Analyte-Induced Disruption of Luminescence Quenching (AIDLuQ) for Femtomolar Detection of Biomarkers

Abstract

Here, we present a graphene-based analyte induced disruption of luminescence quenching (AIDLuQ) assay for specific detection of biomarkers with femtomolar sensitivity. In the AIDLuQ assay, antibody (Ab)-conjugated quantum dots (QDs) are initially deposited on a graphene coated paper. However, the emission from QDs is quenched due to resonance energy transfer to graphene. Upon the addition of an analyte (An) corresponding to Ab, QDs-Ab-An complex is lifted above the surface resulting in the disruption of the quenching from graphene and recovery of the luminescence of the QDs. The percentage of recovery depends upon the concentration of analyte allowing one to create standard curves for effective quantification. Despite its rapidity in assay time (15-20 mins), the graphene platform has limited sensitivities. To further enhance this sensitivity, we embedded gold nanoparticle (Au NPs) into graphene paper. The graphene-Au paper exhibited excellent sensitivity in our model assay and was able to detect ~10 fM of biotin and IgG unlike graphene that showed only ~1 nM and ~10 pM sensitivities.

Introduction:

Graphene, a sheet of sp² hybridized carbon atoms, has received much attention due to its unique physicochemical properties. It is ideally suited for the realization of biosensors in various transduction modes ranging from electrical/electrochemical transduction to optical detection [1–4]. Specifically, graphene presents an excellent platform for fluorescence energy transfer or quenching [5,6] due to its linear electronic energy bands with high optical absorption (~2.3% per layer [7–11]). Others and we have previously shown that emission from traditional dyes and quantum dots [6] can be quenched by resonant energy transfer via the excitation of electron-hole pairs in graphene. This ability of graphene has also been utilized in supressing background fluorescence and enhancing Raman signals in graphene-enhanced Raman spectroscopy or GERS [12]. Building on fluorescence quenching properties of graphene, this article presents a novel flexible immunosensor for highly sensitive and rapid detection of proteins and biomarkers.

Traditionally, antibody-based immunoassays have been extensively used for the detection of proteins due to their high sensitivity and specificity. Despite these advantages, conventional immunoassays (e.g., lateral flow, microarray, enzyme-linked immunosorbent assay) [13] require long incubation times, multiple washing steps, and are unsuitable for point-of-care (POC) testing [14]. Over the past two decades, there has been a steady increase in the availability and use of flexible and inexpensive POC sensors for disease diagnosis and monitoring [15–17]. Previously, we used surface plasmon (SP) coupled emission [18] to develop highly sensitive fluorescent POC platforms for pathogen detection. Although such SP platforms improve the sensitivity and flexibility needed for POC testing, they do not alleviate long incubation and multiple washing steps. In this regard, there is a great need for rapid, highly sensitive (up to femto-molar concentrations), flexible, and inexpensive optical POC sensors for detection of specific biomarkers. Here, we used the fluorescence quenching properties of graphene to build a rapid, flexible, cost-effective, extremely sensitive and highly efficient immunosensing platform.

In many conventional immunoassays, at least two antibodies are necessary for detecting an analyte. While the primary antibody captures the analyte, a labelled secondary antibody (often labelled with a dye or an optically active molecule) is used to obtain a measurable optical signal to quantify the analyte [19]. Some immunoassays (such as the capture sandwich assay) necessitate the use of three antibodies. Unlike these methods where the detection antibody is added at the end, our graphene sensing platform is precoated with detection antibodies (see Figure 1) labelled highly luminescent quantum dots (QDs). However, the luminescence of QDs on the platform is almost completely quenched due to the presence of graphene. Upon the addition of analyte, the analyte binds with the detection antibody and raises QDs away from graphene leading to reappearance of QD emission (Figure 1a). Thus, in a single step, the analyte can be detected. While this simple platform allows for rapid detection of analytes, its sensitivity is limited by the emission intensity and quantum yield of QDs. To improve the sensitivity, we incorporated gold nanoparticles (Au NPs) into graphene platform (Figure 1b). In the graphene-Au platform, the analyte raises QDs away from the surface similar to graphene platform leading to recovery in QD emission. The analyte-antibody pair acts as a spacer between Au NPs embedded in graphene and QDs, which leads to surface plasmon-induced increase in emission and consequently higher sensitivity.

