key: cord-0220755-da07qocp authors: Miyamura, Shogo; Oe, Ryo; Nakahara, Takuya; Okada, Shota; Kajisa, Taira; Taue, Shuji; Tokizane, Yu; Minamikawa, Takeo; Yano, Taka-aki; Otsuka, Kunihiro; Sakane, Ayuko; Sasaki, Takuya; Yasutomo, Koji; Yasui, Takeshi title: Rapid detection of SARS-CoV-2 nucleocapsid protein using dual-comb biosensing date: 2022-04-25 journal: nan DOI: nan sha: c44e0485004a14a89ad011abbf8ac0e19a5edb5e doc_id: 220755 cord_uid: da07qocp Testing of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is essential along with vaccination and inactivation to fight against the coronavirus disease 2019 (COVID-19) pandemic. Reverse-transcription polymerase chain reaction (RT-PCR), based on reverse transcription of RNA into DNA and amplification of specific DNA targets, is the current standard for COVID-19 testing; however, it hampers from laborious and time-consuming multiple steps. If the testing is largely simplified and shortened, it will be a powerful deterrent to the spread of COVID-19. Here we demonstrate the optical biosensing based on optical frequency comb (OFC), enabling the rapid detection of SARS-CoV-2 nucleocapsid protein. The virus-concentration-dependent optical spectral shift caused by antigen-antibody interaction and multimode-interference fiber sensor is transformed into a photonic radio-frequency (RF) shift by coherent frequency link between optical and RF regions in OFC, benefiting from high precision, rapid, simple, and low cost in electric frequency measurements. Furthermore, the active-dummy compensation of temperature drift with dual-comb configuration extracts the imperceptible change of virus-concentration-dependent signal from the large background signal that changes moment by moment. Such the dual-comb biosensing has a potential to reduce the testing time down of COVID-19 to a few tens minute, which is one order of magnitude shorter than that of RT-PCR (typically, 5 hours). Furthermore, it will be applied for sensitive sensing of not only virus of emerging and re-emerging infectious diseases but also RNA, bio-marker, and endocrine disruptor by selecting the surface modification of biomolecule interaction. The coronavirus disease 2019 , caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has rapidly spread in no time and is still going around all over the world. One reason not to suppress the rapid spread is in time-consuming testing of COVID-19. The current standard for COVID-19 testing is reverse transcription polymerase chain reaction (RT-PCR) [1] [2] [3] , benefiting from sensitive detection of SARS-CoV-2 by selective amplification of nucleic acid sequence of target gene. However, it hampers from time-consuming multiple steps which involve purification, nucleic acid amplification, and fluorescence detection. To avoid overwhelming hospitals, there is a considerable need for rapid detection of SARS-CoV-2. Optical biosensor [4, 5] is a potential method for rapid detection of SARS-CoV-2 due to their sensitivity, cost-effectiveness, versatility, ease of testing, point-ofcare (POC) prospects, and more importantly no need for nucleic acid amplification. For example, surface plasmon resonance (SPR) [6, 7] has been widely used for analyzing bio-molecules and virus, in which the spectral shift of SPR dip is measured in wavelength or angular spectrum. Due to high quality kinetics, real-time data acquisition, label-free analysis, faster automated experiments, and lower sample consumption, SPR has been applied for the detection of human immunodeficiency virus [8] , Ebola virus [9] , norovirus [10] , influenza virus [11] , and even SARS-CoV-2 [12] [13] [14] . However, the relatively broad SPR dip and/or the instrumentational resolution -6 -often spoil the precision of virus detection, hampering from the further enhancement of sensing performance. If the virus-concentration-dependent optical spectral shift is transformed into a photonic radio-frequency (RF) signal in combination with a sharpened spectrum, the optical biosensing benefits from high precision, high functionality, convenience, and low cost by making use of frequency standards and precise measurement apparatuses in the electric RF region. Recently, an optical frequency comb (OFC) [15] [16] [17] [18] has attracted attention for a photonic RF sensor via coherent frequency link between optical and RF regions [19, 20] . For example, a refractive-index-dependent optical spectrum shift was converted into a change of an OFC mode spacing frep around several tens MHz by placing a multimode-interference (MMI) refractive-index (RI) fiber sensor [21, 22] inside a fiber OFC cavity [23] [24] [25] . Due to the ultra-narrow linewidth of the frep signal, this RI-sensing OFC enables us to precisely measure the sample RI with a resolution of 4.9×10 -6 refractive index units (RIU). The RI-sensing OFC has a potential to be further extended for optical biosensing, namely biosensing OFC, by surface modification of biomolecule interaction in the MMI fiber sensor similar to SPR. However, the residual temperature drift of frep signal hampers from high precision in the RI-sensing OFC and its