key: cord-0728359-l59odcv1
authors: Wu, Kai; Chugh, Vinit Kumar; D. Krishna, Venkatramana; di Girolamo, Arturo; Wang, Yongqiang Andrew; Saha, Renata; Liang, Shuang; Cheeran, Maxim C-J; Wang, Jian-Ping
title: One-Step, Wash-free, Nanoparticle Clustering-Based Magnetic Particle Spectroscopy Bioassay Method for Detection of SARS-CoV-2 Spike and Nucleocapsid Proteins in the Liquid Phase
date: 2021-09-09
journal: ACS Appl Mater Interfaces
DOI: 10.1021/acsami.1c14657
sha: f7d358d5fca2cbd2d79d2d0a2274811f6114b110
doc_id: 728359
cord_uid: l59odcv1

[Image: see text] With the ongoing global pandemic of coronavirus disease 2019 (COVID-19), there is an increasing quest for more accessible, easy-to-use, rapid, inexpensive, and high-accuracy diagnostic tools. Traditional disease diagnostic methods such as qRT-PCR (quantitative reverse transcription-PCR) and ELISA (enzyme-linked immunosorbent assay) require multiple steps, trained technicians, and long turnaround time that may worsen the disease surveillance and pandemic control. In sight of this situation, a rapid, one-step, easy-to-use, and high-accuracy diagnostic platform will be valuable for future epidemic control, especially for regions with scarce medical resources. Herein, we report a magnetic particle spectroscopy (MPS) platform for the detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) biomarkers: spike and nucleocapsid proteins. This technique monitors the dynamic magnetic responses of magnetic nanoparticles (MNPs) and uses their higher harmonics as a measure of the nanoparticles’ binding states. By anchoring polyclonal antibodies (pAbs) onto MNP surfaces, these nanoparticles function as nanoprobes to specifically bind to target analytes (SARS-CoV-2 spike and nucleocapsid proteins in this work) and form nanoparticle clusters. This binding event causes detectable changes in higher harmonics and allows for quantitative and qualitative detection of target analytes in the liquid phase. We have achieved detection limits of 1.56 nM (equivalent to 125 fmole) and 12.5 nM (equivalent to 1 pmole) for detecting SARS-CoV-2 spike and nucleocapsid proteins, respectively. This MPS platform combined with the one-step, wash-free, nanoparticle clustering-based assay method is intrinsically versatile and allows for the detection of a variety of other disease biomarkers by simply changing the surface functional groups on MNPs.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a novel human corona virus, is the causative agent of ongoing coronavirus disease 2019 (COVID- 19) pandemic. It was first identified in patients with acute respiratory illness in Wuhan, China, in December 2019 and soon after became a public health emergency of international concern. 1 Since its emergence, SARS-CoV-2 infected 167 million people worldwide and caused 3.48 million deaths (as of May 25, 2021) . The clinical presentation of SARS-CoV-2 infection ranges from asymptomatic or mild flu-like symptoms to fatal pneumonia and acute respiratory illness. 2 The SARS-CoV-2 belongs to the genus betacoronavirus of the family Coronaviridae. They are enveloped, single-stranded, positive-sense RNA viruses with a genome size of approximately 29.9 kb. 3−5 The SARS-CoV-2 is about 60−140 nm in diameter and consists of 4 structural proteins and 16 nonstructural proteins. 6 The structural proteins such as spike (S), membrane (M), and envelope (E) are associated with the viral envelope, and the nucleocapsid (N) protein forms the capsid outside genomic RNA.

