Open Access
Add to BibSonomy
Abstract
Circulating tumor DNA (ctDNA) has recently emerged as an ideal target for biomarker analytes. Thus, the development of rapid and ultrasensitive ctDNA detection methods is essential. In this study, a high-throughput surface-enhanced Raman scattering (SERS)-based lateral flow assay (LFA) strip is proposed. The aim of this method is to achieve accurate quantification of TP53 and PIK3CA E545K, two types of ctDNAs associated with head and neck squamous cell carcinoma (HNSCC), particularly for point-of-care testing (POCT). Raman reporters and hairpin DNAs are used to functionalize the Pd-Au core-shell nanorods (Pd-AuNRs), which serve as the SERS probes. During the detection process, the existence of targets could open the hairpins on the surface of Pd-AuNRs and trigger the first step of catalytic hairpin assembly (CHA) amplification. The next stage of CHA amplification is initiated by the hairpins prefixed on the test lines, generating numerous “hot spots” to enhance the SERS signal significantly. By the combination of high-performing SERS probes and a target-specific signal amplification strategy, TP53 and PIK3CA E545K are directly quantified in the range of 100 aM-1 nM, with the respective limits of detection (LOD) calculated as 33.1 aM and 20.0 aM in the PBS buffer and 37.8 aM and 23.1 aM in human serum, which are significantly lower than for traditional colorimetric LFA methods. The entire detection process is completed within 45 min, and the multichannel design realizes the parallel detection of multiple groups of samples. Moreover, the analytical performance is validated, including reproducibility, uniformity, and specificity. Finally, the SERS-LFA biosensor is employed to analyze the expression levels of TP53 and PIK3CA E545K in the serum of patients with HNSCC. The results are verified as consistent with those of qRT-PCR. Thus, the SERS-LFA biosensor can be considered as a noninvasive liquid biopsy assay for clinical cancer diagnosis.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Head and neck squamous cell carcinoma (HNSCC) is the sixth most common malignancy worldwide, accounting for 4% of all new cancers and responsible for approximately 3.7% of cancer-related deaths [1,2]. Patients with HNSCC are frequently diagnosed at advanced stages, with a poor five-year relative survival rate of approximately 50%. An estimated 30% of patients relapse within two years of treatment, despite aggressive treatment with surgery, radiotherapy, chemotherapy, or any combination thereof [3,4]. Thus, there is an urgent need to develop innovative technologies and specific biomarkers to predict tumor aggressiveness, metastasis potential, and treatment response.
Recently, circulating tumor DNA (ctDNA), a fraction of cell-free DNA (cfDNA) derived from cancer cells typically found in plasma or serum, has attracted increasing attention as a promising biomarker for early-stage cancer diagnosis [5–7]. Researchers have found that ctDNA expression levels differ depending on the progression of the primary tumor, and the efficacy of ctDNA as a cancer biomarker has been validated. Thus, ctDNA analysis may provide the possibility of noninvasive detection of HNSCC progression [8]. Traditional methods for ctDNA detection mainly include polymerase chain reaction (PCR)-based methods (real-time PCR and digital PCR) and sequencing-based methods [9–11]. However, PCR-based methods usually require a high integrity of the template DNA, and sequencing-based techniques are restricted by their high cost and cumbersome processes. Thus, sensitive detection methods are required to monitor low-frequency ctDNAs in patients with early-stage HNSCC, particularly in point-of-care testing (POCT).
The lateral flow assay (LFA), a paper-based POC diagnostic tool, has attracted significant interest because of its advantages, including low cost, long-term stability, user-friendly format, and rapid detection time [12–15]. Previously, LFA strips have been extensively used to analyze various disease-related biomarkers [16–18], but they remain unable to meet the detection requirements for low-abundance ctDNAs. To overcome these limitations, surface-enhanced Raman scattering (SERS) has been introduced to LFA systems, owing to its merits, such as fingerprinting resolution and single-molecule-level sensitivity features [19–22]. Typically, electromagnetic enhancement (EM) and chemical enhancement (CM) are the two main enhancement mechanisms attributed to the ultrahigh sensitivity of SERS, and EM is widely accepted as the primary factor [23]. Thus, noble metals (e.g., Au and Ag) with unique architectures that can generate strong SERS signals, have been widely applied.
Among various SERS-active substrates, Pd-Au core-shell nanorods (Pd-AuNRs), which are bimetallic nanomaterials, have received significant attention [24]. The relatively thin Pd shell means Pd-AuNRs exhibit a well-defined localized surface plasmon resonance (LSPR) band in the visible and near-infrared regions, leading to a high dielectric sensitivity. The core-shell structure enables the Pd shell to borrow high SERS activity from the inner Au core, owing to long-range electromagnetic enhancement. It should be noted that the SERS signals in Pd-AuNRs gaps are more robust than those on the surfaces, owing to the significant effects of the EM. Hence, combining the LFA system with the Pd-AuNR-based SERS technique could present new avenues for the highly efficient quantification of ctDNAs.
To further enhance the sensitivity of ctDNA detection, various signal amplification strategies have been developed, such as rolling circle amplification (RCA), strand displacement amplification (SDA), duplex-specific nuclease (DSN), and the hybridization chain reaction (HCR) [25–28]. Although these strategies can significantly improve the detection sensitivity, their applications remain limited by poor reproducibility, high cost, susceptibility to reaction conditions, and complicated operation processes. Catalytic hairpin assembly (CHA), an enzyme-free isothermal signal amplification technique, has emerged as a highly promising tool for signal enhancement, owing to its rapid turnover rates, leading to high sensitivity and low background associated with target-specific reactions [29–32]. During the detection process, double-stranded structures can form via a hybridization reaction between two hairpins, only when the targets are present. Thus, the CHA can significantly amplify the detection signal to ensure specificity. In addition, hairpins can be easily modified onto nanomaterial surfaces by functional groups, providing the possibility of introducing CHA to SERS.
