key: cord-0996666-bqrpqlun
authors: Lee, Seul-Lee; Kim, Jihoon; Choi, Sungwook; Han, Jinsil; Seo, Giwan; Lee, Yong Wook
title: Fiber-optic label-free biosensor for SARS‐CoV‐2 spike protein detection using biofunctionalized long-period fiber grating
date: 2021-08-13
journal: Talanta
DOI: 10.1016/j.talanta.2021.122801
sha: d783b4bb550e81aed300ca30a46abf81d2ba9bda
doc_id: 996666
cord_uid: bqrpqlun

With COVID-19 widespread worldwide, people are still struggling to develop faster and more accurate diagnostic methods. Here we demonstrated the label-free detection of SARS-CoV-2 spike protein by employing a SARS-CoV-2 spike antibody-conjugated phase-shifted long-period fiber grating (PS-LPFG) inscribed with a CO(2) laser. At a specific cladding mode, the wavelength separation (λ(D)) between the two split dips of a PS-LPFG varies with the external refractive index, although it is virtually insensitive to ambient temperature variations. To detect SARS-CoV-2 spike protein, SARS-CoV-2 spike antibodies were immobilized on the fiber surface of the fabricated PS-LPFG functionalized through chemical modification. When exposed to SARS-CoV-2 spike protein with different concentrations, the antibody-immobilized PS-LPFG exhibited the variation of λ(D) according to the protein concentration, which was caused by bioaffinity binding-induced local changes in the refractive index at its surface. In particular, we also confirmed the potential of our sensor for clinical application by detecting SARS-CoV-2 spike protein in virus transport medium. Moreover, our sensor could distinguish SARS-CoV-2 spike protein from those of MERS-CoV and offer efficient properties such as reusability and storage stability. Hence, we have successfully fabricated a promising optical transducer for the detection of SARS-CoV-2 spike protein, which can be unperturbed by external temperature disturbances.

The recent outbreak of novel coronavirus disease 2019 , caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has globally threatened human health. What was worse, it is rapidly spreading across the globe from Asia to other continents because of its rapid human-to-human transmission [1] . As of March 04, 2021, more than 114,428,211 cases of COVID-19 have been confirmed around the world, resulting in 2,543,755 deaths [2] . Since no specific drugs for COVID-19 are yet available, early diagnosis is still critical for controlling the outbreak. Thus, highly sensitive diagnostic methods capable of rapid and direct detection of SARS-CoV-2 are required to obtain worthy information that can be provided to medical teams in the treatment of COVID-19. There are ongoing efforts to develop diagnostic platforms integrated with current biosensing protocols for rapid and selective detection of the virus. As one of the promising alternatives to conventional diagnostic methods, biosensor technology has been developed and reported for the determination of SARS-CoV-2 by employing various kinds of detection platforms. For example, reverse transcription-polymerase chain reaction (RT-PCR) is a predominant diagnostic method for COVID-19 but too time-consuming to be used for on-site detection [3] . Besides, assays based on reverse transcription loop-mediated isothermal amplification (RT-LAMP) were reported for rapid detection of RNA samples extracted from SARS-CoV-2 in about 30 min [4] . More recently, a dual-functional plasmonic biosensor was reported to detect SARS-CoV-2 with a low limit of detection (LOD) of 0.22 pM and distinguish genes of SARS-CoV from ones of SARS-COV-2 [5] . These methods are highly accurate and sensitive diagnostic technologies to detect SARS-CoV-2 but depend on the DNA detection-based method with disadvantages of the complex extraction and denaturation of viral RNA or DNA [6] . By incorporating a detection method based on antigen-antibody reaction, a graphene-based J o u r n a l P r e -p r o o f 5 recognition elements be covalently bonded onto the surface. The biofunctionalized LPFG offers the feasibility of an LPFG-based biosensor that can detect a minute amount of analytes with a high sensitivity to the SMRI variation induced by biological activity. In brief, the interaction between the biological recognition element and the analyte on the surface of the biofunctionalized LPFG can be detected directly through observation of its resonance wavelength shift, and such a detection approach, referred to as a label-free detection, can provide the possibility of the real-time measurement associated with biological activities without suffering from the inconvenience of laborious labeling processes. To date, numerous LPFGbased biosensors have already been proposed to prove that an LPFG is a useful apparatus for the label-free detection of various types of bioanalytes. For example, a graphene oxide (GO)-coated dual-peak LPFG was presented for the detection of anti-IgG with a LOD of 7 ng/mL by analyzing quantitative variations of the wavelength separation between the dual peaks, which occur by SMRI changes induced by bioaffinity binding between the immobilized IgG and the target anti-IgG at the GO-coated interface [14] . For the specific detection of glucose, a glucose oxidase (GOD)-immobilized LPFG was implemented and showed the linear correlation between the wavelength shift of its transmission spectrum and the glucose concentration in the range of 0.1-3 mg/mL [15] . By harnessing the specific binding between bacteria and bacteriophage, Escherichia coli of 10 3 CFU/mL was detectable with a bacteriophage-immobilized LPFG [16] , and T7 bacteriophage detection using an LPFG-based immunosensor was also achieved with a LOD of 5×10 3 PFU/mL [17] . Previous works show that LPFG-based optical biosensors can detect various types of biological targets in a label-free manner through appropriate surface biofunctionalization. However, despite the urgent need for the rapid and sensitive immunological J o u r n a l P r e -p r o o f 6 diagnosis of COVID-19, any study on LPFG-based biosensors for the detection of SARS-CoV-2 spike protein has not been reported yet.

