Multiplexed Detection of Respiratory Virus RNA Using Optical pH Sensors and Injection Molded Centrifugal Microfluidics

doi:10.21203/rs.3.rs-4205382/v1

Download PDF

Article

Multiplexed Detection of Respiratory Virus RNA Using Optical pH Sensors and Injection Molded Centrifugal Microfluidics

https://doi.org/10.21203/rs.3.rs-4205382/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Infectious pathogens, such as the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), are a threat to global health and prosperity, with the coronavirus disease 2019 (COVID-19) pandemic causing deaths and negative economic impacts worldwide. Pathogens continuously mutate, evading vaccines and treatments; monitoring is therefore crucial to preventing future outbreaks. But there are still many shortcomings in available diagnostic technologies, and scalable and convenient point-of-care technologies are highly demanded. In this work, we demonstrate the application of injection molded centrifugal microfluidic chips with integrated optical pH sensors for multiplexed detection of SARS-CoV-2, influenza A, and influenza B RNA.

The optical pH sensors generated sensitive fluorescent readouts from diagnostic reverse transcription loop-mediated isothermal amplification (RT-LAMP) reactions; limits of detection for influenzas A and B, and SARS-CoV-2 of 89, 245, and 38 RNA copies per reaction, respectively, were attained. Results were obtainable within 44 minutes for SARS-CoV-2 and influenza A, and 48 minutes for influenza B. We implemented a data processing strategy that allowed for reliable, quantitative thresholds for deciding reaction outcomes based on numerical derivatives of the fluorescence curves, enabling 100% specificity. This work demonstrates the utility of optical pH sensors and injection molded centrifugal microfluidics for multiplexed infectious disease diagnostics with point-of-care applications.

Physical sciences/Optics and photonics/Applied optics/Optical sensors

Physical sciences/Engineering/Electrical and electronic engineering

Physical sciences/Chemistry

Physical sciences/Nanoscience and technology/Nanobiotechnology/Microfluidics

Infectious diseases are illnesses caused by infectious agents, including bacteria, parasites, fungi, and viruses.¹ Infectious diseases are responsible for substantial morbidity and mortality worldwide: in 2019, they caused 420 million lost disability-adjusted life years and nearly 8 million deaths (over 10% of deaths globally).² They are a continuing and growing problem. This can be attributed to, among others, the following factors: re-emergence of certain infectious diseases such as tuberculosis,³ antibiotic resistance in bacteria,⁴ the interplay between the necessity for expanding food production and disease emergence,⁵ and emerging infectious diseases of zoonotic origin.⁶ The impact of infectious diseases is underscored by the coronavirus disease 2019 (COVID-19) pandemic, presumably caused by the zoonotic transmission of a novel coronavirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).^7–9 SARS-CoV-2 was first identified in January of 2020 and in March of 2020 a global pandemic was declared.^10,11 Over 350,000 deaths worldwide were recorded by May of 2020, rising to over 6.5 million by October of 2022.^11,12 In addition to substantial loss of life, SARS-CoV-2 negatively impacted the world economy on a huge scale, highlighting the multi-faceted ways in which infectious diseases can harm society.^13,14 Diagnosing and monitoring infectious diseases is therefore important, since it promotes better individual treatments and improves public health.^15,16

The initial spread of the coronavirus can be attributed in part to a lack of efficacious interventional therapies available at the time.^17,18 The lack of treatments consequently made routine testing and diagnosis of infected individuals critical to monitoring and preventing the virus’ transmission and spread, especially because of the prevalence of asymptomatic carriers.^19,20 Hence, accurate and scalable diagnostic platforms were needed to mitigate the spread of SARS-CoV-2 and reduce mortalities. To achieve diagnostic capacity at the scale needed for containing future outbreaks, reliable and rapid point-of-care (POC) diagnostic methods are needed.²¹ While an unbridled level of resources and scientific enquiry was devoted to characterizing and combating SARS-CoV-2, there are still other diseases, such as influenza, which pose challenges to public health.^22,23 Therefore, emerging diagnostic technologies should also be multiplexed to screen multiple pathogens simultaneously.²⁴