Here, this study proposes a novel CHA-assisted SERS-LFA biosensor to quantitatively investigate the expression levels of TP53 and PIK3CA E545K (two HNSCC-related ctDNAs). Raman reporters, hairpins, and mouse monoclonal antibodies (MIgG) are used to functionalize Pd-AuNRs as two SERS probes. Cross-reactivity between TP53 and PIK3CA E545K is evaluated to determine the feasibility of simultaneous detection. To ensure a strong analytical performance, several experimental conditions are conducted. Analytical performance parameters, such as reproducibility, uniformity, specificity, and sensitivity are evaluated. Finally, the SERS-LFA biosensor is employed for the expression level analysis of TP53 and PIK3CA E545K in the serum of HNSCC patients and healthy subjects. The assay results are compared with those obtained by qRT-PCR to validate the accuracy of the proposed biosensor for clinical applications. To the best of the authors’ knowledge, few studies have been reported on high-throughput SERS-based LFA (SERS-LFA) biosensors for ultrasensitive detection of ctDNAs using CHA as the signal amplification strategy. Thus, the proposed SERS-LFA biosensor provides new insights into POC diagnostics of HNSCC.
2. Experimental section
2.1 Reagents and materials
Chloroauric acid tetrahydrate (HAuCl4), palladium chloride (PdCl2), ammonium sulfate ((NH4)2SO4), silver nitrate (AgNO3), and hydrochloric acid (HCl) were obtained from Sigma-Aldrich. Sodium hydroxide (NaOH), bovine serum albumin (BSA), sodium oleate (NaOL), sodium borohydride (NaBH4), ascorbic acid (AA), carbodiimide (EDC), hexadecyl trimethyl ammonium bromide (CTAB), and tris acetate (C6H15NO5) were obtained from Sinopharm Chemical Reagent Co. Ltd. 5,5'-dithiobis-2-nitrobenzoic acid (DTNB), 4-aminothiophenol (4-ATP), N-hydroxysuccinimide (NHS), and tris-(2-carboxyethyl) phosphine hydrochloride (TCEP) were obtained from J&K Scientific Ltd. Mouse monoclonal antibody (MIgG), goat anti-mouse antibody (GMIgG), streptavidin (SA), and qRT-PCR kits were obtained from GeteinBiotech (China). All the reagents were used without further purification. Milli-Q water (resistivity > 18 M) was used for the experiments. The sample pads, nitrocellulose (NC) membranes, PVC sheets, absorption pads, and conjugate pads were obtained from Jieyi Biotechnology Co. Ltd. Oligonucleotides used in this study (Table 1) were obtained from R&D Systems (Minneapolis, USA).
Table 1. Name and sequence of oligonucleotides used in the experiment
2.2 Clinical specimens
Human blood samples from healthy subjects (n = 30) and patients with HNSCC at different stages (n = 120) were collected from the College of Clinical Medicine of Yangzhou University. All serum samples were obtained via venipuncture into vacutainer tubes and centrifugation (12,000 rpm, 4 °C, 10 min), followed by storage at -80 °C until required. All experiments were approved by the Ethics Committee of Yangzhou University. The details of the samples are presented in Table 2.
Table 2. Details of study volunteers
2.3 Preparation of Pd-AuNRs
According to the method reported by Wang et al., Au nanorods (AuNRs) were obtained [33]. Then, a 90 µL AuNRs solution was mixed with 0.5 mL of 0.1 M CTAB followed by 1 h of ultrasound and centrifugation. Subsequently, the precipitate was mixed with 0.1 M CTAB to produce 150 µL, then 25 µL of the resulting solution was mixed with 1 mL of 10 mM CTAB. After 10 min, 10 µL of 1 mM CuCl2 and 25 µL of 10 mM H2PdCl4 were added, before adding 12.5 µL of 0.05 M AA 5 min later and shaking to mix. Finally, centrifugation (9000 rpm, 15 min) was performed after storage at 30 °C for 10 h to synthesize Pd-AuNRs.
2.4 Preparation of SERS probes
The SERS probes were prepared as follows. Briefly, 0.5 µL of 0.1 mM 4-ATP solution was mixed with with 1.0 mL of 1 nM Pd-AuNRs, then reacted for 0.5 h under continuous stirring. Prior to use, 50 µL of 10−4 M bio-hp1-1 was mixed with 50 µL of 5 mM TECP in a 50 mM Tris-HCl solution (pH 7.4) and incubated at room temperature for 1 h). Thus, the disulfide groups on the hairpins were activated and converted into free thiol groups. Then, 25 µL of activated hp1-1 was added to 1 mL of Pd-AuNRs@4-ATP solution, and the mixture was incubated in dark conditions for 24 h. Subsequently, 10 µL of 30 mM NHS and 10 µL of 150 mM EDC were mixed with bio-hp1-1 modified Pd-AuNRs@4-ATP and incubated for 2 h to activate the carboxyl group of 4-ATP. Next, 100 µL of 4 µg/µL MIgG was added and reacted for 4 h. Then, 10 µL of 3% BSA solution in PBS was added to shield the bare sites. After that, salt aging of the obtained mixture solution was performed by adding 10 µL of 5 M NaCl three times at 8 h intervals. The mixture was purified by centrifugation (7000 rpm, 10 min). Thus, SERS probes (Pd-AuNRs@4-ATP@bio-hp1-1@MIgG) and (Pd-AuNRs@DTNB@ bio-hp1-2@MIgG) were prepared similarly.