Figure 1:

a) The sensing platform consists of receptor (e.g. streptavidin or IgG antibody) functionalized CdSe quantum dots (QDs) deposited on a flexible graphene paper. The fluorescence emission from CdSe QDs is quenched when they are spatially close to graphene. Upon the addition of analyte (e.g. biotin or IgG), some CdSe QDs are lifted off from the graphene surface due to binding with receptors on the surface of CdSe QDs. The binding increases the spatial gap between CdSe QDs and graphene, which disrupts quenching and results in an increased emission from QDs. The number of CdSe QDs lifted off from graphene surface is proportional to the analyte concentration. Thus, the increase in the emission of CdSe QDs could be used as a tool to detect different analytes, b) A platform similar to the one presented in (a) with the extra addition of gold nanoparticle (Au NPs) is shown. In this case, the binding of analyte disrupts quenching caused by graphene while plasmonic Au NPs enhance the emission of CdSe QDs to enable highly sensitive detection of low analyte concentrations.

In other words, as described in Figure 1, this process relies on analyte induced disruption of luminesce quenching (AIDLuQ). The AIDLuQ platform has multiple advantages such as: i) easy scalability in manufacturing, ii) flexibility, iii) inexpensive, iv) alleviation of the use of a second antibody as is the case in most of traditional antigen-antibody immunosensing techniques, and v) shorter incubation time (15-30 min). We demonstrated two model assays using biotin-streptavidin and human immunoglobulin assay to illustrate the practical applications of AIDLuQ sensors. Our results show clear evidence that AIDLuQ sensors are successfully able to achieve selective and specific detection up to 10 fM on graphene-Au platform, which is superior to other sensing platforms (Table S1).

Experimental Section:

Synthesis of Graphene/Graphene-Au NP Paper:

Exfoliated graphene nanoplatelets (Grade M) were obtained from XG Science (Mason, MI). Detailed atomic force microscopy images of graphene platelets are provided in the supporting information (Figure S1). A 5 mg/ml suspension of exfoliated graphene nanoplatelets was prepared in 75 ml acetone via tip sonication for 30 minutes (Branson 250W, 1/8” tip sonicator). This solution was then spray-coated thrice using an industrial spray gun (Iwata 5095 WS400; 1.3 mm nozzle, 29 psi ambient air pressure) to coat a layer of graphene on a standard copier paper (21 x 29.7 cm). A representative scanning electron micrograph (obtained using Hitachi S-4800) of graphene coated paper is shown in Fig. 2a. For preparing graphene-Au NP platform, 7.5 ml of 1.5 mg/ml ~10 nm Au NPs (Vive Nano, Inc.) was added to the 5 mg/ml graphene suspension before the third coat. This corresponds to a net areal Au NP density of ~18 μg/cm², which was found to be optimal for sensing (see Figure S2 in supporting information for further details). The paper was dried for 15 min in air after each coating. The dried graphene/graphene-Au NP papers were stored in dark and dry conditions.

Figure 2:

a) A representative scanning electron micrograph of graphene coated paper. The graphene flakes are ~5 nm thick with lateral dimensions in the range of few μm. The scale bar is 20 μm, b) The emission of CdSe QDs on graphene is strongly quenched due to interactions between pi-electron clouds of graphene and d-orbitals of CdSe QDs. The inset shows a very weak quenched emission peak for CdSe QDs on graphene.

Preparation of Standard Buffer:

A standard buffer was prepared by mixing 0.5% (v/v) Tween-20 and 1% (w/v) of BSA in 0.01 M phosphate buffer saline (PBS). This standard buffer was used as a solvent for all the further studies.

Conjugation of CdSe QDs with Antibodies:

CdSe QDs were purchased from Thermo Fisher Scientific (Invitrogen, Cat. # Q10123MP). QDs were mixed with anti-human IgG antibody (Ab) (Abcam, Cat. # ab109489 ) in the standard buffer to attain a final concentration of 100 nM for QDs and 200 μg ml⁻¹ for Ab. These optimal concentrations for QDs and Ab was chosen based on previous studies [20]. The conjugation was carried out by continuous shaking at 650 rpm and 4°C for 30 min.

