Close this search box.

Biporous silica nanostructure-induced nanovortex in microfluidics for nucleic acid enrichment, isolation, and PCR-free detection – Nature Communications

Synthesis of PSNF and BSNF

2.992 g of triethanolamine (Sigma-Aldrich, 90279) was dissolved in 1.045 L of distilled water at 80 °C and 100 μm thin PET films (Kemafoil hydrophilic film, HNW-100, COVEME, Italy, 10 cm by 7 cm) were immersed in the solution. After 30 min, 61.16 ml of cetyltrimethylammonium chloride (CTAC) (Sigma-Aldrich, 292737) and 12.496 g of sodium salicylate (NaSal) (Sigma-Aldrich, S3007) were added under vigorous stirring to form micelle templates36. Subsequently, 110 ml of tetraethyl orthosilicate (TEOS) (Sigma-Aldrich, 86578) was added into the solution and stirred at 80 °C for 3 h. The PSNF-synthesized PET films were washed three times with ethanol and water and dried for further use.

The PSNF with small pores around 60 nm was employed not only as the binding site with high surface area, but also as the seed layer for the synthesis of BSNF. A mixture of 1.045 L distilled water, 61.16 ml of CTAC and 12.496 g of NaSal were stirred vigorously to form giant micelle templates at 80 °C for 2 h. PSNF-synthesized PET films were immersed in the solution for 2 h under stirring. Then, 110 ml of TEOS was added and stirred for 2 h. Finally, BSNF-synthesized PET films were washed three times with ethanol and water and stored in dry.

Design and fabrication of the BSNFs-chip

Microfluidic chips, which include the Flat-chip, PSNFs-chip, and (BSNFs-chip, are structured as three-layered devices. Each layer is composed of either a 100 μm flat PET film, PSNF, or BSNF for the top and bottom layers. In the center, a layer features double-sided tape structures, which are 300 μm in height and arranged in a spiral shape. These structures connect the top and bottom films and manage the circulation of fluid within the microchannels. The overall external height of the microfluidic chip reaches 500 μm, while the internal microchannels maintain a height of 300 μm. The designs for the film and the double-sided tape were drafted using AutoCAD 2019 software (Autodesk, Inc., San Rafael, CA, USA). Both the film and the double-sided tape structures were shaped into rectangles measuring 70.16 mm by 85 mm. Specifically, the double-sided tape structures were designed as a repeated sequence of 12 microfluidic channels, each composed of 13–14 half-oval shapes with a diameter of 4.5 mm. This design enabled a serpentine flow and the injection of ~650 µl of liquid. Following the design process, the layouts were cut into their desired shapes using the VLS3.50 laser machine (Universal Laser Systems, Scottsdale, AZ, USA)33.

Before usage, the microfluidic chips were assembled, and the internal channels were treated to promote amine group functionality. The initial step involved subjecting the films to O2 plasma treatment for surface cleansing and alteration, which led to the production of hydroxyl groups (OH) to enhance hydrophilicity and chemical reactivity. Once treated, the layers were precisely aligned and fastened together. Acrylic adaptors were then affixed to the inlet and outlet of the microfluidic chip’s upper layer to facilitate sample injection and expulsion. Tygon tubing was linked to the adaptor openings using epoxy, thus ensuring a robust connection for fluid movement. Following this, the inner surface of the microfluidic chips was functionalized by introducing a 2% solution of 3-aminopropyl(diethoxy)methylsilane (APDMS; Cat no. 371890, Sigma-Aldrich, St. Louis, MO, USA) into ultra-pure deionized water (DW). Subsequently, the microfluidic chips were incubated at 65 °C for 60 min and then washed thrice with 1 ml of DW each time to eliminate any residual silane. After the fabrication process was concluded, the amine group functionalized Flat-chip, PSNFs-chip, and BSNFs-chip were stored at ambient temperature until needed for further use.