Early diagnosis is critical for clinical management of patients and controlling the spread of COVID-19 by isolating infected patients. Several nucleic acid detection tests targeting N, E, S, or RNA-dependent RNA polymerase genes of the virus are commercially available for the diagnosis of COVID-19. 7−9 Viral RNA detection by quantitative reverse transcription polymerase chain reaction (RT-qPCR) is currently the gold standard test for SARS-CoV-2 infection. 10, 11 Although this nucleic acid amplification test is highly sensitive with the limit of detection as low as 1 RNA copy/μL, RT-PCR needs expensive laboratory instrument, skilled technicians, and days to get the result. Therefore, there is an urgent need for a simple, rapid, sensitive, and accurate point-of-care diagnostic platform that can detect SARS-CoV-2 directly from clinical samples. The antigen detection test can directly detect SARS-CoV-2 in respiratory specimens by detecting virus specific antigens. 12−14 However, some of the commercially available antigen detection tests have very low sensitivity. 15 Magnetic particle spectroscopy (MPS) is one of the most promising candidates for rapid, inexpensive, and high-accuracy bioassays. 16−22 It is a homogeneous biosensing tool that monitors the dynamic magnetic responses of magnetic nanoparticles (MNPs) in the liquid phase. 23−26 By applying AC magnetic fields to MNPs, the magnetic moments of MNPs follow the time-varying external field directions through a combined Neél and Brownian relaxation mechanism. 16,18,27−29 The dominant relaxation of these two processes is dependent on the magnetic core sizes of MNPs (assuming unconstrained MNPs in liquid without surface binding of any chemical substances). 30 −32 The Brownian relaxation process is domi-nant for single-core iron oxide MNPs with core sizes above 20 nm (this critical size may vary for different magnetic materials of different magnetic properties). 33−37 This relaxation process reflects the degree of freedom of physical rotational motion of MNPs. Conjugation of any chemical substances including protein molecules, aptamers, and other nonmagnetic materials onto MNPs can hinder or even block this Brownian relaxation. 23,27,38−45 As a result, this binding event causes a phase lag between magnetic moments and external AC fields (except at very low frequencies), and thus, weaker dynamic magnetic responses are observed from clustered MNPs. By exploiting this unique property of Brownian relaxationdominant MNPs, researchers have applied it along with the MPS platform for the detection of H1N1 virus, 39 thrombin, 38 staphylococcal toxins, 20 SARS-CoV-2, 23,46 botulinum neurotoxins A, B, and E, 21 and so forth. Compared to the traditional optical, mechanical, and electrochemical sensing techniques, this magnetic sensing technique is immune to the background noise from biological samples that may interfere with the signal reading because most biological substrates are nonmagnetic (or paramagnetic) and MNPs are the only sources of the magnetic signal.

Our group has successfully built up both benchtop and portable MPS systems for one-step, wash-free bioassays. 22, 39 In this paper, we have modified the MPS system to include a onestage lock-in scheme for the advantages of down sampling and phase-sensitive rejection of noise. Herein, we report a one-step, wash-free, nanoparticle clustering-based MPS bioassay method for the detection of SARS-CoV-2 spike and nucleocapsid proteins in the liquid phase using this one-stage lock-in MPS system. By surface-conjugating polyclonal antibodies (pAbs) to MNPs, each MNP will be able to specifically bind to target protein molecules. On the other hand, each protein molecule has multiple epitopes that allow for the specific bindings of pAbs. Thus, the pAbs anchored to MNP surfaces and the nature of different epitopes from each protein molecule allow us to amplify the binding events caused by dynamic magnetic response change. We demonstrate that this nanoparticle clustering-based detection method can detect target analytes by monitoring the dynamic magnetic responses of MNPs.

2.1. Materials. The SARS-CoV-2 recombinant nucleocapsid protein (His tag, consists of 430 amino acids and predicts a molecular mass of 47.08 kDa, Cat: 40588-V08B), the recombinant spike protein (RBD, His tag, consists of 234 amino acids with a molecular mass of 26.54 kDa, Cat: 40592-V08H), the nucleocapsid antibody (polyclonal rabbit IgG, Cat: 40588-T62), and the spike RBD antibody (polyclonal rabbit IgG, Cat: 40592-T62) were purchased from Sino Biological Inc. The IPG30 MNPs were 30 nm iron oxide nanoparticles functionalized with protein G, with a weight concentration of 1.7 mg/mL and particle concentration of 57.8 nM, provided by Ocean NanoTech. The phosphate-buffered saline (PBS, Cat: 79378) was purchased from Sigma-Aldrich Inc.

2.2. Magnetic Property Characterization. 10 μL of IPG30 MNP suspensions were pipetted onto a filter paper and air-dried. Then, the static magnetic hysteresis loops were collected on a Physical Properties Measurement System (PPMS, Quantum Design Inc.) to obtain the magnetic properties of these nanoparticles such as saturation (M s ) and coercivity (H c ). The static magnetic hysteresis loops of IPG30 MNPs were measured under external fields of −5000−5000 Oe and −500−500 Oe, and the magnetic properties were analyzed in the Supporting Information S1.