Nucleic acid amplification tests (NAATs) are well suited to the detection of infectious pathogens.²⁵ The polymerase chain reaction (PCR), for example, has exceptional sensitivity and specificity.²⁶ Reverse transcription quantitative PCR (RT-qPCR), a variant of PCR used for detecting RNA, has been able to detect SARS-CoV-2 in patients with sensitivities of approximately 80% and specificities of 100%.^27,28 For comparison, widely used rapid antigen tests (RATs) had much lower sensitivities ranging from 23–71%.²⁹ Furthermore, assay design is relatively straightforward for NAAT’s, requiring only the sequence being targeted and the production of oligonucleotides recognizing that sequence.³⁰ In contrast, RATs require the integration of antibodies, which can be more difficult to produce, especially quickly.^31–33 However, a key drawback to RT-qPCR, and PCR in general, is that it requires expensive machinery and relatively high levels of expertise to perform.³⁴ Therefore, PCR is generally performed in centralized laboratories, meaning facilities may be overwhelmed in low-resource settings.²⁴ This can hamper public health and medical efforts to curtail outbreaks and treat patients, respectively. These technical limitations mean that POC alternatives to gold standard methods are needed, enabled by newer technologies such as isothermal amplification and microfluidics.³⁵

One route to implementing point-of-care NAATs is through microfluidics. Microfluidic devices are characterized by their handling of minute liquid volumes (on the order of microliters or less) using sub-millimeter sized structures, and have emerging applications in healthcare.^36–40 Microfluidics can make it easier to perform diagnostic assays by automating the setup of reactions and reducing expenses through minimization of reagent usage.^41,42 A promising class of microfluidics, especially for POC applications, is centrifugal microfluidics. Centrifugal microfluidics utilize rotational forces (i.e., centrifugal, Coriolis, and Euler forces) to move fluids through channels and into chambers.^43,44 The integration of passive valves to regulate fluid flow simplifies the fabrication of centrifugal microfluidic chips, and metering chambers enable precise aliquoting and distribution of fluids to separate chambers.⁴³ These microfluidics at minimum require only a motor for actuation.^45,46

While microfluidic chips can enable multiplexing of diagnostics reactions, the reactions themselves must be amenable to POC applications.^47,48 Isothermal amplification reactions, such as loop-mediated isothermal amplification (LAMP), are a promising alternative to PCR since they do not require thermocycling, making assay system design and construction more straightforward.^49–51 LAMP reactions can be performed on microfluidic chips and can detect DNA or RNA in the case of reverse transcription LAMP (RT-LAMP). ^52,53 LAMP reactions confer microfluidic assays with adaptability since they use readily synthesized oligomer primers as the specific detection reagents.^49,50 This means that diagnostic panels can be rapidly customized to detect a variety of emerging infectious agents and their variants. LAMP also generates substantial quantities of products and byproducts, which can be easily detected through various readout mechanisms.^54,55

Various methods have been used to generate readouts from LAMP reactions, detecting either the products (i.e., DNA) or byproducts (i.e., magnesium pyrophosphate and hydrogen ions).⁵¹ A prevalent method involves using pH indicators, either colorimetric or fluorescent, to determine pH changes in minimally buffered LAMP reactions.^51,56,57 In another work, we built upon these methods by utilizing fluorescent optical pH sensors to generate readouts from LAMP reactions to detect DNA fragments.⁵⁸ The optical pH sensors comprised fluorescein isothiocyanate (FITC) covalently immobilized within a poly(2-hydroxyethyl methacrylate) (pHEMA) matrix and were drop-casted as a precursor solution into microfluidic reaction chambers; they were used because of their advantages over conventional pH indicators.

Fluorescent optical pH sensors are non-invasive and have better sensitivity and selectivity than colorimetric indicators,^59,60 and they are amenable to quantitative readouts, obviating skewed color responses that can occur with colorimetric indicators.^61,62 Additionally, pH-sensitive fluorophores exhibit improved stability when conjugated to a polymer.⁶³ Fluorescent optical pH sensors have been utilized in microfluidic systems for cell culture^64–69, single cell experiments⁷⁰, enzymatic reaction monitoring^71–73, free-flow electrophoresis^74–77, and oral biofilm characterization⁷⁸, among other applications. In this study, we expand their applications by utilizing them for RNA-based RT-LAMP diagnostic methods in combination with centrifugal microfluidics.

