Detection of SARS-CoV-2 and influenza a virus using colorimetric LFA strips

The N-proteins used in this work are suitable as target biomarkers because they are virus genomic RNA-binding proteins that exist largely inside the virus. Most of the commercial LFA strips also use the N-proteins as target biomarkers. Although N-proteins for SARS-CoV-2 and influenza A virus have the same name, they have different binding domains because their structures and amino acid sequences are entirely different [27, 28]. Since the N-proteins of SARS-CoV-2 and influenza A have low cross-reactivity with each other, we chose these proteins to distinguish the two viruses. Figure 1a and b show the detection results of SARS-CoV-2 and influenza A virus lysates using commercially available colorimetric LFA strips from the SD Biosensor. As shown in Fig. 1a, the N-protein target, extracted from SARS-CoV-2, and detection antibody-conjugated AuNPs were sequentially captured by N-protein capture antibodies immobilized on the test line. At this time, the color of the test line changed from colorless to red when SARS-CoV-2 was present. On the other hand, in the absence of SARS-CoV-2, there was no color change because immunocomplexes were not formed on the test line. When various concentrations of SARS-CoV-2 were measured with commercial LFA strips, the limit of detection (LoD) for SARS-CoV-2 was estimated to be approximately 500 plaque-forming unit (PFU)/mL (Fig. 1a). The assays were conducted similarly using commercial LFA strips for influenza A virus. The LFA tests were performed for various influenza A virus concentrations from 0 hemagglutinin unit (HAU)/mL to 8064 HAU/mL. As shown in Fig. 1b, no color change was observed when the concentration of influenza A virus was less than 2016 HAU/mL. Therefore, the LoD of the commercial LFA strip for influenza A was estimated to be approximately 2016 HAU/mL.

When self-diagnosing using commercial LFA strips, it is actually positive but negative in many cases because of the poor sensitivity of the diagnostic strip. In other words, commercial LFA strips appeared to be negative for SARS-CoV-2 in the 1–500 PFU/mL range and influenza A virus in the 1–2016 HAU/mL range, respectively, but they are actually positive. To solve this false-negative diagnostic problem of the commercial LFA strip, a new diagnostic technology that can dramatically improve the detection sensitivity of each virus is essentially required. Therefore, we expected that the dual-mode SERS-LFA strip would significantly reduce the false-negative diagnosis rate by enhancing the detection sensitivity of SARS-CoV-2 and influenza A virus, thereby accurately distinguishing these two respiratory viruses.

Fig. 1
figure 1

Detection results of SARS-CoV-2 and influenza A virus lysates using commercially available colorimetric LFA strips. a Assay results of SARS-CoV-2 lysates along with the increase of SARS-CoV-2 concentration in the 0–1000 PFU/mL range. N-protein in SARS-CoV-2 was used as a target antigen. b Assay results of influenza A virus lysates along with the increase of influenza A virus concentration in the 0–8064 HAU/mL range. N-protein in influenza A virus was used as a target antigen

Working principle of dual-mode SERS-LFA strips for simultaneous detection of SARS-CoV-2 and influenza a virus

Figure 2 shows the working principle of the dual-mode SERS-LFA strip for simultaneous detection of SARS-CoV-2 and influenza A virus. A target virus solution was prepared by spiking SARS-CoV-2 and influenza A virus lysates into a nasopharyngeal solution of normal individuals to make conditions similar to the clinical diagnosis process (Fig. 2a). The LFA strip consists of two test lines and one control line. N-protein antibodies for SARS-CoV-2 and influenza A were immobilized on test lines 1 and 2, respectively, and anti-mouse IgG antibodies were immobilized on the control line. We also prepared different concentrations of a target virus solution, detection antibody-conjugated SERS nanotags for SARS-CoV-2 and influenza A virus, and a running buffer solution in an ELISA 96-well plate. When the prepared LFA strip is immersed in each well, the solution in the well moves toward the absorbent pad by capillary force.