2.3. One-Stage Lock-In MPS System. The MPS platform used in this work utilizes the frequency mixing approach where the excitation magnetic field consists of a low-frequency ( f L ) component and a high-frequency (f H ) component. 17, [19] [20] [21] [22] 26, 39, 43 A low frequency field having 50 Hz frequency and 250 Oe amplitude and a highfrequency field having 5000 Hz frequency and 25 Oe amplitude were used for the magnetic excitation of the MNPs. In this work, the MPS system is modified with the addition of a one-stage lock-in implementation to improve the detection sensitivity of the system. Figure 1a shows the schematic representation of the sensing scheme used for decoding the MNP response signal in the one-stage lock-in MPS system. The signal from pick-up coils is amplified using a high-precision instrumentation amplifier, INA828 from Texas Instruments. Figure 1b ,c shows the time-domain and frequency-domain representations of the MNP signal at this stage. The amplified signal is then processed by a one-stage lock-in implementation consisting of a synchronous demodulator followed by bandpass filtering. AD630 from Analog Devices is used for the synchronous demodulator application. Figure 1d ,e depicts the corresponding time-domain and frequency-domain presentations of the MNP signal after the lock-in implementation stage. This one-stage lock-in MPS system helps remove the feedthrough signals corresponding to the excitation magnetic field frequencies and only records the dynamic magnetic responses of MNPs, hence improving the sensitivity. The filtered signal is sampled at a 100 KSPS sampling rate and communicated to a laptop for postprocessing and data analysis. Each MPS reading for a vial contains 170,000 samples, which are collected over a period of roughly 2 s. If multiple readings are required from a sample, we use a cooldown time of 3 min between the readings to ensure neither the coils nor the analogue driving circuitry heats up too much.

2.4. MPS-Based Bioassay Method. The MPS-based bioassays for the detection of target analytes in the liquid phase allows for onestep, wash-free assays that can be handled by a layperson without much technical training requirements. This method relies on the intentional interference on the Brownian relaxation of MNPs in liquid. When subjected to external AC magnetic fields (excitation fields), the magnetic moments of MNPs try to realign to external field directions through the Brownian relaxation. This process will be hindered if MNPs are surface-bound with polymers, protein molecules (such as antibody and antigen), aptamers, and so forth, 18, 23, 26, 38 fixed on a surface/substrate, 20, 21, 46, 47 or form nanoparticle clusters. 26, 38, 39 Figure 2b shows a schematic view that by surface coating an increased number of antibodies, the dynamic magnetic responses of MNPs (in the form of MPS spectra, i.e., the amplitudes of higher harmonics) become weaker. In the presence of target analytes, the antibody-functionalized MNPs bind to these analytes and form nanoparticle clusters. As a result, this nanoparticle clustering events further weaken the dynamic magnetic response, thus lowering the harmonic amplitudes, as shown in Figure 2c . On the other hand, if less target analytes are present, then less nanoparticle clusters are formed and, as a result, the dynamic magnetic responses of MNPs yield stronger harmonics ( Figure 2d ).

Groups. We first evaluated the detection limit of our one-stage lock-in MPS system for the detection of IPG30 MNPs. Samples containing a varying concentration/amount of IPG30 MNPs are prepared by two-fold dilutions, as shown in Table  S1 from Supporting Information S2. Three independent MPS readings are taken from each sample, and the amplitudes of higher harmonics are summarized in Figure S2 from Supporting Information S2. Results show that the developed one-stage lock-in topology-based MPS system can detect as low as 266 ng (equal to 9 fmole) of IPG30 MNPs, which equals to 512-fold dilutions of the original concentration. Compared with our previous work, this modified MPS system has higher detection sensitivity. 22, 26, 39 Thus, for the bioassays in this work, we used diluted IPG30 MNPs with the final particle concentration of 3.06 nM and weight concentration of 90 μg/ mL. This feasibility attempt allows us to explore inexpensive bioassays on our MPS platform for future high-volume tests at the users' ends and in regions with scarce medical resources. Furthermore, it is reported that for MPS-based homogeneous bioassays, lower MNP concentrations could improve the detection sensitivities of biomolecules. 24, 48 The design of one-step, wash-free, nanoparticle clustering-based MPS bioassays is highly relied on the surface functionalization of pAbs on MNPs. It has been reported in our previous work that the pAbs functionalized on MNPs can trigger the cross-linking between MNPs and target analytes, thus forming MNP clusters. 39 In this work, the SARS-CoV-2 spike and nucleocapsid proteins have multiple epitopes that can be recognized by their corresponding pAbs. As a result, each protein molecule (i.e., spike and nucleocapsid) can bind to multiple pAbs from multiple MNPs. On the other hand, each MNP functionalized with multiple pAbs can bind to multiple protein molecules. A schematic example is given in Figure 2c .