Computational modeling and simulations were carried out via the commercial solver of the GeoDict® 2023 software package (Math2Market GmbH, Germany). GeoDict is a voxel-based tool and is used for prediction of the surface area and flow performance of the simulated virtual models. Virtual models were generated with the FiberGeo module and the Sinter & Crystallization algorithm of the GrainGeo module. The pore diameter and the thickness of pore wall were constructed by the approximate values of the experimental values. Domains discretized with a voxel length of 1 nm were made to be 500 × 500 nm2 (x–z plane), and 3000 nm (y-axis) in flat, 3145 nm (y-axis) in PSNF, and 3400 (y-axis) nm in BSNF, respectively. Then, MatDict module was used to calculate the surface area of the simulated virtual models.

A FlowDict module was used to calculate the flow profiles for the simulated virtual models. The flow of water was obtained from the Stoke’s equation considering at a given flow condition. The water flow was passed in z-direction, and 100 μl min−1 of flow rate and 21.66 cm2 of area (inner surface area of microchannel) were imposed as the experimental setup. The periodic boundary condition was employed in the peripheral region to model the infinite periodic domains.

Particle proximity test

To confirm the slip effect of BSNF, we conducted a particle proximity test through the visualization of particle flow paths. A micro-scale channel (3 mm width, 0.1 mm height) was fabricated using transparent polydimethylsiloxane (PDMS) for visualization. An inlet and outlet were placed on both sides of the channel, and a 2 mm wide PET film was placed at the middle of the channel vertically (Fig. 3k). Then, an aqueous solution containing 10 μm fluorescent polystyrene beads (Phosphorex, Inc., USA) was injected into the channel at a flow rate of 20 µl min−1 using a syringe pump (Fusion 200, Chemyx Inc., USA). The trajectories of the particles were captured at every 0.05 s using an inverted microscopy (IX-71, Olympus Co., Japan) and EMCCD (ImagEMx2, Hamamatsu Co., Japan). The particle flow paths were visualized by stacking 50 frames of images (i.e., during 2.5 s).

Material characterization

SEM images were obtained using a ZEISS Sigma 300 field-emission SEM (FE-SEM) (ZEISS, Germany) at the Center for Polymers and Composite Materials, Hanyang University, Korea. The zeta potential was measured by a Malvern Zetasizer (Malvern, UK). Feret diameter and mean gray value were analyzed in ImageJ software. SAXS measurement was performed using a Xeuss 2.0 (Xenocs, France) with a Cu Kα radiation (λ = 0.154 nm) and a sample-to-detector distance of 1500 mm. Fluorescence images were observed under fluorescence imaging system (EVOS M7000 Imaging System, Thermo Fisher Scientific, USA) using Qdot™ ITK™ Carboxyl Quantum Dots (ThermoFisher Scientific, USA). FT-IR spectrum was obtained using a PerkinElmer Frontier™ FT-IR spectrometer with a diamond ATR (Perkin Elmer, USA), and water contact angle was measured using pendant drop tensiometer DSA100 (Kruss, Germany).