Figure 2b shows SERS nanotags for identifying SARS-CoV-2 and influenza A virus. After conjugation of MGITC Raman reporters and N-protein detection antibodies on AuNPs, 0.1% casein was used as a blocking agent to stabilize the remaining area of the AuNP surface. Additional file 1: Fig. S1 shows TEM images of (a) AuNPs and (b) SERS nanotags, and corresponding (c) DLS distributions and (d) UV–vis absorption spectra. TEM images show that AuNPs have uniform size distributions. Since antibodies or Raman reporter molecules attached to the surface of AuNPs do not appear in the TEM image, we confirmed their attachment through DLS distribution (Additional file 1: Fig. S1c) and UV–vis spectra (Additional file 1: Fig. S1d). According to the DLS distribution, the average diameter of AuNPs and SERS nanotags increased from 50 to 68 nm, and the UV–vis absorption maximum was red-shifted from 530 to 535 nm. Therefore, we could confirm that SERS nanotags were successfully synthesized from DLS and UV–vis spectral data. As shown in Fig. 2c, SERS nanotags and target N-proteins form sandwich immunocomplexes on test line 1 in the presence of SARS-CoV-2 lysates. On the other hand, sandwich immunocomplexes are formed on test line 2 when influenza A virus lysates are present. Additional file 1: Fig. S2 shows SEM images for each test line in the presence and absence of viruses. In both cases, it was observed that many nanoparticles were bound to each test line by the interaction between SERS nanotags and N-proteins when the viruses were present. SERS nanotags are always captured on the control line through antibody–antibody interactions regardless of the presence of viruses. After the flow through the LFA strip is finished, SERS signals were collected on test and control lines to evaluate SARS-CoV-2 and influenza A virus quantitatively.

Fig. 2
figure 2

Working principle of the dual-mode SERS-LFA strip for simultaneous detection of SARS-CoV-2 and influenza A virus. a Preparation of virus lysates in a nasopharyngeal solution of normal individuals. This virus solution was mixed with SERS nanotags and a running buffer solution in a 96-well plate. b N-protein antibody-conjugated SERS nanotags for the detection of SARS-CoV-2 and influenza A virus. c SERS nanotags and running buffer move toward the SERS-LFA strip by capillary force. Formation of sandwich immunocomplexes for SARS-CoV-2 (test line 1) and influenza A virus (test line 2)

Simultaneous detection of SARS-CoV-2 and influenza a virus using dual-mode SERS-LFA strips

In SERS-based assays, it is critical to secure the reproducibility that is formed from localized surface plasmon effects. Therefore, multiple SERS spectra were measured and averaged. As shown in Fig. 3a, pixel-to-pixel detections for an area of 2600 μm × 500 μm with a 100 μm interval were sequentially performed for test and control lines. Then the Raman signal intensities for 130 (26 × 5) pixels were baseline-subtracted to obtain reliable Raman signal intensity in Additional file 1: Fig. S3. In this way, we evaluated the analytical performance of the dual-mode SERS-LFA strip by changing the concentration of one virus type only. The color change of the SERS-LFA strip and corresponding SERS spectra of the test and control lines were measured when the concentration of SARS-CoV-2 lysate changed in the range of 0–1000 PFU/mL as shown in Fig. 3b. According to the photographic images of the SERS-LFA strips, we can see a certain intensity of red color, indicating that the control line is correctly operating. On the other hand, test line 1 for SARS-CoV-2 showed an apparent color change at a concentration of 500 PFU/mL and more, but a color change at the concentration lower than that was difficult to observe. Test line 2 for influenza A virus showed no color change at this time.

In the case of influenza A virus, on the contrary, test line 1 for SARS-CoV-2 showed no color change at all, and test line 2 showed an evident color change at a concentration of 1008 HAU/mL and more (Fig. 3c). These experimental results show that the dual-mode SERS-LFA strip has a specific selectivity for SARS-CoV-2 and influenza A virus. Nevertheless, naked-eye identification has a false-negative diagnostic problem for an infected person with a low concentration of SARS-CoV-2 or influenza A virus. Virus assays using dual-mode SERS-LFAs were performed to reduce this false-diagnostic rate. Raman spectra were averaged after measuring 130 pixels for test and control lines (Fig. 3b and c). In both SARS-CoV-2 and influenza A virus cases, the SERS intensity decreases consistently as the virus concentration decreases. On the other hand, the control lines’ SERS intensities were constant in both cases. Our experimental results indicate that the SERS-LFA strips can simultaneously detect SARS-CoV-2 and influenza A virus with high sensitivity.