2.5 Preparation of LFA test strips
The LFA test strips consisted of a sample pad, conjugate pad, NC membrane, and absorption pad. Each pad was overlapped by 1.5 mm to ensure continuous flow of the sample solution via capillary action. SERS probes were added and dried at room temperature to prepare the conjugated pads. Then, 3 µL of hp2-1 (0.1 mM) and 10 µL of SA (0.3 mg/mL) were dispensed onto the NC membrane, forming test line 1 (T1 line). Similarly, test line 2 (T2 line) was prepared by spreading 4 µL of hp2-2 (0.1 mM) and 10 mL of SA (0.3 mg/mL). Subsequently, the control line (C line) was prepared by spreading 10 µL of 1 mg/mL GM IgG. Each component was assembled in turn and cut into 4 mm widths. Finally, eight strips prepared from the same batch were assembled on a disc-like PVC plate.
2.6 LFA-SERS procedure for ctDNA detection
A total of 150 µL of sample solutions with varying concentrations were added dropwise to the sample pad. The solution then flowed toward the absorption pad via capillary action. When the conjugate pad was reached, the targets hybridized with the hairpins on the SERS probes and subsequently flowed together. Color on the T1, T2, and C lines developed over time, and all procedures were completed within 45 min. The SERS probes were anchored onto the T lines, owing to the reaction between SA and biotin, resulting in the aggregation of Pd-AuNRs and enhancement of the SERS signal. Then, SERS spectra on the T lines were recorded with a 785 nm laser excitation, and the acquisition time and laser power were fixed at 10 s and 10 mW, respectively. Each SERS spectrum was measured at 10 random spots on the T lines.
2.7 Instrumentation
Transmission electron microscopy (TEM) images of the morphologies of the Pd-AuNRs were obtained using a TECNAI 12 microscope at an accelerating voltage of 60 kV. A S-4800II scanning electron microscopy (SEM) system (Hitachi) was used to obtain the SEM images. High-resolution transmission electron microscopy (HRTEM) images were obtained using a FEI field-emission TEM at 200 kV. Ultraviolet-visible-near-infrared (UV-Vis-NIR) spectrum images were recorded using a Shimadzu UV-3600 plus spectrometer. Raman spectra were obtained using a Renishaw Invia Raman microspectrometer system.
3. Results and discussion
3.1 Principle of high-throughput SERS-LFA biosensor
As shown in Fig. 1(A), DTNB and 4-ATP were anchored on Pd-AuNR surfaces via Au-S bonds, serving as Raman reporters. Using the Au-S bond, biotinylated hairpins (bio-hp1-1 and bio-hp1-2) were modified on 4-ATP and DTNB-modified Pd-AuNRs. Using this stable amide bond, MIgG was connected to 4-ATP and DTNB. Thus, Pd-AuNRs were functionalized to form SERS probes (Pd-AuNRs@4-ATP@bio-hp1-1@MIgG and Pd-AuNRs@DTNB@bio-hp1-2@MIgG). Figure 1(B) clearly indicates that the proposed SERS-LFA biosensor was disc-like and consisted of eight parallel channels, enabling the simultaneous detection of target ctDNAs in different samples. During the detection process, the sample solution was added to the sample pad. When flowing to the conjugated pad, the presence of TP53 (PIK3CA E545K) opened the hairpin structure of bio-hp1-1 (bio-hp1-2) followed by a hybridization reaction, forming the unstable intermediate Pd-AuNRs@4-ATP@ bio-hp1-1-TP53@MIgG (or Pd-AuNRs@DTNB@bio-hp1-2-PIK3CA E545K@MIgG) and exposing biotin molecules.
Fig. 1. (A) Schematic illustration of the preparation of two types of SERS probes. (B) Assembly of the SERS-LFA strip and the schematic diagram of the mechanism for ctDNA detection.
As the fluid continued to flow and arrived at T lines, the exposed biotin molecules were connected to SA prefixed on T lines, and the exposed sequence of bio-hp1-1 (bio-hp1-2) was assembled with hp2-1 (hp2-2) on the T1 (T2) line to displace TP53 (PIK3CA E545K). Thus, stable duplexes (Pd-AuNRs@4-ATP@bio-hp1-1-hp2-1@MIgG (or Pd-AuNRs@DTNB@ bio-hp1-2-hp2-2@MIgG) were formed on T1 and T2 lines, respectively, and the released TP53 (or PIK3CA E545K) triggered the next cycle of the CHA process. As the reaction progressed, an increasing number of SERS probes aggregated on the T lines, with two strong gray lines. Owing to the aggregation of Pd-AuNRs on the T lines, caused by the CHA reaction, strong electromagnetic enhancement amplified the signal significantly. Excess SERS probes continued to flow, and the MIgG on the Pd-AuNR surfaces was captured by the GMIgG on the C line, with one strong gray line. The quantitative analysis showed that the color and SERS intensities on the test line were inversely proportional to the concentrations of TP53 and PIK3CA E545K.
3.2 Characterization of Pd-AuNRs
The SEM images in Fig. 2(A) indicate that the Pd-AuNRs had a homogeneous morphology with an average size of 71 nm and a width of 30 nm. In addition, the dispersion of Pd-AuNRs was satisfactory. The TEM images in Fig. 2(B) validate the core-shell structure. High-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) and energy-dispersive X-ray spectroscopy (EDX) mapping were used to study the structural composition further (Fig. 2(C)). These results indicate that Pd-AuNRs consisted of a thin Pd shell and a compact AuNR core. The EDX spectrum in Fig. 2(F) further confirms the basic composition of Pd-AuNRs (Cu from the copper net).
Fig. 2. (A) Typical SEM image and (B) TEM image of Pd-AuNRs. (C) HAADF-STEM images and EDX mapping of Pd-AuNRs. (D) EDX spectrum, (E) HRTEM images and (F) SAED pattern of Pd-AuNRs. (G) UV-Vis-NIR absorption spectrum of Pd-AuNRs. (H) SERS spectra of NBA-labeled Pd-AuNRs (1×10−6 M) and pure NBA (1×10−2 M).