Preparation of Graphene/Graphene-Au Sensing Platforms:

For sensing experiments, graphene or graphene-Au NP coated paper was cut ~1 x 1cm pieces. 2 μl of the QD-Ab conjugate was drop cast on to each piece and left to dry at room temperature for 1 hour. Detailed electron microscopy images of graphene/graphene-Au NP sensing platforms are provided in the supporting information (see Figs. S3–S7). Upon drying, the fluorescence intensity of QDs was measured for at least three spots on each piece with the excitation of 532 nm using Renishaw InVia micro-spectrometer. The average (n=3) spectrum was considered as the background fluorescence.

Using the Sensing Platform for IgG Detection:

Human IgG (Abcam, Cat. # ab91102) solution were prepared in different concentrations (10 fM – 1 nM) in the standard buffer and stored at 4°C. 20 μl of the analyte solutions were then added to the spots where the QD-Ab was drop cast. The solution was immediately absorbed by the paper and took around 15 min of incubation at room temperature for drying. Once dried, the fluorescence was recorded from at least three spots for each concentration with a 532 nm excitation using Renishaw InVia micro-spectrometer. For evaluating the specificity of our sensors (discussed later in Fig. 5), human IgG solution was prepared in 10% fetal bovine serum (FBS). The data shown is the average fluorescence collected from multiple spots to avoid artefacts arising from inhomogeneous spatial distribution of the antibody/antigen.

Figure 5:

Graphene-Au paper with IgG antibody-coated CdSe QDs is used for detecting IgG in the presence of fetal bovine serum (FBS) to evaluate the specificity. Clearly, even in the presence of FBS, the addition of IgG disrupts the quenching between graphene and CdSe QDs and thereby increases the photoluminescence intensity.

Results and Discussion:

As shown in Fig.1a, AIDLuQ works in the following manner on graphene paper. Initially, antibody-coated CdSe QDs (QD-Ab) are adsorbed on the surface of graphene paper. The interaction between CdSe QDs and graphene results in the quenching of emission from CdSe QDs (see Fig. 2b). Upon the addition of antigen, some CdSe QDs are raised from the surface, which disrupts quenching from graphene and leads to increased emission of CdSe QDs. In the case of graphene-Au paper (Fig. 1b), the emission from CdSe QDs is further enhanced due to the presence of surface plasmons of Au NPs.

To test the quenching efficiency of graphene, we dropcasted ~2 μl of 100 nM QD-Abs on both a plain paper and a graphene-coated paper. As seen in Fig. 2b, there is a significant decrease in the fluorescence intensity in the case of the graphene-coated paper due to fluorescence quenching. This observation concurs with previous studies on highly efficient fluorescence quenching of CdSe/ZnS QDs on few-layer graphene [6]. To validate our hypothesis of AIDLuQ, we recorded emission from streptavidin-coated CdSe QDs on graphene paper in the presence of different biotin concentrations. The complex of biotin and streptavidin is the strongest known non-covalent interaction with an extremely low dissociation constant of 10⁻¹⁵ M [21]. We performed a concentration study for the increasing concentrations of biotin and recorded the corresponding fluorescence intensities. As evident from Fig. 3a, there is a steady increase in the recorded fluorescence intensity with the increase in the concentration of biotin added to the graphene paper. While we were able to observe significant recovery in emission of CdSe QDs at >1 nM biotin, no discernible changes in the emission of CdSe QDs were observable at lower concentrations.

Figure 3:

a) Graphene paper with streptavidin-coated CdSe QDs is used for detecting biotin. Upon the addition of biotin, CdSe QDs are lifted off from the surface leading to increase in emission. In this case, a concentration as low as ~1 nM was able to disrupt the quenching. No discernable changes were observed for biotin concentrations below 1 nM, b) A similar experiment of graphene-Au paper with streptavidin-coated QDs demonstrated increased sensitivity due to the presence of Au NPs, which enhance emission by increasing the local electric field, c) A plot showing the background corrected intensity of CdSe QDs emission as a function of biotin concentration. Clearly, graphene-Au provides better sensitivity compared to graphene paper alone.