Fig. 3
figure 3

a Pixel-to-pixel SERS detections for an area of 2600 μm × 500 μm with a 100 μm interval for test and control lines. Raman signal intensities for 130 spots were averaged to obtain reliable Raman signal intensity. b Visual color changes of the SERS-LFA strip and corresponding SERS spectra of the test and control lines when the concentration of SARS-CoV-2 lysate changed in the range of 0–1000 PFU/mL. c Visual color changes of the SERS-LFA strip and corresponding SERS spectra of the test and control lines when the concentration of influenza A virus lysate changed in the range of 0–8064 HAU/mL

Performance evaluation of dual-mode SERS-LFA strips for detection of SARS-CoV-2 and influenza a virus

The calibration curves for SARS-CoV-2 and influenza A virus, determined by the measurement results of ELISA and SERS-LFA, are compared in Fig. 4. The characteristic Raman peak intensity at 1615 cm− 1 of MGITC was used for quantitative analysis of both viruses. Each calibration curve was determined using a four-parameter sigmoidal fitting equation. The y-axis represents the optical density for ELISA and the scattering intensity ratio of the test line and control line for SERS-LFA. In both ELISA and SERS-LFA, virus concentration and optical signal intensity showed a good correlation. For SARS-CoV-2 and influenza A virus, each LoD was determined from calibration curve data measured by ELISA and SERS-LFA.

Fig. 4
figure 4

Comparison of calibration curves for a SARS-CoV-2 and b influenza A virus, determined by the measurement results of ELISA and dual-mode SERS-LFA. Each calibration curve was determined using a four-parameter sigmoidal fitting equation (black: ELISA, purple and red: SERS-LFA). The y-axis represents the optical density for ELISA and the Raman scattering intensity ratio of the test and control lines for SERS-LFA

In Fig. 5a and b, we compared LoD values ​​for the commercial colorimetric strip, ELISA, and dual-mode SERS-LFA strip for SARS-CoV-2 and influenza A virus. The intensity values were normalized based on the highest concentration of each virus. In the case of SARS-CoV-2, LoDs were determined to be 48 PFU/mL for ELISA and 5.2 PFU/mL for SERS-LFA strip, respectively. The SERS-LFA strip exhibits approximately 10 times better sensitivity than ELISA. For the colorimetric LFA strip, LoD was 500 PFU/mL (Fig. 1a). In the case of influenza A virus, LoDs were estimated to be 880 HAU/mL for ELISA and 23 HAU/mL for SERS-LFA strip, proving that the SERS-LFA strip has approximately 40 times better sensitivity than ELISA. The LoD of the colorimetric LFA strip was 1008 HAU/mL (Fig. 1b). Overall, the dual-mode SERS-LFA strip has a good selectivity against SARS-CoV-2 and influenza A virus. It is also more sensitive than the colorimetric LFA strip and ELISA assay currently used for immunoassays of these viruses. Furthermore, using the SERS-LFA strip, it is possible to distinguish between SARS-CoV-2 and influenza A virus.

Figure 5c and d show the cross-reactivity test results of the dual-mode SERS-LFA strip against SARS-CoV-2 and influenza A virus. SERS intensities of test lines 1 and 2 were measured when the concentration of SARS-CoV-2 was changed in the range of 50−1000 PFU/mL, but the concentration of influenza A virus was fixed at 8064 HAU/mL (Fig. 5c). The SERS intensity for SARS-CoV-2 in test line 1 increased consistently as the concentration increased in the 50−1000 PFU/mL range while the SERS intensity for influenza A virus was kept constant at 8064 HAU/mL (Fig. 5c). Conversely, when the concentration of influenza A virus was changed in the range of 168–8064 HAU/mL, but the concentration of SARS-CoV-2 was kept constant at 200 PFU/mL, only the SERS intensity of influenza A virus was changed along with its concentration (Fig. 5d). These experimental results show that the SARS-CoV-2/influenza A virus dual-mode LFA strip developed in this study has good selectivity for each virus. Additionally, a selectivity test for five different respiratory viruses was performed using a dual-mode SERS-LFA strip. As shown in Additional file 1: Fig. S4, our dual-mode SERS-LFA strip showed good selectivity for only influenza A virus (H1N1 and H3N2 types) and SARS-CoV-2 among the five viruses.

Fig. 5
figure 5

Cross-reactivity test results of the dual-mode SERS-LFA strip against SARS-CoV-2 and influenza A virus. Normalized intensity variations of SERS-LFA (Raman peak intensity at 1615 cm− 1), ELISA (absorbance intensity), and colorimetric LFA (phase contrast intensity) for a SARS-CoV-2 (0–1000 PFU/mL) and b influenza A virus (0–8064 HAU/mL). c SERS intensity ratio variations of test lines 1 and 2 when the SARS-CoV-2 concentration was changed in the range of 50–1000 PFU/mL, but the concentration of influenza A virus was fixed at 8064 HAU/mL. d SERS intensity ratio variations of test lines 1 and 2 when the influenza A virus concentration was changed in the 168–8064 HAU/mL range, but the concentration of SARS-CoV-2 was fixed at 200 PFU/mL