Here we demonstrate label-free detection of SARS-CoV-2 spike protein by employing a SARS-CoV-2 spike antibody (SSA)-conjugated phase-shifted LPFG (PS-LPFG) inscribed by CO2 laser pulses as a biosensor head. The SSA-immobilized PS-LPFG serves as a label-free optical transducer to directly detect SARS-CoV-2 spike protein whose binding with the SSAs induces local changes in the refractive index of the bio-interaction layer. For the fabrication of a PS-LPFG, a  phase shift was introduced in the middle of an LPFG with periodical index modulation along the core of single-mode fiber (SMF). This  phase shift can convert destructive interference into constructive interference at the PMC, resulting in a splitting of a wavelengthdependent loss dip at the resonance wavelength. For two split dips with a specific cladding mode, their wavelength separation (λD) varies with the SMRI, although it remains virtually constant for ambient temperature variations owing to the grating formation based on CO2 laser irradiation [18] [19] [20] . To achieve high selectivity to SARS-CoV-2 spike protein in the sensor head, the surface of the fabricated PS-LPFG was functionalized with the immobilization of antibodies against SARS-CoV-2 spike protein on its unjacketed fiber surface, and its antibody binding capability could be enhanced through premodification of the fiber surface with (3-Aminopropyl) triethoxysilane (APTES) and glutaraldehyde (GA), an efficient interface coupling agent used as a probe linker. When the PS-LPFG conjugated with SSAs was exposed to SARS-CoV-2 spike protein, it could be observed that λD varied with increasing antigen concentration owing to bioaffinity binding-induced changes in the refractive index at its surface. Hence, the SSAconjugated PS-LPFG can act as a temperature-insensitive optical transducer for the detection of SARS-CoV-2 spike protein through quantitative analysis of λD as a function of the binding event J o u r n a l P r e -p r o o f between antigen and antibody at the fiber surface. In particular, we confirmed the potential of our sensor for clinical application by detecting SARS-CoV-2 spike protein in the virus transport medium used to collect and transport clinical virus specimens. Moreover, our sensor could distinguish SARS-CoV-2 spike protein from those of MERS-CoV and also provide the reusability and storage stability of the sensor head. Furthermore, the use of a CO2 laser for the grating writing can considerably reduce the fabrication cost of PS-LPFGs compared with UV grating inscription, which enables the potential supply of cost-effective sensor heads. This study is the demonstration of an optical transducer based on the integration of SSAs with a PS-LPFG, enabling the sensitive and selective detection of SARS-CoV-2 spike protein in a clinical virus transport medium as well as a standard buffer while minimizing the effect of environmental temperature perturbations. 