As mentioned earlier, the graphene paper provides a rapid sensing platform but lacks sensitivity to detect lower concentrations. To achieve better sensitivity, we embedded Au NPs into the graphene platform in order to increase local electric field around CdSe QDs through surface plasmons. The graphene-Au paper exhibited much higher sensitivity even at lower concentrations (~10 fM) of biotin (see Fig. 3b). As shown in Fig. 3c, the graphene-Au paper is effective in sensing a wider concentration range from 10 fM-1μM due to the increase in the emission intensity of CdSe QDs. This increase in sensitivity can be rationalized as follows. The quantum yield (QY) of CdSe QDs in the presence of graphene can be expressed as [21]

$$QY=\frac{\Gamma }{\Gamma +{\Gamma }_{Qu}+k_{nr}},$$

where, Γ is radiative decay rate, Γ_Qu is non-radiative decay rate arising from the quencher or graphene, k_nr is the inherent non-radiative decay rate of CdSe QDs. In the case of graphene paper, analyte binds the detection antibody and raises CdSe QDs from the surface resulting in a situation where the effects of Γ_Qu can be ignored. It is worth noting that the quenching in graphene occurs through resonant energy transfer, which decreases very rapidly with distance d as d⁻⁶ [6]. Thus, the original QY and the emission of CdSe QDs is recovered. In the presence of Au NPs, the QY of CdSe QDs is altered and may be expressed as

$$QY=\frac{\Gamma +{\Gamma }_m}{\Gamma +{\Gamma }_m+{\Gamma }_{Qu}+k_{nr}},$$

where, Γ_m is the change in the radiative decay rate due to the presence of metallic NPs. When analyte binds the detection antibody on graphene-Au paper, the quenching effects are removed and the radiative decay rate is increased leading to an increase in QY. While the presence of metallic Au NPs has an oscillatory effect on radiative decay rates, it results in a positive Γ_m when the fluorophore is 5-20 nm away from it [22,23,24,25]. Given that ~20 nm is the typical dimensions for antigen-antibody pairs that are sensed, the presence of Au NP increases the emission of CdSe QDs even at low concentrations of biotin (Fig. 2b).

To validate the sensitivity of AIDLuQ platform for specific biomarkers, we evaluated the performance of graphene and graphene-Au papers for detecting human IgG. IgG is an important antibody, which is found in all body fluids and protects against bacterial and viral infections. IgG levels are often measured to diagnose immunodeficiencies, infections, and detect auto-immune diseases.

As shown in Fig. 4, the graphene/graphene-Au platforms performed similar to the case of biotin. While graphene platform was able to sense up to ~10 pM, graphene-Au platform could distinguish up to ~10 fM of IgG (see Figs. 4c and 5). To further test the specificity of our assay, we evaluated the performance of graphene-Au NP platform for sensing IgG in the presence of fetal bovine serum (FBS), which is a complex mixture of several gamma globulins (i.e., antibodies similar to IgG), proteins, amino acids, sugars, lipids, and hormones. Despite the presence of several interfering molecules, we observed that the PL intensity increased with increasing IgG concentration similar to Fig. 4c (Fig. 5). A clear difference could be observed even at IgG concentrations as low as 10 fM.

Figure 4:

Graphene (a) and graphene-Au papers (b) with IgG antibody-coated CdSe QDs are used for detecting IgG. Similar to the case of biotin, an increase in the emission of CdSe QDs was observed for both graphene and graphene-Au papers, c) A plot showing the background corrected intensity of CdSe QDs emission as a function of IgG concentration reveals that it is possible to sense up to ~10 fM of IgG using the graphene-Au platform.

Conclusions:

In this study, we fabricated a facile single-step immunoassay sensing platform based on graphene’s fluorescence quenching ability. Antibody-conjugated CdSe QDs deposited on graphene paper showed low emission due to resonance energy transfer. We hypothesized that the addition of appropriate analyte will bind antibody, raise CdSe QDs from the surface, and consequently help recover its emission. We validated this “analyte-induced disruption in luminescence quenching or AIDLuQ” scheme using two model assays: biotin and IgG. On graphene coated with CdSe-Streptavidin, we were able to discernibly sense up to 1 nM biotin. Similarly, using graphene coated with CdSe-anti-IgG antibodies, we were able to detect up to ~10 pM IgG. To increase the sensitivity of this platform, we embedded Au NPs (~10 nm) into the graphene paper. The graphene-Au paper outperformed the graphene paper and showed ~10 fM sensitivity for biotin and IgG. In summary, spray coated graphene/graphene-Au provides a rapid, specific, and highly sensitive platform for biosensing.