HRTEM images and selective area electron diffraction (SAED) patterns were employed to investigate the detailed structure. The interplanar spacing of 0.19 nm in Fig. 2(D) corresponds to the Pd (100) lattice spacing. The cylindrical shape of the AuNRs was formed because of the side facets consisting of four (110) facets and four (100) facets. The square Pd shell was formed because of the disappearance of the (110) facets, caused by competition for growth between the Pd (110) and Pd (100) side facets. The SAED pattern in Fig. 2(E) confirms that the obtained Pd-AuNRs were polycrystalline, and the bright circular rings correspond to the (111), (200), (220), and (311) planes. To study the plasmonic characteristics of the prepared Pd-AuNRs, UV-Vis-NIR spectra were recorded (Fig. 2(G)). They exhibited distinct longitudinal and transverse surface plasmon resonances (SPR) peaks at approximately 749 and 520 nm, respectively.
To verify the SERS enhancement effect, the SERS spectra of the pure NBA- and NBA-modified Pd-AuNRs were measured (Fig. 2(H)). The characteristic peak observed at 592 cm-1 was attributed to the in-plane vibration of CCC and CNC of NBA [34]. These results demonstrate that the SERS signal of Pd-AuNRs@NBA was significantly stronger than that of pure NBA. This is because the bimetallic core-shell structure generated numerous electromagnetic “hot spots” around gaps, tips, and edges of intra- and interparticles. The analytical enhancement factor (AEF) was calculated using the following equation: AEF= (ISERS/CSERS)/(IRS/CRS), which can be interpreted as the ratio of the SERS intensity (ISERS) of a given mode to the intensity of the Raman signal (ISR) of the same mode for a given analyte, both normalized to the respective concentration (C). The AEF was calculated as 7.17×106, indicating a strong SERS enhancement effect of Pd-AuNRs.
3.3 Characterization of SERS probes
The SERS probes applied in the experiment were prepared by labeling Raman molecules, biotinylated hairpins, and MIgG onto the Pd-AuNR surfaces. Figure 3(A) displays the absorption spectra of the SERS probes (Pd-AuNRs@DTNB@bio-hp1-2@MIgG) and Pd-AuNRs. This clearly demonstrates a slight red shift from 749 to 774 nm. The dynamic light scattering (DLS) distribution (Fig. 3(B)) revealed that the average diameter changed from 71 to 75 nm, proving that Pd-AuNRs were successfully functionalized. Moreover, the SERS spectra of native Pd-AuNRs and SERS probes were recorded (Fig. 3(C)). The characteristic peak at 1330 cm-1 was attributed to the C-N stretching vibrational modes of DTNB [35]. The results indicate that the SERS probes showed a strong signal intensity, whereas the native Pd-AuNRs had no SERS signal. To verify the feasibility under harsh conditions, the stability of the Pd-AuNRs and SERS probes was investigated under various salt concentrations, as NaCl is known to induce nanoparticle aggregation. Figure 3(D) illustrates the intensity changes of the normalized UV-Vis-NIR absorption maxima (749 nm for Pd-AuNRs and 774 nm for SERS probes). For Pd-AuNRs, the normalized intensity decreased significantly with an increasing NaCl concentration. Alternatively, the intensity changes of SERS probes were modest, even under high NaCl concentrations because BSA provided effective stabilization against aggregation.
Fig. 3. (A) UV-Vis-NIR absorption spectrum of Pd-AuNRs and SERS probes. (B) Dynamic light scattering distribution of Pd-AuNRs (blue columns) and SERS probes (yellow columns). (C) SERS spectra of Pd-AuNRs (yellow line) and SERS probes (blue line). (D) Changes in the normalized UV-Vis-NIR absorption intensity of Pd-AuNRs and SERS probes at various NaCl concentrations.
3.4 Evaluation of cross-reactivity
To simultaneously detect TP53 and PIK3CA E545K, it was necessary to evaluate cross-reactivity. In this study, 1 nM PIK3CA E545K was mixed with different concentrations of TP53 (100 aM-1 nM). As shown in Fig. 4(A), the T2 line color was identical in all samples, regardless of the concentration of TP53, indicating that the presence of TP53 did not affect the detection results of PIK3CA E545K. In contrast, the T1 line color changed to light gray at 1 pM, and no gray bands were found when the concentration was lower than 1 pM. Then, the SERS intensities at 1083 cm-1 (C-S stretching vibrational mode of 4-ATP) of the T1 line and 1330 cm-1 of the T2 line were measured (Fig. 4(B)) [36]. The results reveal that the SERS intensity on the T2 line was constant, whereas the SERS intensity on the T1 line increased with the TP53 concentration. Therefore, the cross-reactivity was satisfactory and simultaneous detection was feasible.
Fig. 4. (A) Digital photographic images and (B) corresponding SERS intensity of T1 and T2 lines at 1083 cm-1 and 1330 cm-1. I: PIK3CA E545K, 1 nM; TP53, 100 aM; II: PIK3CA E545K, 1 nM; TP53, 1 fM; III: PIK3CA E545K, 1 nM; TP53, 10 fM; IV: PIK3CA E545K, 1 nM; TP53, 100 fM; V: PIK3CA E545K, 1 nM; TP53, 1 pM; VI: PIK3CA E545K, 1 nM; TP53, 10 pM; VII: PIK3CA E545K, 1 nM; TP53, 100 pM; VIII: PIK3CA E545K, 1 nM; TP53, 1 nM.
3.5 Qualitative analysis
Figure 5(A) shows images of the simultaneous analysis of TP53 and PIK3CA E545K, with the corresponding SERS spectra presented in Figs. 5(B)–5(E). Notably, the color of the C line was constant, regardless of the target concentration, indicating that the C line worked effectively by capturing SERS probes through the GMIgG-IgG reaction. As shown in Fig. 5(A)(IV), the colors of the T1 and T2 lines did not change when no target was included in the sample solution. When both TP53 and PIK3CA E545K were present, two visible gray bands were observed at T1 and T2 (Fig. 5(A)(I)). When only TP53 or PIK3CA E545K was included, only one band changed to grey (Fig. 5(A)(III) and Fig. 5(A)(II)). Then, the SERS spectra of the T lines were recorded to verify their accuracy, with satisfactory results.