As shown in Figure 2a Table 1 . For the SARS-CoV-2 spike protein tests, we prepared four groups of samples along with one negative control sample (vial #9) that contains bare IPG30 MNP without pAb coating (labeled as "bare MNP" in this work). In each group, the vials #X-8 (X = I, II, III, and IV) serve as the negative control that contain MNPs functionalized with a varying amount of pAbs, while 0 nM protein (we replaced the protein with PBS buffer) is added. Thus, a S and N stand for SARS-CoV-2 spike (S) and nucleocapsid (N) proteins, respectively. b X = I, II, III, and IV. There are four groups of samples prepared for S and N protein tests. As shown in Figure 2a , the diluted IPG30 MNPs (particle concentration of 3.06 nM) are incubated with spike or nucleocapsid pAbs for 12 h at 4°C on an incubating shaker. This step allows the Fc part of IgG pAbs to bind to protein G from IPG30 MNP surfaces. Then, 80 μL of pAb-functionalized MNPs is mixed with 80 μL of spike or nucleocapsid proteins and incubated at room temperature for 1 h with continuous shaking. After this point, the specific binding between pAbs and target protein molecules reaches to an equilibrium state, and sample vials are then stored at 4°C before MPS tests. It should be noted that the 1 h incubation time of pAb-functionalized MNPs with protein biomarkers is not optimized. To ensure the antibody−antigen binding to fully saturate, we set an incubation time of 1 h. Further works should be carried out to reduce the incubation time by means of changing incubation conditions (such as temperature, 49 Figure 4a −d. Two-sample ttest is applied to compare the differences between two sets of harmonic data collected from different samples.

For the scenario of coating four pAbs onto each MNP (Group I, MNP/pAb = 1:4, Figure 4a ), adding 400 nM (equivalent to 32 pmole) of spike protein to vial #I-1 causes the 3rd harmonic to be significantly lower (p < 0.001) than the 3rd harmonics from other vials in Group I. This weaker harmonic signal is due to the abundancy of target spike protein molecules from the sample that causes the nanoparticle clustering, as explained in Section 2.4 and Figure 2c ,d. However, vials #I-2 to #I-8 show no significant differences.

In Group II (MNP/pAb = 1:3, Figure 4b ), vials #II-1 to #II-4 show similar harmonic amplitudes and are significantly lower than other vials from the same group, which indicates that in this scenario (MNP/pAb = 1:3), adding 32 pmole (400 nM), 8 pmole (100 nM), 2 pmole (25 nM), and 1 pmole (12.5 nM) SARS-CoV-2 spike protein can completely block the Brownian relaxation of MNPs and cause the lowest achievable MPS spectra (or weakest dynamic magnetic responses). By reducing the number of spike protein in vials #II-5 to #II-8, the 3rd harmonic amplitudes gradually increase due to the less amount of the target spike protein molecule present in the sample that can cause the nanoparticle clustering. From Group II, a clear linear concentration-response curve is observed from vials #II-4 to #II-8. Indicating that this bioassay design could be used for quantitative assays of spike protein samples within a concentration range of 0−12.5 nM (corresponding to the spike protein amount of 0−1 pmole) and qualitative assays of spike protein samples with concentrations above 12.5 nM.

On the other hand, by functionalizing two pAbs (Group III, MNP/pAb = 1:2, Figure 4c ) and one pAb (Group IV, MNP/ pAb = 1:1, Figure 4d ) per MNP, no significant response curves are observed. There is a clear trend showing that the harmonic signal amplitude decreases, as we add a higher number of spike protein molecules. Figure 4e compares the 3rd harmonic amplitudes from vials #X-1 (X = I, II, III, and IV), bare MNP (vial #9), and blank (no IPG30 MNP, i.e., noise floor). All the vials added with IPG30 MNPs show significantly higher harmonic signal intensities than the blank sample, of which the bare MNP (vial #9) sample shows the highest harmonic signal because no pAbs are functionalized and the Brownian relaxation of MNPs is not hindered even by adding 400 nM (32 pmole) spike protein to it, which confirms that MNPs without pAb functionalization will not form clusters in the presence of target analytes. In addition, the 3rd harmonics from vials #X-3, #X-5, #X-7, and #X-8 (X = I, II, III, and IV) are compared in Figure 4f −i, respectively. A clear trend is observed that by adding the same amount of spike protein molecules, MNPs that functionalized with more pAbs show a lower harmonic signal. Figure 4i proves the schematic drawing in Figure 2b that functionalizing more pAbs per MNP will cause weaker dynamic magnetic responses and lower harmonic signals, which also confirms that the pAbs have been successfully functionalized onto IPG30 MNPs.

From the concentration-response profiles of SARS-CoV-2 spike protein, it is concluded that functionalizing three pAbs per MNP (Group II, MNP/pAb = 1:3, Figure 4b) gives us the best linear response curve with a detection limit of 1.56 nM (equivalent to 125 fmole of spike protein molecules).

