Detection of SARS-CoV-2 receptor binding domain using fluorescence probe and DNA flowers enabled by rolling circle amplification

Using rolling circle amplification (RCA) and two different ways of signal readout, we developed analytical methods to detect the receptor-binding domain (RBD) of SARS-CoV-2 spike protein (S protein). We modified streptavidin-coated magnetic beads with an aptamer of RBD through a biotin-tagged complementary DNA strand (biotin-cDNA). Binding of RBD caused the aptamer to dissociate from the biotin-cDNA, making the cDNA available to initiate RCA on the magnetic beads. Detection of RBD was achieved using a dual signal output. For fluorescence signaling, the RCA products were mixed with a dsDNA probe labeled with fluorophore and quencher. Hybridization of the RCA products caused the dsDNA to separate and to emit fluorescence (λex = 488 nm, λem = 520 nm). To generate easily detectable UV–vis absorbance signal, the RCA amplification was extended to produce DNA flower to encapsulate horseradish peroxidase (HRP). The HRP-encapsulated DNA flower catalyzed a colorimetric reaction between H2O2 and 3,3′,5,5′-tetramethylbenzidine (TMB) to generate an optical signal (λabs = 450 nm). The fluorescence and colorimetric assays for RBD have low detection limits (0.11 pg mL−1 and 0.904 pg mL−1) and a wide linear range (0.001–100 ng mL−1). For detection of RBD in human saliva, the recovery was 93.0–100% for the fluorescence assay and 87.2–107% for the colorimetric assay. By combining fluorescence and colorimetric detection with RCA, detection of the target RBD in human saliva was achieved with high sensitivity and selectivity. Graphical Abstract Supplementary Information The online version contains supplementary material available at 10.1007/s00604-023-05747-6.


Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is an enveloped virus with a surface heavily decorated with stick-like glycosylated spike (S) structural proteins, as well as three other major structural proteins: nucleocapsid (N), matrix (M), and envelope (E) proteins [1][2][3]. The structure of the S protein consists of three protomers, each containing S1 and S2 subunits, consisting of 672 and 588 amino acids, respectively [4,5]. The S1 subunit is particularly important because it contains the receptor-binding domain (RBD) that plays a crucial role in SARS-CoV-2 virus infection. It is now well known that SARS-CoV-2 infects epithelial cells of the human respiratory system. The interaction between the RBD of the S protein and angiotensin-converting enzyme II (ACE2), which is expressed on the host cells, and the S2 subdomain are required for membrane fusion [6,7]. In addition, the RBD determines the binding affinity to the host receptor, viral specificity, and mortality [8,9]. For these reasons, the RBD of S protein provides a suitable target to develop biosensors for detection of SARS-CoV-2 [10].
There are many different tools to detect SARS-CoV-2 [11]. Some research groups have reported biosensors and molecular biology methods for SARS-CoV-2 detection by targeting RNA [12,13], antigen and antibodies [14][15][16], and the spike and nucleocapsid proteins of the virus [17][18][19]. However, these assays have their own advantages and limitations. For example, RNA-based assays are rapid but can be affected by sample degradation, while antibody-based assays are highly specific but can be expensive. Nanomaterial-based biosensors have good specificity and can offer high-throughput, but reproducibility issues and potential toxicity exist. To overcome these limitations, we choose to develop aptamer-based biosensors for detection of SARS-CoV-2 RBD protein.
Aptamers have become a popular choice for constructing biosensors because of their high specificity and affinity for binding analytical targets. They offer advantages over antibodies [20], including unique 3D structure and variable conformation, stability across a wider temperature and pH range, and lower production costs [21][22][23]. Motivated by reported aptamers targeting the RBD of S protein [24], we decided to use aptamer as a recognition element to develop protein assays using isothermal amplification to achieve high sensitivity. Several isothermal amplification methods, for example, hybrid chain reaction (HCR) [25,26] and rolling circle amplification (RCA) [27][28][29][30], are suitable for in situ signal amplification on the surface of solid materials. As DNA-encoded information is stable and can be converted easily into optical or electrochemical signals, DNA-based probes are routinely used in bioanalytical assays. The nucleic acid sequences can also be designed to form various interesting structures for signal output [31,32].
The aim of this work is to develop analytical method for detection of SARS-CoV-2 RBD by fluorometric and colorimetric measurements. The optical assays are based on displacement of aptamer from its cDNA by the target RBD. A rolling circle amplification on magnetic beads is used to enhance the target-induced signal output. Two alternative means are considered for signal readout. For fluorometric reading, the RCA product is used to trigger fluorescence emission after hybridization with a dsDNA labeled with fluorophore and quencher. For colorimetric reading, the RCA is extended to produce DNA flower [33] to encapsulate horseradish peroxidase (HRP) [34,35], which catalyzes the reaction between H 2 O 2 and 3,3′,5,5′-tetramethylbenzidine (TMB) to alter the UV-vis absorbance. Using SARS-CoV-2 RBD as a model analytical target, we wish to highlight the potential of aptamers and RCA in biosensing applications.
Preparation of cDNA/aptamer-functionalized magnetic beads (MBs-cDNA/apt) and synthesis of circular DNA template are described in the Electronic Supplementary Material (ESM).