Conventional methods for NA isolation and detection

Conventional methods for NA isolation and detection, such as the spin column method and qPCR, were employed to evaluate and validate the BSNFs-chip for pathogen and NA enrichment/isolation, along with the LRET assay for PCR-free detection. For the NA isolation, serial diluted HCT116 cells (from 1 × 104 to 1 × 100 cell ml−1) and SARS-CoV-2 culture fluid (from 0.96 × 104 to 0.96 × 10−1 PFU ml−1) were prepared in 1.5 mL microcentrifuge tubes. Genomic DNA, RNA, and viral RNA were directly extracted from 100 µl of the serial diluted samples and 100 µl of pathogens concentrated and isolated by the BSNFs-chip. This was achieved using QIAamp DNA Mini kit (Cat no. 51306, Qiagen, Hilden, Germany), QIAamp RNA Blood Mini kit (Cat no. 52304, Qiagen), and QIAamp Viral RNA Mini Kit (Cat no. 52906, Qiagen), respectively. The NAs were then eluted with 100 µL of elution buffer supplied in the kits and stored at −20 or −80 °C until further use. For NA detection, primer and probe sets were designed for each target NAs and mixed with 5 µl of eluted NAs to prepare PCR mixes. qPCR was performed for DNA of the ACTB gene (40 cycles; denaturation at 95 °C for 10 s, annealing at 60 °C for 20 s, and extension at 72 °C for 20 s) using Brilliant III Ultra-Fast SYBR Green qPCR Master Mix (Cat no. 600882, Agilent Technologies, Santa Clara, CA, USA). qRT-PCR was performed for RNA of 18 s rRNA gene (reverse transcription at 50 °C for 20 min and 40 cycles; denaturation at 95 °C for 15 s, annealing at 58 °C for 15 s, and extension at 72 °C for 30 s) using Brilliant III Ultra-Fast SYBR Green qRT-PCR Master Mix (Cat no. 600886, Agilent Technologies). qRT-PCR for viral RNA of the S gene of SARS-CoV-2 was performed (reverse transcription at 50 °C for 10 min and 45 cycles; denaturation at 95 °C for 5 s and annealing/extension at 60 °C for 30 s) using LightCycler Multiplex RNA Virus Master Mix (Cat no. 06754155001, Roche, Basel, Switzerland). The primers and probe used are listed in Supplementary Table 2. The amplified products were detected through SYBR green and Cy5 fluorescence signals using the CFX96 Real-Time PCR System (Bio-Rad, Hercules, CA, USA).

Pathogen and NA enrichment/isolation using the BSNFs-chip

Microfluidic chips, namely Flat-chip, PSNFs-chip, and BSNFs-chip, were utilized in two different strategies: a 2-Step method focused solely on pathogen enrichment/isolation without NA enrichment/isolation, and a 1-Step method allowing simultaneous enrichment/isolation of both pathogens and NAs, as depicted in Fig. 4a. We obtained the HCT116 cell line (KCLB No. 10247) from the Korean Cell Line Bank (Seoul, Korea) and heat-inactivated SARS-CoV-2 Culture Fluid (0810590CFHI) from ZeptoMetrix (Buffalo, NY, USA). To begin with, HCT116 cells were serially diluted from 1 × 104 to 1 × 100 cells ml−1, while SARS-CoV-2 culture fluid was diluted from 0.96 × 104 to 0.96 × 10−1 PFU ml−1. For the enrichment/isolation of pathogens and NAs, a volume of one milliliter from these serially diluted samples was combined with 100 mg of adipic acid dihydrazide (ADH; Cat no. 8.41689.0050, Merck Millipore, Billerica, MA, USA). The blend was then injected into the microfluidic chip’s internal channel through the inlet Tygon tubing using a syringe and syringe pump at a rate of 100 µl min−1. The microfluidic chip was left to incubate at room temperature for 15 min, which allowed for pathogen capture on the flat, porous, or biporous structured surface of the film. Post-incubation, debris and unreacted ADH were eliminated using an air-filled syringe, and any remaining residue was washed away with 1 ml of PBS and air.

In the 2-Step method, pathogen enrichment/isolation was achieved without NA enrichment/isolation. The concentrated pathogens were isolated using 100 µl of elution buffer with a pH of 10–11, at a flow rate of 25 µl min−1. Conversely, the 1-Step method involved the simultaneous enrichment/isolation of both pathogens and NAs. The microfluidic chips were filled with pathogen lysis buffer, consisting of 20 μl Proteinase K (Cat No. 19133, Qiagen, Hilden, Germany), 100 mM pH 8.0 Tris-HCl (Cat No. 15568025, Invitrogen, Carlsbad, CA, USA), 10 mM ethylenediaminetetraacetic acid (Cat No. AM9260G, Invitrogen), 1% sodium dodecyl sulfate (Cat No. AM9822, Invitrogen), 10% Triton X-100 (Cat No. T8787, Sigma-Aldrich), 50 mg ADH, and 10 μl RNase-free DNase (only for RNA) (Cat No. 79254, Qiagen), and incubated at 56 °C (for DNA) or room temperature (for RNA) for 15 min. This procedure facilitated pathogen lysis and NAs capture on the flat, porous, or biporous structured surface of the film. After the incubation, an air-filled syringe was used to remove pathogen debris and unused ADH from the reaction, and the remaining residues were subsequently washed with 1 ml of PBS and air. Finally, the concentrated NAs were isolated at a rate of 25 μl per minute using 100 µl of pH 10–11 elution buffer.