Fig. 5. (A) Digital photographic images and corresponding SERS spectra of the SERS-LFA biosensor in the presence of (B) TP53, 1 nM; PIK3CA E545K, 1 nM; (C) TP53, 1 nM; PIK3CA E545K, 0 nM; (D) TP53, 0 nM; PIK3CA E545K, 1 nM; (E) TP53, 0 nM; PIK3CA E545K, 0 nM.
3.6 Optimization of experimental parameters
For optimal analytical performance, several significant factors were investigated before quantitative detection. The signal-to-noise ratio (S/N) was applied as the index parameter, where S and N represent the SERS intensities in the presence and absence of target ctDNAs, respectively. As the signal intensity on the T lines mainly depends on the number of captured SERS probes, this directly affected the detection sensitivity. As shown in Fig. 6(A), different volumes of the SERS probes for TP53 (2, 3, 4, 5, and 6 µL) were added to the conjugated pad. The results illustrate that a maximum S/N ratio was obtained when the volume was 4 µL. Thus, the optimal volume of SERS probes for TP53 was 4 µL. Similarly, the optimal volume of SERS probes for PIK3CA E545K was 3 µL (Fig. 6(B)).
Fig. 6. Optimization SERS probe volume for (A) TP53 and (B) PIK3CA E545K. Optimization of (C) buffer solution type and (D) incubation time.
As the migration of targets and SERS probes strongly relies on the running buffer, a suitable type of buffer can achieve strong analytical performance. Here, three types of buffer solutions (PBS, HEPES, and tris-acetate) were applied, and the results in Fig. 6(C) indicate that the S/N values for TP53 and PIK3CA E545K in the PBS buffer were significantly better than those in the other two buffers. Therefore, the PBS buffer was used in the subsequent assays. As an amplification strategy for the SERS intensity, the efficiency of the CHA reaction can greatly influence the detection sensitivity. Thus, incubation time was investigated to achieve the best amplification efficiency. As shown in Fig. 6(D), the SERS intensity increased rapidly over time in the window of 25-45 min, whereas it decreased for 45-65 min. In summary, the optimal incubation time was found to be 45 min.
3.7 Analytical performance evaluation
The analytical performance of the proposed high-throughput SERS-LFA biosensor, including reproducibility, uniformity, and specificity, was assessed. Five biosensors, prepared in different batches, were used to evaluate reproducibility (Figs. 7(A) and 7(B)). The results indicate that no obvious difference in the shape of the spectra was observed, and only a slight distinction in the intensity was observed. Therefore, the developed SERS-LFA biosensor demonstrated strong reproducibility between measurements. Furthermore, uniformity was assessed by measuring the SERS signals of 10 random points from the T1 and T2 lines, as shown in Figs. 7(C) and 7(E). Negligible differences between the spectra were observed. As shown in Figs. 7(D) and 7(F), the peak intensities (1083 cm-1 and 1330 cm-1) of the selected points were relatively consistent, and the corresponding relative standard deviations (RSD) were calculated as 6.9% and 7.3%, respectively, indicating that the SERS-LFA biosensor has strong uniformity.
Fig. 7. SERS spectra of the (A) T1 line and (B) T2 line on the SERS-LFA biosensors prepared in different batches. SERS spectra of ten random spots on (C) the T1 line and (E) the T2 line. Corresponding histogram of SERS intensity at (D) 1083 cm-1 and (F) 1330 cm-1. (G) SERS spectra on the T1 line for different analyses and the corresponding histogram of SERS intensity (inset). (H) SERS spectra on the T2 line for different analyses.
When multiple ctDNAs were present in a sample, it was necessary to ensure that the proposed biosensor could distinguish the target from interference. To investigate the specificity of the methodology, negative control tests were performed for both TP53 and PIK3CA E545K. Figure 7(G) shows the SERS spectra of TP53, single-mismatched DNAs (MT1-1), triple-mismatched DNAs (MT3-1), and random and blank DNA. As expected, the SERS signal was significantly enhanced in the presence of TP53 because only TP53 could open the hairpins on the SERS probes to initiate CHA amplification. Similarly, the specificity of PIK3CA E545K was strong (Fig. 7(H)). These results clearly indicate that the proposed biosensor has satisfactory specificity.
3.8 Quantitative determination in PBS and human serum
Under the optimal experimental parameters, the SERS-LFA biosensor was used for quantitative determination of TP53 and PIK3CA E545K in the PBS buffer, to study the sensitivity using standard solutions with different mass concentrations (100 aM, 1 fM, 10 fM, 100 fM, 1 pM, 10 pM, 100 pM, 1 nM). The images clearly indicate that when the target concentrations decreased, fewer SERS probes were captured on the T lines, and the corresponding color also became lighter. The results demonstrate difficulties in observing color changes with the naked eye, when the concentrations of TP53 and PIK3CA E545K were lower than 1 pM and 100 fM, respectively.
To obtain more sensitive and quantitative results, the SERS spectra of the T lines were measured. As shown in Figs. 8(A) and 8(C), the SERS intensity was concomitantly enhanced with increasing target concentrations. The corresponding calibration curves are shown in Fig. 8(B), based on the SERS intensity at 1083 cm-1. The regression equation was y = 1479.44x-1807.52, where y represents the SERS intensity at 1083 cm-1 and x represents the logarithm of TP53 concentration, with an R2 of 0.9911. Similarly, the regression equation for PIK3CA E545K in Fig. 8(D) was y = 1610.93x-2004.41, with R2 = 0.9903. Then, the limit of detection (LOD) for TP53 and PIK3CA E545K was calculated as 33.12 aM and 20.01 aM, respectively.