A PS-LPFG is composed of two identical sections of an LPFG with a grating length and grating period of L and , respectively, between which an unperturbed region of a fiber of LPS in length is inserted. The intensity transmittance of the PS-LPFG can be derived based on the coupled-mode theory with the help of the fundamental matrix method [12] . When the beam propagates through the PS-LPFG whose length is 2L + LPS, the intensity transmittance I of the PS-LPFG can be depicted by

cos 2 tan tan , co cl m co cl m co cl m PS

where Ico and Icl,m are the normalized intensities of the core and the mth cladding modes, respectively, and  is the phase mismatch given by δ = (co − cl,m − 2π/)/2. co (= 2nco/) and

cl,m (= 2ncl,m/) are the propagation constants of the core and the mth cladding modes, respectively. nco and ncl,m are the refractive indices of the core and the mth cladding modes, respectively, and  is the free-space wavelength. κ is the coupling coefficient given by 

which indicates that the transmission spectrum of the -PS-LPFG has two attenuation bands symmetrically split about the resonance wavelength res. Then, the PS-LPFG was rinsed successively with ethanol and DI water several times to completely remove any residues of compounds, followed by thermal annealing for 1 h at 120 °C.

After the formation of the amine-functionalized surface, the PS-LPFG was subsequently treated J o u r n a l P r e -p r o o f with 2% (v/v) GA solution for 3 h at room temperature to activate glutaraldehyde cross-linking.

The aldehyde groups (-CHO) of GA, formed through this GA treatment, can be bound with the -NH2 groups of SSA. Finally, the functionalized PS-LPFG, ready for amine cross-linking, was exposed to 100 g/mL SSA for 3 h at room temperature.

To confirm the efficiency of the surface functionalization for SSA conjugation, NH2- sensor head, but in this study these microbubbles were detached from the surface of the sensor head by simply applying light impact to the sample container using a rubber stick.

For the performance evaluation of the fabricated sensor, we investigated its response to antigen protein. Figure 4 Fig. 4(e) , respectively, did not exhibit any remarkable changes for various sample concentrations. The control experiment indicates that SSA is indispensable for specific binding with SARS-CoV-2 spike protein. Thus, in light of the trend of these two responses, our approach exhibits sufficient signal-on/off detection performance, which indicates the "signal-on" at the treated PS-LPFG and the "signal-off" at the untreated one. Regarding the "signal-on" operation, although overlaps between the error bars are responsible for the low measurement resolution, it is possible to differentiate between SARS-CoV-2 spike protein concentrations that differ by more than 10 3 times for sample concentrations less than 10 6 pg/mL.

Moreover, to investigate the cross-reactivity of SSA immobilized on the PS-LPFG, we compared its sensitivity to SARS-CoV-2 and MERS-CoV spike proteins. Figure 4 (f) shows the ΔD,det variations of the sensor when the concentrations of SARS-CoV-2 and MERS-CoV spike proteins increase from 1 to 10 8 pg/mL, indicated by orange circles and green squares, respectively. As can be confirmed from the figure, no noticeable variations were observed in terms of MERS-CoV spike antigen proteins, suggesting that our SSA-conjugated sensor head is highly specific to SARS-CoV-2 spike proteins.

In general, diagnosis of COVID-19 in the clinic is performed using an FA transport medium LPFG without SSA conjugation. The experimental results revealed that the sensor could detect SARS-CoV-2 spike protein at a concentration of 100 pg/mL (as confirmed in Fig. 5(a) and Supplementary Fig. S2 ). This corroborates that our PS-LPFG sensor can detect antigens in clinical samples without any preprocessing.

In particular, to assess the reusability of the sensor head, repetitive measurements of SARS-CoV-2 spike protein was done using the same sensor head (i.e., SSA-conjugated PS-LPFG) at irregular temporal intervals. As indicated by red circles in Fig. 5(b) , the tests were performed by five times for two weeks (14 days) . For every test, the sensor head was exposed to SARS-CoV-2 spike protein with a concentration of 1 μg/mL, and the normalized value of ΔD,det, referred to as ΔD,det,N, was observed as a sensor signal. Instantly after completing the measurement of the sensor signals, the sensor head was washed with 1 × PBS to remove residue on the sensor surface, which is irrelevant to antibodies and can cause non-specific reactions, and stored at 4 °C in the refrigerator. At the second test (3 days after the first test), the magnitude of the ΔD,det,N signal was maintained at 90% or more. At the third test (6 days after the first test), the magnitude of the signal was maintained at 60%. So, the measurement reliability might be dropped from the third test. Finally, it was decreased to 30% or less at the fifth test (14 days after the first test).