In Figure 5 , we have summarized the ratios of higher harmonics (from the 5th to the 15th harmonics) to the 3rd harmonics collected from each sample. For Groups I, III, and IV (Figure 5a,c,d) , the harmonic ratio curves of vials #X-1 to #X-8 (X = I, III, and IV) are tightly distributed with very narrow gaps or even overlapping, while the harmonic ratio curves from Group II vials #II-1 to #II-8 are sparsely distributed. The significant differences in harmonic ratio curves from vials added with different concentrations/amounts of spike protein molecules allow us to analyze and collect meaningful concentration-response profiles, as shown in Figure  4b .

The harmonic amplitude decreases as the harmonic index increases, namely, the harmonic amplitudes A3 > A5 > A7 > A9 > A11 > A13 > A15, where A represents amplitude (A). It is observed that functionalizing different amount of pAbs per MNP also changes the decay rate of higher harmonics (with a harmonic index), as reflected in the MPS spectra from Figure  3c Research Article 0.52, and 0.51, respectively. Thus, conjugating more pAbs on each MNP will make the higher harmonics decay slower. Besides comparing each sample's higher harmonic amplitudes (A5−A15) to its own 3rd harmonic (A3), we further compared the samples' higher harmonics from Group I−IV with the bare MNP sample (vial #9, IPG30 MNP without any surface functionalization). As shown in Supporting Information S4, the 3rd, 5th, 7th, 9th, 11th, 13th, and 15th harmonics of each sample are compared with the corresponding harmonics from the bare MNP sample (vial #9) and the grayscale heatmaps of the harmonic signal drop (defined as Figure 6a compares the 3rd harmonic of each sample with the bare MNP sample (vial #9). In each row of Figure 6a , by adding the same amount of spike protein molecules, the 3rd harmonic signal drop Δ decreases from I to IV, which agrees with the results shown in Figure 4e −i. Thus, ideally, the color becomes darker from the left column to right column in each grayscale heatmap (as schemed in Figure 6b ). In each column of Figure 6a , with IPG30 MNPs surface functionalized with an identical number of pAbs, adding more spike protein causes a larger harmonic signal drop. Ideally, the color becomes darker from the bottom row (vials #X-1, X = I, II, III, and IV) to the top row (vials #X-8, X = I, II, III, and IV). Again, this ideal trend is shown in Figure 6b . There are two variables in plotting the grayscale heatmaps of the higher harmonic signal drop Δ: (1) the number of pAbs functionalized on each MNP (Groups I−IV with MNP/pAb = 1:4, 1:3, 1:2, and 1:1) and (2) the concentration/amount of the spike protein added to each sample (i.e., vials #X-1 to #X-8, X = I, II, III, and IV). These two variables jointly cause the harmonic signal drop Δ increases from the upper right corner to bottom left corner in diagonal. The 3rd harmonics give a highest signal to noise ratio, and we can clearly see this trend from Figure 6a . Figure 6a(1−4) plots the 3rd harmonic signal drop Δ curves for samples from Groups I−IV, where Group II shows the best monotonic concentration-response curve. Figure 6b schematically shows the scenarios where extra and insufficient target analytes (i.e., spike protein) are added. As a result, the number of target analytes directly affects the degree of nanoparticle clustering and the dynamic magnetic responses.

3.3. Concentration-Response Profiles of SARS-CoV-2 Nucleocapsid Protein. A total of 33 SARS-CoV-2 nucleocapsid samples are prepared in the same manner as the spike protein samples aforementioned. Six independent MPS readings are taken from each sample, and the 3rd harmonics are plotted in Figure 7a −d. Group I (MNP/pAb = 1:4) shows a remarkable monotonic concentration-response curve where the 3rd harmonics monotonically increase from vial #I-1 to vial #I-6, with nucleocapsid protein decreasing from 400 (32 pmole) to 3.13 nM (500 fmole). Similar trends are also observed from Group II (MNP/pAb = 1:3, Figure 7b ) and Group III (MNP/pAb = 1:2, Figure 7c ), while both are less significant than Group I. The detection limit of the SARS-CoV-2 nucleocapsid protein on our MPS system is 12.5 nM (equivalent to 1 pmole) based on the response curves from Group I.