extension to the biosensing OFC. In this article, we first suppress the temperature drift of frep signal by using dual-comb configuration of an active-sensing OFC and a dummy-sensing OFC in the same manner as the active-dummy temperature compensation of strain sensors [26] . Then, for proof of -7 -concept, we apply the active-dummy dual RI-sensing OFCs for RI sensing of a glycerol solution. Finally, we demonstrate the rapid detection of SARS-CoV-2 nucleocapsid protein (N protein) antigen by combining the active-dummy dual RI-sensing OFCs with the surface modification of SARS-CoV-2 N protein antibody, namely dual-comb biosensing. Figure 1a shows the principle of operation for the biosensing OFC. Function of the biosensing OFC is implemented by three steps: (1) antigen-antibody interaction in the antibody-modified sensor surface, (2) RI-dependent tunable bandpass filtering in MMI fiber sensor [21, 22, 27] , and (3) photonic-to-RF conversion via wavelength dispersion of a cavity fiber [23] [24] [25] . In (1), the selective combination of a target antigen with the antibody changes the effective RI on the sensor surface depending on the antigen concentration. In (2) , since the MMI fiber sensor functions as an RI-dependent tunable optical bandpass filter (bandpass center wavelength = lMMI) via the multimode interference and the Goos-Hänchen shift, the OFC shows RI-dependent and hence antigen-concentration-dependent shift of optical spectrum. In (3), the antigenconcentration-dependent shift of optical spectrum is converted into that of optical cavity length nL, where n is RI of the cavity fiber and L is the physical length of OFC cavity, via the wavelength dispersion of the cavity fiber. Finally, a change in the antigen Mechanical sharing dual-comb configuration [28] of an active biosensing OFC (center optical wavelength = 1550 nm, repetition frequency = frep1 = 29.38 MHz) and a dummy biosensing OFC (center optical wavelength = 1550 nm, repetition frequency = frep2 = 29.47 MHz) was adopted for the active-dummy compensation of the temperature drift. In this configuration, although frep1 and frep2 fluctuate depending on the cavity temperature via thermal expansion or shrink of nL, their fluctuations were common to each other due to the same thermal disturbance. Therefore, frequency difference between frep1 and frep2 (= ∆frep = frep1 -frep2 = -85.2 kHz or -88.6 kHz) remains steady regardless of temperature drift in frep1 and frep2. If the active biosensing OFC measures a target antigen solution under a certain temperature condition while the dummy biosensing OFC measures a reference material under the same temperature condition, ∆frep reflects the antigen concentration without the influence of temperature drift. In other words, one-to-one correspondence is established between ∆frep and the antigen concentration independent of temperature drift. A photonic RF signal of the dual-comb biosensing was detected by a pair of photodetectors (PDs) and was measured by a RF frequency counter synchronized to a rubidium frequency standard. Details of the dual-comb biosensing are given in the Methods section together with details on the experimental and analytical methodology employed for the following measurements. -9 - We first evaluated the dependence of frep1 on the temperature in the active RI-sensing OFC. To this end, we measured the temporal fluctuation of frep1 when the cavity temperature of the active RI-sensing OFC was not controlled. A pure water was used for a sample and was put into a glass sample cell together with the MMI fiber sensor. Figure 2a shows the frep1 shift (dfrep1) in the active RI-sensing OFC when the cavity temperature changed within a range of 1 ºC. dfrep1 was calculated as the frequency deviation from the initial value of the measurement. The temporal behavior of dfrep1 indicated a temperature sensitivity of about 400 Hz/ºC. To suppress the dfrep1 fluctuation below 1 Hz, it is required to stabilize the cavity temperature within 0.0025 ºC. Although the cavity temperature was actively controlled in the following experiments (see the Methods section), such the temperature stability is technically challenging. For an alternative method, we applied the dual-comb configuration for the active-dummy compensation of temperature drift. Green and blue lines in Fig. 2b shows a shift of frep1 and frep2, namely dfrep1 and dfrep2, with respect to the elapsed time when a pure water was used for a sample for both active and dummy RI-sensing OFCs. The dfrep1 and dfrep2 suffers from the temperature drift over -38.2 Hz due to the decreased cavity temperature; however, their behaviors were almost common to each other. The resulting ∆frep shift (d∆frep) was stable within a range of 1.18 Hz as shown by a red line in Fig. 2b , being equivalent to the temperature stability within a range of 0.0030 ºC. -10 -We next adopted the active-dummy temperature compensation for the RI sensing of a liquid sample. We used glycerol solution with different mixture ratios between glycerin and pure water (= 0 vol%, 1 vol%, 2 vol%, 3 vol%, 4 vol%, and 5 vol%, corresponding to 1.3334 RIU, 1.3350 