Fig. 8. SERS spectra of (A) TP53 and (C) PIK3CA E545K in the PBS buffer with different concentrations (9: blank; 8: 100 aM; 7: 1 fM; 6: 10 fM; 5: 100 fM; 4: 1 pM; 3: 10 pM; 2: 100 pM; 1: 1 nM). Corresponding calibration curves of SERS intensities at (B) 1083 cm-1 and (D) 1330 cm-1.
The potential practicality of the biosensor in serum was also evaluated, and various concentrations of TP53 and PIK3CA E545K combined with human serum were analyzed. As shown in the images, the color of the T line became lighter with a decrease in target concentration. The SERS spectra in Figs. 9(A) and 9(C) show that the SERS signal increased with the target concentration. As shown in Fig. 9(B), the calibration curves show that a linear relationship exists between the peak intensity and the logarithm of TP53 concentration, with a regression equation of y = 1376.43x-1679.24 (R2 = 0.9821). Figure 9(D) illustrates that the regression equation for PIK3CA E545K was y = 1576.37x-1991.11, with R2 = 0.9927. The LOD for TP53 and PIK3CA E545K was calculated as 37.82 aM and 23.06 aM, respectively, which are significantly lower than those of existing methods (Table 3). The quantitative determination of TP53 and PIK3CA E545K was achieved using the developed SERS-LFA biosensor.
Fig. 9. SERS spectra of (A) TP53 and (C) PIK3CA E545K in human serum with different concentrations (9: blank; 8: 100 aM; 7: 1 fM; 6: 10 fM; 5: 100 fM; 4: 1 pM; 3: 10 pM; 2: 100 pM; 1: 1 nM). Corresponding calibration curves of SERS intensities at (B) 1083 cm-1 and (D) 1330 cm-1.
Table 3. Comparison of the SERS-LFA biosensor with existing detection methods
3.9 Clinical serum sample analysis
To estimate practical applications, serum samples from 30 healthy volunteers and 120 HNSCC patients were collected and modelled. The average SERS spectra of TP53 and PIK3CA E545K in human serum on the test lines are shown in Figs. 10(A) and 10(B). The corresponding histograms of the SERS intensities at 1083 cm-1 and 1330 cm-1 are shown in Fig. 10(C). The results clearly reveal that the expression levels of TP53 and PIK3CA E545K increased with the progression of HNSCC, proving that these two types of ctDNA can be applied as HNSCC-specific biomarkers. The concentrations of TP53 and PIK3CA E545K were then obtained by substituting the peak intensities into the above regression equation. The SERS results were compared with those obtained by qRT-PCR, the gold standard, to validate the accuracy of the proposed SERS-LFA biosensor. As shown in Table 4, the SERS results were consistent with the qRT-PCR results. Therefore, the proposed SERS-LFA biosensor is feasible for the clinical detection of ctDNAs.
Fig. 10. SERS spectra of (A) TP53 and (B) PIK3CA E545K in healthy subjects and HNSCC patients at different stages (I, II, III and IV). (C) Corresponding histogram of SERS intensities at 1083 cm-1 and 1330 cm-1.
Table 4. Results of the SERS-LFA biosensor and qRT-PCR for real sample detection
4. Conclusions
In summary, a high-throughput SERS-LFA biosensor was established for the POC detection of TP53 and PIK3CA E545K, with CHA as the signal amplification strategy. Based on the unique properties of Pd-AuNRs and the CHA strategy, ultralow LODs (37.8 aM of TP53 and 23.1 aM of PIK3CA E545K) in serum were achieved. The proposed SERS-LFA biosensor was successfully applied for the quantitative detection of TP53 and PIK3CA E545K in HNSCC patients, and the results were validated using conventional qRT-PCR. Moreover, the reproducibility, uniformity, and specificity were satisfactory. It is worth noting that the complete detection process can be completed within 45 min, and the multichannel design can realize the parallel detection of multiple groups of samples. These results demonstrate that the proposed SERS-LFA biosensor is an alternative method for efficient detection of ctDNA biomarkers.
Funding
National Natural Science Foundation of China (No.81701825); the Social Development Foundation of Jiangsu (No. BE2018684); the Natural Science Foundation of the Jiangsu Higher Education Institutions of China (No.17KJB416012).