Competition experiment
S protein RBD solution was diluted to 0, 0.001, 0.01, 0.1, 1, 10, and 100 ng/mL in 1 × PBS buffer. The prepared solution (100 μL) was mixed with 50 μL of MBs-cDNA/apt suspension. The mixture was shaken at room temperature for 1 h. During this period, S protein RBD bound to the aptamer to cause it to dissociate from the cDNA immobilized on the MBs. After magnetic separation, the MBs were collected and washed 3 times with 200 μL 1 × PBS buffer. Finally, the MBs were collected and re-suspended in 30 μL 1 × PBS buffer to give MBs-cDNA to be used in RCA.

Rolling circle amplification
Circular DNA template (2 μL) and MBs-cDNA (4 μL) were mixed with 1 μL recombinant albumin, 1-μL phi29 DNA polymerase (10 U/μL), 4-μL dNTPs (10 mM), 3 μL 10 × phi29 DNA polymerase buffer, and 15 μL H 2 O. The mixture was gently stirred and incubated at 30 °C. In the fluorometric assay, the RCA continued for 2 h before the mixture was heated at 65 °C for 10 min to inactivate the enzyme. In the colorimetric assay, 5-μL HRP solution (1 mg/ mL) was added before the RCA was started and continued for 20 h. The last step of enzyme inactivation was omitted in the colorimetric assay.

Fluorometric detection
Two DNA sequences were designed to construct the probe: one was modified with BHQ-1 at the 3′ terminal and the other one with FAM at the 5′ terminal. The two DNA solutions (10-μM stock solution) were mixed at room temperature for 1 h to prepare the F-Q probe. For detection of RCA products, 10 μL of the F-Q probe was added into 30 μL of amplification products and 60 μL H 2 O. The mixture was kept at room temperature for 45 min. For fluorescence measurement, the excitation wavelength was set to 488 nm, and the emission wavelength of the fluorescence peak was 520 nm.

Colorimetric detection
The RCA was extended to 20 h to synthesize the HRP-loaded DNA flowers. The DNA flowers were collected by magnetic separation, then washed with 100-μL water to remove excess HRP. Finally, the DNA flower was re-suspended in 30-μL water. For colorimetric detection, the DNA flower (10 μL) was mixed with 80-μL acetate buffer (pH 5, 0.1 M), then added to 5 μL 35% H 2 O 2 (w/w aqueous solution) and 5-μL TMB (0.025 mM). The mixture was put on a rocking table at room temperature for 10 min. After adding 50 μL of stop solution (0.2 M H 2 SO 4 ) into the mixture, the particles were removed by magnetic separation. The absorbance of the supernatant at 450 nm was measured using a UV-vis spectrophotometer. All data were obtained from triplicate experiments.