For the clinical use, 30 NP swab samples were collected from patients suspected of having COVID-19, of which 20 were clinically confirmed as positive and 10 as negative, to determine the clinical utility of the BSNFs-chip. In brief, a mixture was prepared with 200 µl NP swab samples, 200 µl of lysis buffer, 50 mg ADH, 10 µl RNase-free DNase, and added PBS to reach a total volume of approximately 650 µl. This mixture was then injected into the BSNFs-chip at a rate of 100 µl per minute using a syringe pump. After pathogen lysis and RNA enrichment by incubating at room temperature for 15 min, an air-filled syringe was utilized to clear out debris of clinical samples and unused ADH from the reaction. Subsequently, any remaining impurities were thoroughly removed using 1 ml of PBS and air. The concentrated RNA derived from the concentrated pathogens were efficiently collected at a flow rate of 25 μl per minute, using 100 µl of pH 10–11 elution buffer. It was confirmed that a higher NA yield was obtained at a flow rate of 25 µl min−1 compared to 50 µl min−1 on the BSNFs-chip (Supplementary Fig. 11). All the eluted NAs were stored at either −20 or −80 °C for future use. All used microfluidic chips were disposed of after a single use to avoid contamination risks, and the eluted NAs were stored at either −20 or −80 °C for future use. All primers and probes for the qPCR-based methods were synthesized by BIONICS (Seoul, Korea) and are listed in Supplementary Table 2.

Clinical specimens

The clinical specimens were collected from Asan Medical Center (Seoul, Korea) between March and November 2022. The median age of the COVID-19 positive participants was 61 years (interquartile range 48–66 years) with 45% male, while the COVID-19 negative participants had a median age of 60 years (interquartile range 45–64 years) with 50% male. All patients were enrolled upon confirmation of SARS-CoV-2 infection through nasopharyngeal (NP) RT-PCR. Following enrollment, weekly RT-PCR tests were conducted on respiratory samples, including NP swabs, saliva, or sputum, for up to 12 weeks. If RT-PCR results remained positive after this period, testing continued weekly until two consecutive negative results were obtained. The study protocol was reviewed and approved by the Institutional Review Board of Asan Medical Center (IRB-2022-1054). The research process adhered to the ethical standards for medical research involving human subjects. Participants provided written informed consent before taking part in the study, with the consent form explicitly stating that demographic and clinicopathological information would be used for academic research and potential publication. Participants were not offered any financial compensation for their participation. The study was designed to improve detection sensitivity regardless of patient sex or gender, sex and/or gender were not considered as influencing factors.

In this study, a total of 30 nasopharyngeal (NP) swab samples were used to validate the performance of the BSNFs-chip and PCR-free detection method. These clinical samples comprised of 20 samples from COVID-19 positive patients and 10 samples from patients suspected to have COVID-19, but were later confirmed as negative. All the clinical samples underwent heat inactivation at 60 °C for 30 min and were subsequently stored at −80 °C until they were used. This study was given ethical approval by the Institutional Review Board of the Asan Medical Center (IRB No. 2022-0297), and all the participants in this study provided informed consent. To isolate the viral RNA, 200 µl of each NP swab sample was used. The isolation was carried out with both the QIAamp Viral RNA Mini Kit (Qiagen) and the BSNFs-chip. In both methods, the viral RNAs were obtained using 100 µl of elution buffer and were then stored at −80 °C until they were further used for analysis or testing.