RIU, 1.3366 RIU, 1.3382 RIU, 1.3398 RIU, and 1.3414 RIU) for a target sample in the active RI-sensing OFC. We exchanged the sample by using a peristaltic pump. On the other hand, a pure water (0-vol% glycerol solution, corresponding to 1.3334 RIU) was used for a reference sample in the dummy RI-sensing OFC. To avoid the temperature increase of the pure water in the sample cell during the measurement, the pure water was exchanged into another pure water by another peristaltic pump when the target sample was exchanged into the different RI glycerol solution. Blue and green lines in Fig.3a show sensorgrams of dfrep1 and dfrep2 when the concentration of the glycerol solution was increased from 0 vol% to 5 vol% while the pure water was exchanged into another pure water at the same timing. Red plots of Fig. 3c shows a relation between the sample RI and d∆frep. A negative linear relationship was confirmed between them with a slope coefficient of -5470 Hz/RIU, corresponding to RI sensitivity. For comparison, we also investigated a relation between the sample RI and dfrep1 as shown by the blue plot of Fig. 3c . A linear relationship was confirmed between them with a RI sensitivity of -9906 Hz/RIU; however, it was significantly shifted from that in d∆frep due to the temperature drift of frep1. Importantly, the RI sensitivity of dfrep1 fluctuates from moment to moment depending on the temperature fluctuation because the behavior of temperature drift also fluctuates from moment to moment (see and compare blue lines between Figs. 2a and 2b), spoiling the precision of RI-sensing OFC. In contrast, one-to-one correspondence between the sample RI and d∆frep is always maintained independently of the temperature fluctuation. The resulting enhanced RI precision enables us to apply this dual RI-sensing OFCs for rapid detection of SARS-CoV-2. The antigen-antibody interaction of virus protein was adopted for the detection of SARS-CoV-2 with the dual-comb biosensing. We selected N protein for the antigen-antibody interaction because of abundant existence, low mutationintroducing probability, high specificity, high sensitivity, and high stability. We formed a -12 - was obtained within a range of 1 aM to 1nM in the curve fitting analysis, thus testifying the validity of the proposed method. Importantly, the sensitivity close to that of RT-PCR has been achieved in the rapid measurement. The residual drift of d∆frep in Fig. 4b hampers the more sensitive measurement of SARS-CoV-2 N protein antigen. We first discuss a possibility to further suppress the residual drift of d∆frep. Although the mechanical sharing dual-comb configuration enables the effective active-dummy temperature compensation, temporal behaviors of frep1 and frep2 in Fig. 2b is not exactly common to each other due to use of two different fiber cavities, leading to the residual drift of d∆frep. Recently, -14 -single-cavity dual-comb fiber lasers based on multiplexing mode-locked oscillation in wavelength [29, 30] , polarization [31, 32] , or propagation direction [33, 34] have attracted attentions for a light source of dual-comb spectroscopy. Since these fiber lasers achieve dual-comb oscillation in the same single cavity, the resulting fluctuation of ∆frep can be reduced below 1 Hz. Although installation of the MMI fiber sensor in the single-cavity dual-comb fiber laser is technically challenging, the combination of them will enhance the sensing performance of dual-comb biosensing for SARS-CoV-2 N protein antigen. Assuming that the residual drift of d∆frep is well suppressed, we next discuss a possibility of further enhancement in the biosensing sensitivity while securing the sufficient compensation of temperature drift. Since the dual-comb biosensing is based on the frequency measurement of ∆frep signal, one potential method to increase the sensitivity is a frequency multiplication of ∆frep signal. For example, when a m-order harmonic component of frep1 and frep2 (freq. = mfrep1 and mfrep2) is measured by using a faster photodetector, a frequency difference between them (= mfrep1 -mfrep2 = m∆frep) is corresponding to a m-fold frequency-multiplied ∆frep signal while securing the activedummy temperature compensation. Unfortunately, as a frequency response of fast photodetectors is typically within a range of several GHz, a frequency multiplication factor m is remained around 100. To further increase m, a combination of dual-comb biosensing with a dual-THz-comb spectroscopy [35, 36] We used a pair of linear fiber cavities mode-locked by saturable absorption for the active biosensing OFC and the dummy biosensing OFC in dual-comb configuration (see Fig. 1b Fig. 1b) . We set frequency spacing of the active and dummy sensing OFCs around 29.4~29.5 MHz by adjusting their cavity length; the resulting ∆frep was -85.2 kHz in RI sensing and --18 -88.6 kHz in biosensing. We adopted mechanically sharing of those two fiber cavities to implement equivalent environmental temperature disturbance to them [28] . The output light from dual OFCs was detected by a pair of photodetectors (PDs, PDA05CF2, Thorlabs, wavelength = 800~1700 nm, frequency