Characterization of DNA flowers
DNA flowers were characterized by scanning electron microscopy (SEM). Prior to SEM imaging, the samples (10 μL) were dried in an oven and coated with palladium/gold to increase conductivity. The DNA flowers were characterized using excitation voltage of 3 ~ 10 kV.

Detection of S protein RBD in saliva
Human saliva samples (100 μL) were analyzed directly using the fluorometric and colorimetric assays without pre-treatment. Saliva (100 μL) was mixed with 100 μL of S protein RBD solution (1-100 ng/mL). The spiked samples were used to measure the RBD concentration with the fluorometric and colorimetric assays. All the measurements were repeated in triplicate.

Principle of fluorometric and colorimetric detection based on rolling circle amplification
The principle of analyte detection is presented in Scheme 1. In this work, we used streptavidin-coated magnetic beads to immobilize biotin-cDNA/aptamer to form MBs-cDNA/aptamer. In the presence of S protein RBD, there is a competition between RBD and the cDNA to bind the aptamer. After a magnetic separation, the solution including the aptamer-S protein RBD can be removed. Therefore, there is a positive correlation between the concentration of S protein RBD and the un-complexed cDNA on the MBs. The un-complexed cDNA acts as a primer to trigger RCA after addition of the circular DNA template and phi29 DNA polymerase. After the polymerization, the first signal readout is achieved using the F-Q probe that hybridizes with the RCA products (RCAPs). Because of the complimentary sequences between RCAPs and the F probe, strand displacement occurs to eliminate the quenching caused by the Q probe. The second signal readout is achieved by increasing the reaction time of RCA to form HRP-loaded DNA flower. After magnetic separation and washing, the HRP loaded in the DNA flower is used to catalyze the colorimetric reaction between TMB and H 2 O 2 to generate UV-visible signal.

Rolling circle amplification
The importance of primer to match the padlock DNA for synthesis of the circular DNA template was also investigated. In Fig. 1a, the steps for synthesis of the circular DNA template are illustrated. A circularized DNA template can be formed only when the primer is fully complementary to pair with the padlock DNA. A mismatched ssDNA is not able to allow the T-4 DNA ligase to produce the circularized DNA template from the padlock DNA. Figure 1b shows the electrophoretic DNA bands from the padlock DNA, the primer, the circular DNA template, and the products of the rolling circle amplification using the circular DNA template.

Fluorometric detection
To further prove the successful RCA-enabled fluorometric detection, we performed a series of fluorescence detection experiments by omitting one essential component of the RCA reaction. The results are compared with the RCAenabled fluorescence detection (Fig. 2a). In contrast to the RCA-enabled fluorescence detection that gives a strong emission, when a single RCA reaction component (phi29 DNA polymerase, T-4 DNA ligase, dNTPs, or 10 × phi29 DNA polymerase buffer) is omitted, the fluorescence signal became significantly reduced.

Colorimetric detection
To realize colorimetric detection for RBD, we selected to use HRP-loaded DNA flower to catalyze the commonly used colorimetric reaction between TMB and H 2 O 2 . As shown in Fig. 2b, when one of the essential components of the RCA reaction was missing, it was not possible to obtain a clear absorbance band as a response to RBD. Also, omitting any component of the colorimetric reaction (HRP, TMB, or H 2 O 2 ) made it impossible to observe the absorbance band at 450 nm. These results confirm that both the RCA-enabled DNA flower and the HRP-catalyzed reaction are indispensable for the colorimetric detection of RBD.
Formation of HRP-loaded DNA flower can be achieved using the same RCA reaction as used in the fluorometric detection (Fig. 3a). The morphology of the DNA flowers (DFs), with and without loaded HRP, was studied by scanning electron microscopy (SEM) (Figs. 3b, c). The extended rolling circle amplification generated long DNA strands, which complexed with Mg 2 PPi to form flower-like microparticles. The uneven, petal-like structure of the DFs provides a larger surface area, making the encapsulated HRP readily accessible to the substrates TMB and H 2 O 2 . For colorimetric detection, the signal output depends on the amount of HRP trapped in the DNA flowers, as well as the concentration of TMB and H 2 O 2 .