The decrease in the signal magnitude seems to be caused by the destruction or detachment of SSAs weakly bound to the surface of the sensor head during the washing step. Moreover, the storage stability of the sensor head was also appraised by measuring the sensor signal when exposed to SARS-CoV-2 spike protein two weeks after the initial measurement. Similarly to the J o u r n a l P r e -p r o o f case of the reusability assessment, the sensor head was washed with 1 × PBS and stored at 4 °C in the refrigerator, instantly after the initial signal measurement of SARS-CoV-2 spike protein.

Then, the next signal measurement was done after two weeks. The measured sensor signals are displayed as blue squares in Fig. 5(b) . As can be seen from the figure, our sensor could offer the signal magnitude kept at 70% or more. This indicates that our sensor can be reused at least one more time within 14 days after the first test, exhibiting reasonable storage stability for a certain period and reliable sensing performance.

Furthermore, to check whether the sensor performance is affected by ambient temperature changes, we investigated the temperature dependence of the SSA-conjugated PS-LPFG regarding two aspects: (1) Whether antigen-antibody interaction during the sensing operation creates temperature changes on the surface of the sensor head and (2) how the sensor signal is affected by external temperature disturbances. Since the ambient temperature changes can lead to the refractive index changes of the core and cladding of the sensor head as well as the thermal expansion of the buffer solution, it is important to control the ambient temperature and minimize thermally induced noises to accurately measure very small changes in the refractive index occurring in bioassays [22] . While antigen-antibody interaction occurred on the surface of the temperature. The actual temperature variation may be less than 0.1 °C because the temperature resolution of the thermometer is 0.1 °C. Next, for scrutinization of the external temperature response of the SSA-conjugated PS-LPFG, it was placed in a temperature chamber, and D of the dual-resonance dips in its transmission spectrum was monitored for the ambient temperature increase (red circles) and decrease (blue squares) in a range of 22 to 26 °C with a step of 0.5 °C.

As seen in Fig. 6(b) , the detuning parameter of D, i.e., ΔD,det, shows no specific temperature dependence. As described earlier, ΔλD,det was defined as 100  (ΔD/D,0), where D,0 was set as 91.84 nm because it was the value of D at the onset of the measurement. In the above temperature range, ΔD,det was measured as ~0.141 for a temperature change of 4 °C, and, in particular, ΔD,det below 0.0218 was observed in a temperature range of 2324 °C. It can be inferred from this result that, in the room-temperature antigen detection, ΔD,det due to temperature perturbations induced by bioaffinity binding is much less than 0.0109. Thus, it is concluded that a D deviation induced by ambient temperature changes rarely affects the measurement accuracy of our antigen detection experiments carried out at room temperature, because the ΔD,det value of ~0.0109, obtained by a D deviation of 0.01 nm, is close to the ΔD,det magnitude obtained in the bare PS-LPFG. For further implementation of a completely temperature-insensitive system, it is necessary to include a temperature control system based on thermo-electric cooling devices such as Peltier elements in our sensor system [23] .

In sum, we demonstrated a fiber-optic biosensing platform suited for the label-free detection of SARS-CoV-2 spike protein by employing an SSA-conjugated PS-LPFG as a sensing head to directly detect the bio reaction between bound SSAs and free SARS-CoV-2 spike protein, which J o u r n a l P r e -p r o o f induces local changes in the refractive index of a bio-interaction layer deposited on its surface through immobilization of SSAs. A PS-LPFG was fabricated by inserting a  phase shift in the middle of an LPFG inscribed with a CO2 laser, which results in a splitting of its resonance dip.