We also observed that when functionalizing one pAb per MNP (Group IV), no remarkable response curves are observed from either spike or nucleocapsid proteins (Figures 4d and  7d ). For Group IV (MNP/pAb = 1:1) in spike protein tests, vials #IV-1 to #IV-4 are significantly different (p < 0.001) from vials #IV-5 to #IV-8, but no significant differences are observed within vials #IV-1 to #IV-4 or vials #IV-5 to #IV-8, as shown in Figure 4d . Similar results are observed from Group IV (MNP/ pAb = 1:1) in nucleocapsid protein tests, and vials #IV-1 to #IV-5 are significantly different (p < 0.001) from vials #IV-6 to #IV-8. Thus, for the detection of spike and nucleocapsid proteins, it is not practical to functionalize one pAb per MNP. The grayscale heatmaps of the harmonic signal drop Δ from nucleocapsid samples are also provided in Supporting Information S5. The ratios of higher harmonics to the 3rd harmonics recorded from Groups I to IV for the nucleocapsid protein can be found in Supporting Information S6.

Herein, we have successfully applied a one-stage lock-in MPS system for the detection of SARS-CoV-2 spike and nucleocapsid proteins directly in the liquid phase. This onestep, wash-free detection method allows a layperson to carry out the tests without much technical training requirements. By conjugating pAbs to MNPs, it increases the chances of MNP binding events to target protein molecules (i.e., spike and nucleocapsid proteins in this work). Combined with the nature that each protein molecule has different epitopes that allow for the binding of multiple pAbs. We are able to amplify the effect of MNP binding events that caused obstruction of physical rotational motion of MNPs, thus amplifying the changes in higher harmonics.

In addition, we explored the effect of the pAbs amount functionalized on each MNP. Four groups of samples with MNP/pAb = 1:4 (Group I), 1:3 (Group II), 1:2 (Group III), and 1:1 (Group IV) are designed. These pAb-functionalized MNPs are used to quantitatively detect spike and nucleocapsid proteins varying from 0 fmole (vials #X-8) to 32 pmole (vials #X-1). The concentration-response profiles for spike and nucleocapsid proteins are generated based on the higher harmonics from MNPs. It is concluded that conjugating three pAbs per MNP yields a linear concentration-response curve and reaches a detection limit of 1.56 nM (equivalent to 125 fmole) for spike protein molecules. On the other hand, conjugating four pAb per MNP yields a linear concentrationresponse curve for the detection of nucleocapsid proteins and a detection limit of 12.5 nM (equivalent to 1 pmole) is achieved. The higher harmonics from all the control groups, bare MNPs (vial #9) and vials #X-8 (X = I, II, III, and IV, where no proteins are added), confirmed that the pAbs have been successfully conjugated to MNPs.

To sum up, this nanoparticle clustering-based bioassay mechanism combined with the MPS platform is intrinsically versatile and can be applied for the detections of other biomarkers such as proteins, viruses, nucleic acids, and so forth. Its nature of one-step wash-free experimental procedure allows for simpler and more convenient on-field tests in the future. Zhong et al. also reported the SARS-CoV-2 spike protein detection from the liquid phase using a similar MPS platform, 23 where they coated anti-SARS-CoV-2 spike protein monoclonal antibodies (mAbs) to the MNPs and spike protein to 100 nm polystyrene (PS) beads. The binding events of MNPs to 100 nm PS beads through antibody−antigen-specific recognition hinder the physical rotational motion of MNPs. Thus, in a similar manner, higher harmonics are recorded as indicators of spike protein abundancy. Each PS bead was coated with 100 SARS-CoV-2 spike proteins to mimic one SARS-CoV-2 virus particle. They demonstrated the detection limit of 84 fM (equal to 5.9 fmole) mimic virus particles, which is equal to 590 fmole of spike protein molecules. The main differences between our work and Zhong et al.'s work are that we utilized MNP-pAb-protein-pAb-MNP-pAb-protein-pAb-MNP-... clustering that caused obstruction of physical rotational motion of MNPs to achieve specific detection of the target SARS-CoV-2 spike and nucleocapsid protein, while Zhong et al. used an MNP-mAb-protein-PS bead to block the physical rotational motion of MNPs. Although the detection strategy is different, the detection mechanism is the same, that is, relying on the change of the harmonic signal caused by the reduced rotational freedom of the MNPs to quantitatively and/ or qualitatively detect the target the SARS-CoV-2 protein biomarker.

