockhart SJ, Monaghan PJ, Armston A, Chadwick CA, Wilshaw H, Freire M, Perry L, Keevil BG, 2016. Clinical chemistry62(9), 1220–1229. [PubMed] [Google Scholar]
- Lee J, Kim L, Shin Y, Lee J, Lee J, Kim E, Moon J-K, Kim J-H, 2017. Journal of agricultural and food chemistry65(16), 3387–3395. [PubMed] [Google Scholar]
- Lee K-S, El-Sayed MA, 2005. The Journal of Physical Chemistry B109(43), 20331–20338. [PubMed] [Google Scholar]
- Li Y-F, Sun Y-M, Beier RC, Lei H-T, Gee S, Hammock BD, Wang H, Wang Z, Sun X, Shen Y-D, 2017. TrAC Trends in Analytical Chemistry88, 25–40. [Google Scholar]
- Li Y, Yu F, Chen S, Zhang N, Wang H, Zhao Q, 2021. Analytical Chemistry93(11), 4794–4799. [PubMed] [Google Scholar]
- Lin S, Gan N, Qiao L, Zhang J, Cao Y, Chen Y, 2015. Talanta144, 1139–1145. [PubMed] [Google Scholar]
- Liu X, Xu Y, Xiong Y. h., Tu Z, Li Y. p., He Z. y., Qiu Y. l., Fu J. h., Gee SJ, Hammock BD, 2014. Analytical chemistry86(15), 7471–7477. [PMC free article] [PubMed] [Google Scholar]
- Luan J, Morrissey JJ, Wang Z, Derami HG, Liu K-K, Cao S, Jiang Q, Wang C, Kharasch ED, Naik RR, 2018. Light: Science & Applications7(1), 1–13. [PMC free article] [PubMed] [Google Scholar]
- Luan J, Seth A, Gupta R, Wang Z, Rathi P, Cao S, Derami HG, Tang R, Xu B, Achilefu S, 2020a. Nature biomedical engineering4(5), 518–530. [PMC free article] [PubMed] [Google Scholar]
- Luan J, Seth A, Gupta R, Wang Z, Rathi P, Cao S, Gholami Derami H, Tang R, Xu B, Achilefu S, Morrissey JJ, Singamaneni S, 2020b. Nat Biomed Eng4(5), 518–530. [PMC free article] [PubMed] [Google Scholar]
- Peteu SF, Boukherroub R, Szunerits S, 2014. Biosensors and Bioelectronics58, 359–373. [PubMed] [Google Scholar]
- Rissin DM, Kan CW, Campbell TG, Howes SC, Fournier DR, Song L, Piech T, Patel PP, Chang L, Rivnak AJ, 2010. Nature biotechnology28(6), 595–599. [PMC free article] [PubMed] [Google Scholar]
- Rissin DM, Walt DR, 2006. Journal of the American Chemical Society128(19), 6286–6287. [PubMed] [Google Scholar]
- Tabakman SM, Lau L, Robinson JT, Price J, Sherlock SP, Wang H, Zhang B, Chen Z, Tangsombatvisit S, Jarrell JA, 2011. Nature communications2(1), 1–9. [PMC free article] [PubMed] [Google Scholar]
- Wang X, Cohen L, Wang J, Walt DR, 2018. Journal of the American Chemical Society140(51), 18132–18139. [PubMed] [Google Scholar]
- Wang X, Wang S, Cai Z, 2013. TrAC Trends in Analytical Chemistry52, 170–185. [Google Scholar]
- Wang Y-P, Lei Q-Y, 2018. Signal transduction and targeted therapy3(1), 1–9. [PMC free article] [PubMed] [Google Scholar]
- Watanabe E, Miyake S, Yogo Y, 2013. Journal of agricultural and food chemistry61(51), 12459–12472. [PubMed] [Google Scholar]
- Wu K, Zhang H, Chen Y, Luo Q, Xu K, 2021. IEEE Electron Device Letters42(4), 541–544. [Google Scholar]
- Xu K, 2021. Journal of Micromechanics and Microengineering31(5), 054001. [PMC free article] [PubMed] [Google Scholar]
- Zhang Y, Noji H, 2017. Analytical chemistry89(1), 92–101. [PubMed] [Google Scholar]