To achieve high selectivity to SARS-CoV-2 spike protein, its spike antibodies were immobilized on the unjacketed fiber surface of the fabricated PS-LPFG functionalized through chemical modification using APTES and GA. When the SSA-immobilized PS-LPFG was exposed to SARS-CoV-2 spike protein, λD varied with increasing antigen concentration owing to bioaffinity binding-induced changes in the refractive index at its surface. Through quantitative analysis of λD as a function of the binding event between antigen and antibody at the fiber surface, we could detect SARS-CoV-2 spike protein with the proposed sensor not only in the standard buffer but also in virus transport medium. Moreover, our sensor could discriminate SARS-CoV-2 spike protein from those of MERS-CoV and also offer the reusability and storage stability of the sensor head. Furthermore, we confirmed that the temperature-induced measurement uncertainty was negligible in the room-temperature antigen detection. As a result, the SSA-conjugated PS-LPFG can act as an optical transducer capable of sensitive and selective detection of SARS-

CoV-2 spike protein in a clinical virus transport medium as well as a standard buffer. Findings from this study highlight the potential of the SSA-immobilized PS-LPFG for the development of the immunological diagnosis of COVID-19, and we wish our cost-effective fiber-optic biosensor to be a step forward for the rapid extinction of the global pandemic. 

Covid-19 -Navigating the Uncharted

A simple magnetic nanoparticlesbased viral RNA extraction method for efficient detection of SARS-CoV-2, bioRxiv

Rapid detection of COVID-19 coronavirus using a reverse transcriptional loop-mediated isothermal amplification (RT-LAMP) diagnostic platform

Dual-functional plasmonic photothermal biosensors for highly accurate severe acute respiratory syndrome coronavirus 2 detection

Facile preparation of a DNA sensor for rapid herpes virus detection

Rapid detection of COVID-19 causative virus (SARS-CoV-2) in human nasopharyngeal swab specimens using fieldeffect transistor-based biosensor

Optical fibre gratings as tools for chemical and biochemical sensing

Long period grating working in transition mode as promising technological platform for label-free biosensing

Fiber grating sensors in medicine: Current and emerging applications

Biosensors employing ionic self-assembled multilayers adsorbed on long-period fiber gratings

Fiber grating spectra

Analysis of the response of long period fiber gratings to external index of refraction

Graphene oxide functionalized long period grating for ultrasensitive label-free immunosensing

Immobilization of enzyme on long period grating fibers for sensitive glucose detection

Long period grating based biosensor for the detection of Escherichia coli bacteria

Long-period fiber grating sensor for detection of viruses

Temperature-and strain-insensitive torsion sensor based on phase-shifted ultra-long-period grating

Fabrication of phase-shifted long-period fiber grating and its application to strain measurement

Resonance-temperature-insensitive phase-shifted long-period fiber gratings induced by surface deformation with anomalous strain characteristics

Long period fiber grating based temperature-compensated high performance sensor for bio-chemical sensing applications

Flow cell for strain-and temperature-compensated refractive index measurements by means of cascaded optical fibre long period and Bragg gratings

Long period grating in double cladding fiber coated with graphene oxide as high-performance optical platform for biosensing

Sensitive and selective detection of SARS-CoV-2 spike protein. (a) Superposed transmission spectra of SSA-conjugated PS-LPFG

Magnified plots of (a) near two attenuation dips with dip wavelengths of (b) 1 and (c) 2, shifting towards shorter and longer wavelength regions with increasing concentration of SARS-CoV-2 spike protein, respectively. (d) Variations of detuning parameters Δ1,det and Δ2,det of SSA-conjugated PS-LPFG, indicated by red circles and blue squares, respectively, and variations of Δ1,det and Δ2,det of bare (untreated) PS-LPFG, indicated by black triangles and gray inverted triangles, respectively, according to antigen protein concentration. (e) Variations of ΔD,det of SSA-conjugated and bare PS-LPFGs with respect to antigen protein concentration

ΔD,det variations of SSA-conjugated PS-LPFG according to concentrations of SARS-CoV-2 and MERS-CoV spike proteins, indicated by orange circles and green squares, respectively. All data points related with the SSA-conjugated PS-LPFG, shown in (d), (e) and (f), were nonlinearly fitted based on the Hill adsorption model with a correlation coefficient