It should also be noted that the results presented in this work are based on detecting purified spike and nucleocapsid proteins from PBS buffer. This is a feasibility demonstration of the one-stage lock-in MPS system for protein biomarker detection, and we did not measure SARS-CoV-2 biomarkers from clinical samples such as upper respiratory tract mucus including saliva and sputum. There are several factors that may impair the sensitivity and repeatability of MPS platforms when testing clinical samples, such as viscosity of biofluid samples, temperature, and nonspecific binding of unknown chemical compounds from biofluids. Because this one-step liquid phase ACS Applied Materials & Interfaces www.acsami.org

Research Article detection mechanism relies on the binding event-caused Brownian relaxation change in MNPs, the viscosity and temperature of biofluid samples can cause different degrees of deviations in MPS harmonic signals and the nonspecific binding can cause repeatability issues (as well as false positive results). To avoid the viscosity variation caused by MPS signal bias, several sample preparation and collection methods can be adopted, such as using the exhaled breath condensate that contains respiratory droplets from the lower respiratory tract. 54−56 The condensed water droplets containing SARS-CoV-2 virions can effectively avoid the body fluid (i.e., sputum and saliva) that caused viscosity variations between patients. In addition to the sample collection method, the sample preparation method can also be altered for this specific application. Usually, the upper respiratory tract mucus samples (nasopharyngeal swabs, throat swabs, nasopharyngeal wash, saliva, sputum, etc.) have high virus titer, but the complex internal matrix could interfere with the MPS test results due to the high viscosity and unknown chemical compounds that nonspecifically bind to MNPs. 57, 58 The binding specificities of pAbs used in this work have been confirmed by ELISA testing (see Supporting Information S7). A mature sample preparation process has been applied to dilute specimens by washing buffers and filter out larger particle complexes. 59 The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsami.1c14657. Static magnetic hysteresis loops and magnetic properties of IPG30 MNPs; minimum detectable amount of IPG30 by the homebuilt MPS system; hydrodynamic size, zeta potential, and the 3rd harmonic amplitudes of bare MNPs and MNPs functionalized with different amounts of pAbs; grayscale heatmaps of the higher harmonic signal drop Δ (in %) compared to bare MNPs for SARS-CoV-2 spike proteins; grayscale heatmaps of the higher harmonic signal drop n (in %) compared to bare MNPs for SARS-CoV-2 nucleocapsid proteins; ratios of higher harmonics to the 3rd harmonics recorded from Groups I to IV for the SARS-CoV-2 nucleocapsid protein; and specificity of the SARS-CoV-2 spike and nucleocapsid pAbs confirmed by ELISA (PDF)

Clinical Features of Patients Infected with 2019 Novel Coronavirus in Wuhan

Genomic Characterisation and Epidemiology of 2019 Novel Coronavirus: Implications for Virus Origins and Receptor Binding

Genomic Characterization of the 2019 Novel Human-Pathogenic Coronavirus Isolated from a Patient with Atypical Pneumonia after Visiting Wuhan

A New Coronavirus Associated with Human Respiratory Disease in China

A Novel Coronavirus from Patients with Pneumonia in China

Tools and Techniques for Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2)/COVID-19 Detection

Challenges in Laboratory Diagnosis of the Novel Coronavirus SARS-CoV-2. Viruses 2020

Molecular Diagnosis of a Novel Coronavirus (2019-NCoV) Causing an Outbreak of Pneumonia

Laboratory Diagnosis of Emerging Human Coronavirus Infections−the State of the Art

Evaluation of a Novel Antigen-Based Rapid Detection Test for the Diagnosis of SARS-CoV-2 in Respiratory Samples

Development of a Portable, Ultra-Rapid and Ultra-Sensitive Cell-Based Biosensor for the Direct Detection of the SARS-CoV-2 S1 Spike Protein Antigen

Microfluidic Immunoassays for Sensitive and Simultaneous Detection of IgG/IgM/Antigen of SARS-CoV-2 within 15 Min

Evaluation of Rapid Antigen Test for Detection of SARS-CoV-2 Virus

Magnetic Particle Spectroscopy: A Short Review of Applications Using Magnetic Nanoparticles

New Type of Biosensor Based on Magnetic Nanoparticle Detection

Magnetic Particle Spectroscopy-Based Bioassays: Methods, Applications, Advances, and Future Opportunities

Magnetic Particle Detection by Frequency Mixing for Immunoassay Applications

Magnetic Immunoassay for Detection of Staphylococcal Toxins in Complex Media

Multiplex Biosensing Based on Highly Sensitive Magnetic Nanolabel Quantification: Rapid Detection of Botulinum Neurotoxins A, B, and E in Liquids

A Portable Magnetic Particle Spectrometer for Future Rapid and Wash-Free Bioassays

Toward Rapid and Sensitive Detection of SARS-CoV-2 with Functionalized Magnetic Nanoparticles