Clinical validation of dual-mode SERS-LFA strips for detection of SARS-CoV-2 and influenza a virus

Table 1 shows clinical assay results using dual-mode SERS-LFA strips for 39 patient samples (28 SARS-CoV-2 positives, 6 influenza A virus positives, and 5 negatives). RT-PCR results for ORF1 of SARS-CoV-2 were used as control data [29, 30]. SARS-CoV-2 positive samples were classified into four levels (< 20, 20–25, 25–30, and > 30) according to the Ct levels. It is known that clinical samples in Ct < 20 have a high virus concentration. On the other hand, clinical samples in Ct > 30 need more thermo-cycling amplification steps, which means that the virus concentration is relatively low. We compared the assay results tested by the dual-mode SERS-LFA strips with those tested by commercial SARS-CoV-2 and influenza A virus LFA strips to evaluate the diagnostic efficacy of the proposed dual-mode SERS-LFA strips. For SARS-CoV-2, both LFA and SERS-LFA strips showed positive results for one clinical sample in the Ct < 20 levels. For the seven samples in the 20 < Ct < 25 levels, six (85%) and seven (100%) samples showed positive for LFA and SERS-LFA, respectively. For the nine samples in the 25 < Ct < 30 levels, six (67%) and eight (88%) clinical samples showed positive for LFA and SERS-LFA, respectively. Finally, for the 11 samples in the Ct > 30 levels, one (9%) and six (60%) showed positive for LFA and SERS-LFA, respectively. In the case of influenza A virus, four out of six clinical samples showed positive (66%) for LFA, but all the six samples showed positive (100%) for SERS-LFA. Both LFA and SERS-LFA showed all negative (100%) for five negative clinical samples.

Additional file 1: Table S1 shows the SERS intensity ratios (TL1/CL) for eight different SARS-CoV-2 concentrations. The corresponding calibration curve (Fig. 4a) was determined by the four-parameter sigmoidal fitting equation in Additional file 1: Table S3. Here, the LoD of TL1/CL for SARS-CoV-2 was determined to be 0.062, and the corresponding concentration was 5.2 PFU/mL. Therefore, if the TL1/CL value measured for the clinical sample was greater than 0.062, it was determined as positive, and if it was less than 0.062, it was determined as negative. Positive/negative discrimination was performed on 33 SARS-CoV-2 samples, and the results are listed in Table 1. Positive/negative discrimination for influenza A virus was also performed similarly to SARS-CoV-2. Additional file 1: Table S2 shows the SERS intensity ratios (TL2/CL) for eight different influenza A virus concentrations. The corresponding calibration curve (Fig. 4b) was determined by the four-parameter sigmoidal fitting equation in Additional file 1: Table S3. The LoD of TL2/CL for influenza A virus was determined to be around 0.048, and the corresponding concentration was 23 HAU/mL. Positive/negative discrimination was performed on 11 influenza virus A samples, and the results are listed in Table 1.

Table 1 Clinical assay results for SARS-CoV-2 and influenza A virus using RT-PCR, commercial LFA, and dual-mode SERS-LFA strips performed on 39 patient samples

Table 2 summarizes the positive/negative discrimination results based on the SARS-CoV-2/influenza A virus assays performed by commercial LFA and our dual-mode SERS-LFA strips on 39 clinical samples. Overall, in the assays using dual-mode SERS-LFA strips, the false-negative rate was significantly reduced compared to commercial LFA strips. In particular, the false-negative rate was reduced considerably in clinical samples with Ct > 25, which is due to the high sensitivity of the SERS-LFA strip. Although a sufficient number of clinical samples could not be tested for influenza A virus, the false-negative rate was also significantly reduced, like SARS-CoV-2 when dual-mode SERS-LFA strips were used for the clinical assays. In conclusion, when using the dual-mode SERS-LFA assay strips developed in this study, the false-negative rate was remarkably reduced compared to individual SARS-CoV-2 or influenza A virus LFA strips.

Table 2 Statistical analysis of the diagnostic results of SARS-CoV-2 and influenza A virus using LFA and dual-mode SERS-LFA strips on 39 clinical samples

Rights and permissions

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://creativecommons.org/licenses/by/4.0/.

Disclaimer:

This article is autogenerated using RSS feeds and has not been created or edited by OA JF.

Click here for Source link (https://www.springeropen.com/)