bandwidth = 150 MHz), and the resulting frequency signals of frep1 and frep2 were measured by an RF frequency counter (53230A, Keysight Technologies, frequency resolution = 12 digit•s -1 ), which was synchronized to a rubidium frequency standard (FS725, Stanford Research Systems, accuracy = 5 × 10 -11 and instability = 2 × 10 -11 at 1s). Supplementary Figure S1 shows a schematic diagram of the intracavity MMI fiber sensor with the surface modification of antibody on its surface. The MMI fiber sensor was composed of a clad-less MMF (FG125LA, Thorlabs, core diameter= 125 μm, fiber length = 58.94 mm) with a pair of SMFs at both ends (core diameter=6μm, clad diameter=125μm, fiber length = 150 mm) [23] [24] [25] . The OFC light passing through the input SMF is diffracted at the entrance face of the clad-less MMF, and then repeats total internal reflection at the boundary between the clad-less MMF surface and the sample solution. Only the OFC modes satisfying the multi-mode interference wavelength lMMI can exit through the clad-less MMF and then goes toward the output SMF. lMMI is given by Fig. 1a ) [23] . The N protein is a promising candidate for the antigen-antibody interaction of Spectral behavior of frequency comb in optical, THz, and RF region. Variation in false-negative rate of reverse transcriptase polymerase chain reaction-based SARS-CoV-2 tests by time since exposure Evaluation of four commercial kits for SARS-CoV-2 real-time reverse-transcription polymerase chain reaction approved by emergency-use-authorization in Korea Stability issues of RT-PCR testing of SARS-CoV-2 for hospitalized patients clinically diagnosed with COVID-19 Optical biosensors Surface plasmon resonance sensors Surface plasmon resonance One-step conjugation chemistry of DNA with highly scattered silver nanoparticles for sandwich detection of DNA Quantitative and selective surface plasmon resonance response based on a reduced graphene oxide-polyamidoamine nanocomposite for detection of dengue virus e-proteins Detection of norovirus virus-like particles using a surface plasmon resonanceassisted fluoroimmunosensor optimized for quantum dot fluorescent labels A SPR aptasensor for detection of avian influenza virus H5N1 A rapid and quantitative serum test for SARS-CoV-2 antibodies with portable surface plasmon resonance sensing Nano-layered surface plasmon resonance-based highly sensitive biosensor for virus detection: A theoretical approach to detect SARS-CoV-2 Ultrasensitive detection of SARS-CoV-2 nucleocapsid protein using large gold nanoparticle-enhanced surface plasmon resonance Accurate measurement of large optical frequency differences with a mode-locked laser Measurement of the hydrogen 1S-2S transition frequency by phase coherent comparison with a microwave cesium fountain clock -envelope phase control of femtosecond mode-locked lasers and direct optical frequency synthesis Optical frequency metrology Passively mode-locked fiber laser sensor for acoustic pressure sensing Strain sensing based on strain to radio-frequency conversion of optical frequency comb Experimental analysis of optical fiber multimode interference structure and its application to refractive index measurement Sensitivity characteristics of multimode-interference optical-fiber temperature-sensor with solid cladding material Refractiveindex-sensing optical comb based on photonic radio-frequency conversion with intracavity multi-mode interference fiber sensor Refractive index sensing with temperature compensation by a multimode-interference fiber-based optical frequency comb sensing cavity Improvement of dynamic range and repeatability in a refractive-index-sensing optical comb by combining saturable-absorber-mirror mode-locking with an intracavity multimode interference fiber sensor Recent advances in strain gauges Tunable multimode-interference bandpass fiber filter Mechanical sharing dual-comb fiber laser based on an all-polarization-maintaining cavity configuration Picometer-resolution dual-comb spectroscopy with a free-running fiber laser Measurement of absolute frequency of continuous-wave terahertz radiation in real time using a free-running, dual-wavelength mode-locked, erbium-doped fibre laser Polarization-multiplexed, dual-comb all-fiber modelocked laser All-polarization-maintaining, polarization-multiplexed, dual-comb fiber laser with a nonlinear amplifying loop mirror Real-time dual-comb spectroscopy with a free-running bidirectionally mode-locked fiber laser High-coherence ultra-broadband bidirectional dualcomb fiber laser Terahertz comb spectroscopy traceable to microwave frequency standard Spectrally interleaved, comb-mode-resolved spectroscopy using swept dual terahertz combs -26 -project from the Promotion of Regional Industries and Universities by the Cabinet Office, as well as from the Plan for Industry Promotion and Young People's Job Creation by the Creation and Application of Next-Generation Photonics by Tokushima Prefecture. The authors acknowledge Prof developed the dual-comb biosensing system, performed the experiments and analyzed the data The authors declare no competing interests.