Disclosures
The authors declare that there are no conflicts of interest related to this article.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. B. Solomon, R. J. Young, and D. Rischin, “Head and neck squamous cell carcinoma: Genomics and emerging biomarkers for immunomodulatory cancer treatments,” Semin. Cancer Biol. 52(Pt 2), 228–240 (2018). [CrossRef]
2. G. Wang, M. Zhang, M. Cheng, X. Wang, K. Li, J. Chen, Z. Chen, S. Chen, J. Chen, G. Xiong, X. Xu, C. Wang, and D. Chen, “Tumor microenvironment in head and neck squamous cell carcinoma: Functions and regulatory mechanisms,” Cancer Lett. 507, 55–69 (2021). [CrossRef]
3. L. Sepiashvili, A. Hui, V. Ignatchenko, W. Shi, S. Su, W. Xu, S. H. Huang, B. O’Sullivan, J. Waldron, J. C. Irish, B. Perez-Ordonez, F.-F. Liu, and T. Kislinger, “Potentially novel candidate biomarkers for head and neck squamous cell carcinoma identified using an integrated cell line-based discovery strategy,” Mol. Cell. Proteomics 11(11), 1404–1415 (2012). [CrossRef]
4. A. Diez-Fraile, J. De Ceulaer, C. Derpoorter, C. Spaas, T. De Backer, P. Lamoral, J. Abeloos, and T. Lammens, “Tracking the Molecular Fingerprint of Head and Neck Cancer for Recurrence Detection in Liquid Biopsies,” Int. J. Mol. Sci. 23(5), 2403 (2022). [CrossRef]
5. A. Campos-Carrillo, J. N. Weitzel, P. Sahoo, R. Rockne, J. V. Mokhnatkin, M. Murtaza, S. W. Gray, L. Goetz, A. Goel, N. Schork, and T. P. Slavin, “Circulating tumor DNA as an early cancer detection tool,” Pharmacol. Ther. 207, 107458 (2020). [CrossRef]
6. J. D. Merker, G. R. Oxnard, C. Compton, M. Diehn, P. Hurley, A. J. Lazar, N. Lindeman, C. M. Lockwood, A. J. Rai, R. L. Schilsky, A. M. Tsimberidou, P. Vasalos, B. L. Billman, T. K. Oliver, S. S. Bruinooge, D. F. Hayes, and N. C. Turner, “Circulating Tumor DNA Analysis in Patients With Cancer: American Society of Clinical Oncology and College of American Pathologists Joint Review,” J. Clin. Oncol. 36(16), 1631–1641 (2018). [CrossRef]
7. M. L. Cheng, E. Pectasides, G. J. Hanna, H. A. Parsons, A. D. Choudhury, and G. R. Oxnard, “Circulating tumor DNA in advanced solid tumors: Clinical relevance and future directions,” Ca-Cancer J. Clin. 71(2), 176–190 (2021). [CrossRef]
8. M. Tellez-Gabriel, E. Knutsen, and M. Perander, “Current Status of Circulating Tumor Cells, Circulating Tumor DNA, and Exosomes in Breast Cancer Liquid Biopsies,” Int. J. Mol. Sci. 21(24), 9457 (2020). [CrossRef]
9. T. Mei, X. Lu, N. Sun, X. Li, J. Chen, M. Liang, X. Zhou, and Z. Fang, “Real-time quantitative PCR detection of circulating tumor cells using tag DNA mediated signal amplification strategy,” J. Pharm. Biomed. Anal. 158, 204–208 (2018). [CrossRef]
10. D. Sefrioui, L. Beaussire, A. Perdrix, F. Clatot, P. Michel, T. Frebourg, F. Di Fiore, and N. Sarafan-Vasseur, “Direct circulating tumor DNA detection from unpurified plasma using a digital PCR platform,” Clin. Biochem. 50(16-17), 963–966 (2017). [CrossRef]
11. A. M. Aravanis, M. Lee, and R. D. Klausner, “Next-Generation Sequencing of Circulating Tumor DNA for Early Cancer Detection,” Cell 168(4), 571–574 (2017). [CrossRef]
12. Y. Liu, L. Zhan, Z. Qin, J. Sackrison, and J. C. Bischof, “Ultrasensitive and Highly Specific Lateral Flow Assays for Point-of-Care Diagnosis,” ACS Nano 15(3), 3593–3611 (2021). [CrossRef]
13. S. Kumar, R. Gallagher, J. Bishop, E. Kline, J. Buser, L. Lafleur, K. Shah, B. Lutz, and P. Yager, “Long-term dry storage of enzyme-based reagents for isothermal nucleic acid amplification in a porous matrix for use in point-of-care diagnostic devices,” Analyst 145(21), 6875–6886 (2020). [CrossRef]
14. G. Liu, A. S. Gurung, and W. Qiu, “Lateral Flow Aptasensor for Simultaneous Detection of Platelet-Derived Growth Factor-BB (PDGF-BB) and Thrombin,” Molecules 24(4), 756 (2019). [CrossRef]
15. B. Lin, Z. Guan, Y. Song, E. Song, Z. Lu, D. Liu, Y. An, Z. Zhu, L. Zhou, and C. Yang, “Lateral flow assay with pressure meter readout for rapid point-of-care detection of disease-associated protein,” Lab Chip 18(6), 965–970 (2018). [CrossRef]
16. M. Magiati, A. Sevastou, and D. P. Kalogianni, “A fluorometric lateral flow assay for visual detection of nucleic acids using a digital camera readout,” Microchim. Acta 185(6), 314 (2018). [CrossRef]
17. J. Li, J. Ding, X.-L. Liu, B. Tang, X. Bai, Y. Wang, W.-D. Qiao, M.-Y. Liu, and X.-L. Wang, “Upconverting phosphor technology-based lateral flow assay for the rapid and sensitive detection of anti-Trichinella spiralis IgG antibodies in pig serum,” Parasites Vectors 14(1), 487 (2021). [CrossRef]
18. S. Yu, S. B. Nimse, J. Kim, K.-S. Song, and T. Kim, “Development of a Lateral Flow Strip Membrane Assay for Rapid and Sensitive Detection of the SARS-CoV-2,” Anal. Chem. 92(20), 14139–14144 (2020). [CrossRef]
19. C. Chen, Y. Li, S. Kerman, P. Neutens, K. Willems, S. Cornelissen, L. Lagae, T. Stakenborg, and P. Van Dorpe, “High spatial resolution nanoslit SERS for single-molecule nucleobase sensing,” Nat. Commun. 9(1), 1733 (2018). [CrossRef]