Synthesis of the lanthanide-doped nanoparticles (LnNPs)

The core LnNPs were synthesized through thermal decomposition of lanthanide acetate precursors. First, 0.4 mmol of Y(CH3CO2)3 (99.9%, Sigma-Aldrich) was mixed with 3 ml oleic acid (90%, Sigma-Aldrich) and 7 ml 1-octadecene (90%, Sigma-Aldrich), and the resulting mixture was heated to 150 °C for 60 min. The solution was then cooled down to 50 °C, and 1 mmol of NaOH (≥98%, Sigma-Aldrich) in methanol and 1.6 mmol of NH4F (≥99.9%, Sigma-Aldrich) in methanol (≥99.8%, Sigma-Aldrich) were added. The mixture was stirred, and the residual methanol was evaporated by heating the solution to 100 °C and keeping it under vacuum. To heat the solution in an argon environment, the vacuum-argon state was changed three times, and then the argon state was maintained. The resulting core nanoparticles were collected by centrifugation and washed after the resulting solution was heated to 300 °C, held for 1 h, and then cooled to room temperature. The resulting product was re-dispersed in cyclohexane (≥99%, Sigma-Aldrich) for later synthesis. The synthesis of each batch of nanoparticles followed a similar procedure to that used for the core nanoparticles. Ln(CH3CO2)3 (Ln = Y, Yb, and Tm, total 0.2 mmol) (99.9%, Sigma-Aldrich) and Y(CH3CO2)3 (0.1 mmol) were used for the synthesis of NaYF4@NaYF4,Yb,Tm and NaYF4@NaYF4,Yb,Tm@NaYF4, respectively.

Characterization of the LRET donor

The morphology of the LnNPs was analyzed on a JEM-2100F (JEOL Ltd., Japan) installed at Hanyang LINC3.0 Analytical Equipment Center (Hanyang University, Seoul, Republic of Korea) at an accelerating voltage of 200 kV. The XRD patterns of the LnNPs were characterized by an XRD-7000 diffractometer. The Fourier transform infrared (FT-IR) spectra of the LnNPs were obtained by using a Nicolet iS50 FT-IR spectrophotometer (Thermo Fisher Scientific Co., USA). The hydrodynamic diameter and zeta potential were measured by a Zetasizer Nano ZSP (Malvern Co., UK). The photoluminescence (PL) emission spectra were recorded by a spectrometer (Andor, Kymera 193i) and an intensified sCMOS detector (Andor, ISTAR-SCMOS-18F-73) under external excitation at 980 nm provided by a 980 nm laser diode (Changchun New Industries Optoelectronics Tech. CO., China, MDL-III-980-2W). The PL lifetime was measured using a photomultiplier tube detector (H10721-01; Hamamatsu, Shizuoka, Japan) attached to the spectrometer and a digital oscilloscope (Rhode & Schwarz, Munich, Germany, RTM3002) under excitation at 980 nm using pulsed laser (optical parametric oscillator (OPO) laser, Q-switched Nd-YAG laser, EKSPLA NT342). The PL emission wavelength was selectively measured using a bandpass filter (Semrock, ff-01-800/12-25) and shortpass filter (Semrock, ff-01-950/sp-25) placed in front of the detectors.

Surface modification of the LnNPs

The LnNPs (15 mg) were dissolved in tetrahydrofuran (≥99.9%, Sigma-Aldrich), and simultaneously, 50 mg of dopamine hydrochloride (≥99.9%, Sigma-Aldrich) was dissolved in water. The solutions were added to the flask and heated to 50 °C with vigorous stirring. After 5 h of incubation under an argon environment, hydrochloric acid (37%, Sigma-Aldrich) solution (1 M) was added, and amine-modified LnNPs were obtained by several washing steps. For preparing maleimide-modified LnNPs, the amine-modified LnNPs (1 mg) and sulfo-SMCC (sulfosuccinimidyl 4-(N-maleimidomethyl) cyclohexane-1-carboxylate, Thermo Fisher Scientific, Waltham, MA, USA) were dispersed in 10 mM HEPES buffer (pH 7.4) (1 M, Gibco). The resultant solution of maleimide-modified LnNPs was collected through several washing steps after 5 h of incubation.