Dependence of Biomolecule Detection on Magnetic Nanoparticle Concentration

Characterizing Physical Properties of Superparamagnetic Nanoparticles in Liquid Phase Using Brownian Relaxation

Magnetic Nanoparticle Relaxation Dynamics-Based Magnetic Particle Spectroscopy for Rapid and Wash-Free Molecular Sensing

Measuring Protein Biomarker Concentrations Using Antibody Tagged Magnetic Nanoparticles

Using Magnetic Nanoparticles and Protein-Protein Interactions to Measure pH at the Nanoscale

Quantification of Magnetic Nanoparticles by Compensating for Multiple Environment Changes Simultaneously

Biological Sensors Based on Brownian Relaxation of Magnetic Nanoparticles

Real-Time Measurement of Brownian Relaxation of Magnetic Nanoparticles by a Mixing-Frequency Method

Magnetic Dynamics of Ferrofluids: Mathematical Models and Experimental Investigations

Effects of Size and Anisotropy of Magnetic Nanoparticles Associated with Dynamics of Easy Axis for Magnetic Particle Imaging

The Role of Anisotropy in Distinguishing Domination of Neél or Brownian Relaxation Contribution to Magnetic Inductive Heating: Orientations for Biomedical Applications

Effects of Particle Diameter and Magnetocrystalline Anisotropy on Magnetic Relaxation and Magnetic Particle Imaging Performance of Magnetic Nanoparticles

Molecular Sensing with Magnetic Nanoparticles Using Magnetic Spectroscopy of Nanoparticle Brownian Motion

Magnetic Particle Spectroscopy for Detection of Influenza A Virus Subtype H1N1

Multiplex Label-Free Biosensor for Detection of Autoantibodies in Human Serum: Tool for New Kinetics-Based Diagnostics of Autoimmune Diseases

Evaluating Blood Clot Progression Using Magnetic Particle Spectroscopy

Blood Clot Detection Using Magnetic Nanoparticles

Multiplex Biosensing with Highly Sensitive Magnetic Nanoparticle Quantification Method

Highly Reproducible and Sensitive Detection of Mycotoxins by Label-Free Biosensors

Ultrasensitive Quantitative Detection of Small Molecules with Rapid Lateral-Flow Assay Based on High-Affinity Bifunctional Ligand and Magnetic Nanolabels

Brief Communication: Magnetic Immuno-Detection o f S A R S -C o V -2 S p e c ifi c A n t ib o d i es

Nanomagnetic Lateral Flow Assay for High-Precision Quantification of Diagnostically Relevant Concentrations of Serum TSH

Magnetic Particle Spectroscopy (MPS) with One-Stage Lock-in Implementation for Magnetic Bioassays with Improved Sensitivities. 2021

Deoxyribonucleic Acid Homology in Bacterial Taxonomy: Effect of Incubation Temperature on Reaction Specificity

Unbinding Forces of Single Antibody-Antigen Complexes Correlate with Their Thermal Dissociation Rates

Jr Use of High Pressure to Accelerate Antibody: Antigen Binding Kinetics Demonstrated in an HIV-1 P24: Anti-HIV-1 P24 Assay

Accelerated Surface-Enhanced Raman Spectroscopy (SERS)-Based Immunoassay on a Gold-Plated Membrane

Kinetic Analysis of a High-Affinity Antibody/Antigen Interaction Performed by Planar Waveguide Fluorescence Immunosensor

Breath Diagnostics in the Era of SARS-CoV-2Clinical and Research Arena

Use of Exhaled Breath Condensate (EBC) in the Diagnosis of SARS-COV-2 (COVID-19)

A Mathematical Model Describing the Localization and Spread of Influenza A Virus Infection within the Human Respiratory Tract

Respiratory Droplet Generation and Dispersal during Nasoendoscopy and Upper Respiratory Swab Testing

Fully Packaged Portable Thin Film Biosensor for the Direct Detection of Highly Pathogenic Viruses from On-Site Samples

Multiplexed Detection of Influenza A Virus Subtype H5 and H9 via Quantum Dot-Based Immunoassay

Rapid Enrichment and Ultrasensitive Detection of Influenza A Virus in Human Specimen Using Magnetic Quantum Dot Nanobeads Based Test Strips

Effect of Antigen Retrieval Methods on Nonspecific Binding of Antibody-Metal Nanoparticle Conjugates on Formalin-Fixed Paraffin-Embedded Tissue

Minimizing the Non-Specific Binding of Nanoparticles to the Brain Enables Active Targeting of Fn14-Positive Glioblastoma Cells

The authors declare no competing financial interest.