20. Q. Ai, J. Zhou, J. Guo, P. Pandey, S. Liu, Q. Fu, Y. Liu, C. Deng, S. Chang, F. Liang, and J. He, “Observing dynamic molecular changes at single-molecule level in a cucurbituril based plasmonic molecular junction,” Nanoscale 12(32), 17103–17112 (2020). [CrossRef]
21. X. Tan, J. Melkersson, S. Wu, L. Wang, and J. Zhang, “Noble-Metal-Free Materials for Surface-Enhanced Raman Spectroscopy Detection,” Chemphyschem. 17(17), 2630–2639 (2016). [CrossRef]
22. L. Zhang, Y. Guo, R. Hao, Y. Shi, H. You, H. Nan, Y. Dai, D. Liu, D. Lei, and J. Fang, “Ultra-rapid and highly efficient enrichment of organic pollutants via magnetic mesoporous nanosponge for ultrasensitive nanosensors,” Nat. Commun. 12(1), 6849 (2021). [CrossRef]
23. H. Wang, Y. Liu, G. Rao, Y. Wang, X. Du, A. Hu, Y. Hu, C. Gong, X. Wang, and J. Xiong, “Coupling enhancement mechanisms, materials, and strategies for surface-enhanced Raman scattering devices,” Analyst 146(16), 5008–5032 (2021). [CrossRef]
24. K. Zhang, Y. Xiang, X. Wu, L. Feng, W. He, J. Liu, W. Zhou, and S. Xie, “Enhanced optical responses of Au@Pd core/shell nanobars,” Langmuir 25(2), 1162–1168 (2009). [CrossRef]
25. S. Sun, S. Yang, X. Hu, C. Zheng, H. Song, L. Wang, Z. Shen, and Z.-S. Wu, “Combination of Immunomagnetic Separation with Aptamer-Mediated Double Rolling Circle Amplification for Highly Sensitive Circulating Tumor Cell Detection,” ACS Sens. 5(12), 3870–3878 (2020). [CrossRef]
26. C. Xue, L. Wang, H. Huang, R. Wang, P. Yuan, and Z.-S. Wu, “Stimuli-Induced Upgrade of Nuclease-Resistant DNA Nanostructure Composed of a Single Molecular Beacon for Detecting Mutant Genes,” ACS Sens. 6(11), 4029–4037 (2021). [CrossRef]
27. M. Mohammadniaei, A. Koyappayil, Y. Sun, J. Min, and M.-H. Lee, “Gold nanoparticle/MXene for multiple and sensitive detection of oncomiRs based on synergetic signal amplification,” Biosens. Bioelectron. 159, 112208 (2020). [CrossRef]
28. R. Li, L. Zou, Y. Luo, M. Zhang, and L. Ling, “Ultrasensitive colorimetric detection of circulating tumor DNA using hybridization chain reaction and the pivot of triplex DNA,” Sci. Rep. 7(1), 44212 (2017). [CrossRef]
29. Y. Wu, C. Fu, W. Shi, and J. Chen, “Recent advances in catalytic hairpin assembly signal amplification-based sensing strategies for microRNA detection,” Talanta 235, 122735 (2021). [CrossRef]
30. J. Wu, Y. Tian, L. He, J. Zhang, Z. Huang, Z. Luo, and Y. Duan, “An efficient localized catalytic hairpin assembly-based DNA nanomachine for miRNA-21 imaging in living cells,” Analyst 146(9), 3041–3051 (2021). [CrossRef]
31. Y. Fan, Y. Liu, Q. Zhou, H. Du, X. Zhao, F. Ye, and H. Zhao, “Catalytic hairpin assembly indirectly covalent on FeO@C nanoparticles with signal amplification for intracellular detection of miRNA,” Talanta 223(Pt 1), 121675 (2021). [CrossRef]
32. L. Cui, J. Zhou, X.-Y. Yang, J. Dong, X. Wang, and C.-Y. Zhang, “Catalytic hairpin assembly-based electrochemical biosensor with tandem signal amplification for sensitive microRNA assay,” Chem. Commun. 56(70), 10191–10194 (2020). [CrossRef]
33. L. Wang, T. Meng, D. Zhao, H. Jia, S. An, X. Yang, H. Wang, and Y. Zhang, “An enzyme-free electrochemical biosensor based on well monodisperse Au nanorods for ultra-sensitive detection of telomerase activity,” Biosens. Bioelectron. 148, 111834 (2020). [CrossRef]
34. R. A. Alvarez-Puebla, R. Contreras-Cáceres, I. Pastoriza-Santos, J. Pérez-Juste, and L. M. Liz-Marzán, “Au@pNIPAM colloids as molecular traps for surface-enhanced, spectroscopic, ultra-sensitive analysis,” Angew. Chem. Int. Ed. 48(1), 138–143 (2009). [CrossRef]
35. D. Li, L. Jiang, J. A. Piper, I. S. Maksymov, A. D. Greentree, E. Wang, and Y. Wang, “Sensitive and Multiplexed SERS Nanotags for the Detection of Cytokines Secreted by Lymphoma,” ACS Sens. 4(9), 2507–2514 (2019). [CrossRef]
36. M. Xiao, K. Xie, X. Dong, L. Wang, C. Huang, F. Xu, W. Xiao, M. Jin, B. Huang, and Y. Tang, “Ultrasensitive detection of avian influenza A (H7N9) virus using surface-enhanced Raman scattering-based lateral flow immunoassay strips,” Anal. Chim. Acta 1053, 139–147 (2019). [CrossRef]
37. J. Ge, Y. Hu, R. Deng, Z. Li, K. Zhang, M. Shi, D. Yang, R. Cai, and W. Tan, “Highly Sensitive MicroRNA Detection by Coupling Nicking-Enhanced Rolling Circle Amplification with MoS Quantum Dots,” Anal. Chem. 92(19), 13588–13594 (2020). [CrossRef]
38. Z. Chen, Y. Xie, W. Huang, C. Qin, A. Yu, and G. Lai, “Exonuclease-assisted target recycling for ultrasensitive electrochemical detection of microRNA at vertically aligned carbon nanotubes,” Nanoscale 11(23), 11262–11269 (2019). [CrossRef]
39. W. Zheng, L. Yao, J. Teng, C. Yan, P. Qin, G. Liu, and W. Chen, “Lateral Flow Test for Visual Detection of Multiple MicroRNAs,” Sensors and Actuators B: Chemical 264, 320–326 (2018). [CrossRef]
40. S. Ye, M. Wang, Z. Wang, N. Zhang, and X. Luo, “A DNA-linker-DNA bifunctional probe for simultaneous SERS detection of miRNAs via symmetric signal amplification,” Chem. Commun. 54(56), 7786–7789 (2018). [CrossRef]