To prepare DNA oligo-modified LnNPs were obtained using a thiol-maleimide click reaction. Free thiol-modified DNA was prepared using oligo in the disulfide form by the following method. To produce free thiol groups, 20 μl of 100 μM disulfide DNA (Genotech, Daejeon, Korea) was mixed with 20 μl of 5 mM tris(2-carboxyethyl)phosphine hydrochloride (TCEP, Sigma-Aldrich) and the mixture was reacted for 30 min at room temperature. Following this, DNA conjugation reaction was performed overnight at room temperature after adding the thiol-modified DNAs to the maleimide-modified LnNPs. The DNA oligo-modified LnNPs (denoted as LRET donor) were obtained by repeated centrifugation three times and dispersed in 150 μl of HEPES buffer. The S-gene was employed as the target gene sequence for LRET-based detection of SARS-CoV-2 in this study. We employed specific DNA oligonucleotides targeting the part of the SARS-CoV-2 S-gene, and these oligonucleotides had a short length, enabling successful energy transfer. Sequences of the LRET donor and acceptor are listed in Supplementary Table 6.

LRET-based viral RNA detection

We evaluated analytical sensitivity of the LRET assay using stock SARS-CoV-2 RNA solution isolated by QIAamp Viral RNA Mini Kit (Cat no. 52906, Qiagen). SARS-CoV-2 RNA solution (15 μl) were mixed with the LRET donor (2 μg) and the LRET acceptor (10 pmol, DNA modified IR800 dye) (Integrated DNA Technology, IDT, USA) in HEPES buffer (10 mM, pH 6.2) (total volume = 150 μl) and incubated at room temperature with 600 rpm shaking for 10 min. After the incubation, the PL intensities of the LRET donor in the mixture were measured by the intensified sCMOS detector under external excitation at 980 nm. In the presence of target RNA, the LRET donor and acceptor are brought into close proximity by oligo hybridization between complementary pairs, resulting in quenching of the LRET donor luminescence by the acceptor. From emission spectra, the relative intensity was calculated by the equation below:



I0 is the PL intensity of the LRET donor and Ix is the PL intensity after incubating with the LRET acceptor in presence of different concentrations of SARS-CoV-2 RNA.

The specificity of the LRET assay was determined using target RNA (SARS-CoV-2) and non-target RNAs including human coronavirus OC43 (hCoV-OC43), hCoV-229E, hCoV-NL63, and H3N2 Influenza A virus (IAV). The hCoV-OC43 and hCoV-229E RNAs were provided by the Korea Bank for Pathogenic Viruses (Seoul, Korea). The hCoV-NL63 RNA was provided by the National Culture Collection for Pathogens (Cheongju, Korea). H3N2 IAV (A/Brisbane/10/2007) RNA was provided by the Korea Research Institute of Bioscience and Biotechnology (KRIBB, Daejeon, Korea).

We validated the clinical applicability of the LRET assay using 30 clinical samples including 20 COVID-19 positive patients and 10 healthy controls. 15 μl of viral RNA samples isolated by the QIAamp Viral RNA Mini Kit (Qiagen) and the BSNFs-chip were mixed with the LRET donor (2 μg) and the LRET acceptor (10 pmol) in HEPES buffer (10 mM, pH 6.2) (total volume = 150 μl). Then, the LRET-based SARS-CoV-2 RNA detection was performed as described above. The cut-off value of relative intensity was determined by applying optimal combinations of clinical sensitivity and specificity from ROC curve based on the Youden index point.

Statistical analysis

Statistical analyses were performed using Origin Pro 2016 and IBM SPSS statistical package (version 27). The mean and ± standard deviations were calculated for each data point from at least triplicate measurement. LODs and linear ranges were determined using linear regression methods, which included assessing the line slope and the standard deviation of the intercept. The statistical significance of differences between SARS-CoV-2 positive samples and negative samples was assessed using a two-tailed unpaired t-test (*p  <  0.05, **p  <  0.01, ***p  <  0.001, and ****p  <  0.0001).

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.