Evaluation of fluorometric and colorimetric methods for RBD detection
Using the optimized conditions, we studied the dose-response curves of the fluorometric and colorimetric methods for detection and quantification of S protein RBD. Standard solution of RBD of different concentrations (0, 0.001, 0.01, 0.1, 1, 10, 100 ng mL −1 ) was added to MBs-cDNA/apt. Complexation with RBD caused the aptamer to dissociate from the cDNA immobilized on the MBs. The un-complexed cDNA became available to trigger the RCA reaction. Figure 4a shows the change of fluorescence intensity in response to the varying RBD concentration. The correlation between RBD concentration and change of fluorescence intensity can be clearly seen in Fig. 4b. The semi-logarithmic plot shows a linear relation between fluorescence intensity and the logarithm of RBD concentration in the range of 0.001-100 ng mL −1 , with a detection limit of 0.11 pg mL −1 based on 3σ/m criterion. For the fluorometric method, the total analysis time is ~ 4 h.   Using the colorimetric method, we obtained a similar calibration curve for detection and quantification of RBD. As shown in Fig. 5, the absorbance of the assay solution increased linearly with increasing RBD concentration, in the semi-logarithmic plot in the range of 0.001 to 100 ng mL −1 . The limit of detection of the colorimetric assay is 0.904 pg mL −1 . For the colorimetric method, although the DNA flower synthesis takes 20 h, the signal readout time is very short (about 10 min).
To evaluate the selectivity of the two analytical methods, we replaced RBD with a series of SARS-related proteins and carried out the protein assays. As shown in Fig. 6, the fluorescence and absorbance signals caused by all the other proteins are much weaker than SARS-CoV-2 RBD. Obviously, the selectivity of the present protein assays is due to the selective molecular recognition (RBD binding) provided by the aptamer. The RCA provides effective amplification of the molecular binding signal, making it possible to detect  Table S3.

Detection of S protein RBD in biological samples
The accuracy of the fluorometric and colorimetric assays was assessed by studying analyte recovery from spiked biological samples. Saliva samples were spiked with S protein RBD and analyzed. The assay protocols for determination of RBD are described in the Electronic Supplementary Material (ESM). As shown in Table 1, the average recoveries of RBD from the spiked saliva samples were 93-100% and 87.2-107% for fluorometric and colorimetric assays, respectively. The relative standard deviation (RSD) of measurements were 4.03-10.6% and 6.15-9.04%, respectively. Therefore, the RCA-enabled protein assays based on fluorescence and absorbance measurements allow to detect and quantify SARS-CoV-2 RBD in untreated biological samples.
For more complex samples such as blood and serum, optical detection in the visible wavelength range may be affected by some interfering substances. This limitation can be mitigated by using alternative labels, e.g., long wavelength fluorescent dyes or upconversion fluorophores.

Conclusions
This work developed competitive fluorometric and colorimetric methods for detecting the receptor-binding domain protein of SARS-CoV-2 using aptamer and rolling circle amplification. The signal readout can be performed using fluorometric or colorimetric measurements, both of which offer accurate and reliable results. The RCA-enhanced methods have potential for virus particle detection. The principles of these protein assays can also be adapted for detecting other low-abundance proteins and biomarkers by changing the aptamer. One limitation of the present assays is the long analysis time, in particular the colorimetric method involving DNA flowers. This issue will be addressed in future studies using, e.g., faster signal amplification reactions.
Author contribution Man Zhang: preparation, creation and writing the initial draft; formal analysis; investigation; Lei Ye: project administration; funding acquisition; writing-review and editing.
Funding Open access funding provided by Lund University. This work was financially supported by the Swedish Research Council VR (grant number 2019-04228).
Data Availability All data generated or analysed during this study are included in this published article and its supplementary information files.

Conflict of interest The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.