Other groups have previously used centrifugal microfluidics for RT-LAMP-based detection of viruses. However, these studies used chips fabricated either with low-throughput prototyping methods, such as CNC machining, or through expensive injection molding services.^79–84 We previously developed variotherm desktop injection molding (VDIM), a technique which bridges the gap between conventional prototyping and mass production methods.⁵⁸ This helps overcome the high costs of commercial injection molding and the incompatibilities of designs prototyped using conventional methods with mass production.

In this work, we demonstrate the application of centrifugal microfluidics with integrated optical pH sensors for performing RT-LAMP reactions and distinguishing SARS-CoV-2, influenza A, and influenza B RNA on a single chip. The straightforward method of integrating the optical pH sensors into the microfluidic chips ensures the scalability of the technology. The sensors’ quantitative, automated outputs yield good assay reliability by minimizing user bias and obviating special training. This technological platform shows promising applications in clinical and domestic settings, where it could inform medical practitioners and laypersons of a disease’s presence and improve public health responses to future outbreaks.

Oligonucleotides

All oligonucleotides were ordered from Integrated DNA Technologies (IDT, United States) desalted and re-suspended at a concentration of 100 µM in water. 5X LAMP primer master mixes were made comprising 8 µM FIP, 8 µM BIP, 1 µM F3, 1 µM B3, 2 µM LF, and 2 µM LB primers in nuclease-free ultrapure water (Beyotime, China).

Optical pH Sensor Titration

Britton-Robinson buffers (BRBs) of varying pH levels were made with 40 mM H₃BO₃, H₃PO₃ and CH₃COOH, respectively, and their pH adjusted with 1 M NaOH. Fluorescein isothiocyanate (FITC, Sigma-Aldrich, United States) was covalently linked to poly-(2-hydroxyethyl methacrylate) (pHEMA, Sigma-Aldrich, United States) to make FITC-pHEMA optical pH probes as described before by Herzog et al.⁷⁴ Briefly, 500 mg of pHEMA was dissolved in 5 mL of N,N-dimethylacetamide (DMAA) and 0.8 mg of FITC was added. The mixture was stirred for 3 hours at 75°C and then cooled to room temperature. FITC-pHEMA was then precipitated with BRB pH 6.0 and washed three times with the same buffer. A 5% (w/w) precursor solution was made by dissolving the product in ethanol/water (9:1 v/v) by stirring for 4 hours at room temperature, before being stored at 4°C.

For the optical pH sensor titration, the precursor solution was diluted to 40% in water and 25 µL was drop-casted into separate wells of a 96-well black polystyrene plate (Greiner Bio-One, Germany). The plates were heated in an oven at 70°C for 10 minutes. The sensors were then covered with 150 µL of BRBs with varying pH levels. Fluorescent readings were taken for each well using a FlexStation 3 Multi-mode Microplate Reader (Molecular Devices, United States), with excitation at 490 nm and emission readings at 525 nm.

Microfluidic Chip Fabrication

Centrifugal microfluidic chips comprised two halves: an injection-molded top with microfluidic features and a CNC-milled bottom layered with a double-sided pressure sensitive adhesive (PSA). The injection-molded upper halves were made in-house using an HJK-12T (Haijiang, China) injection molding machine and PG-383 polystyrene pellets (Chimei, China). The design, fabrication, and usage of injection molds and the usage of the injection molder have been described more thoroughly elsewhere.⁵⁸ The bottom halves were made by layering a double-sided PSA (No. 5605, Nitto, Japan) onto a 0.75 mm thick white polystyrene sheet (Evergreen Scale Models, United States) and cutting the sheet into pieces approximately 40 mm by 50 mm in size. A CNC4030-2.2kW milling machine (Jingyan Instruments, China) was then used to cut these pieces into the shape of the microfluidic halves.

Microfluidic chip halves were post-processed with a CNC-mill and cleaned. Into each of the reaction chambers, 3 µL of FITC-pHEMA sensor precursor solution was drop-casted and dried at 70°C in an oven for 10 minutes. Once dry, any excess sensor layer on the periphery of the chambers or in the adjoining capillary valves was scraped away with a wet sterile toothpick. Primer mix was then deposited in the appropriate reaction chambers and dried at 35°C on a hot plate for 30 minutes. 2 µL of SARS-CoV-2, 0.7 µL of influenza A, and 1.3 µL of influenza B primer mixes were loaded per reaction chamber. Sodium polyacrylate granules (Macklin, China) were then placed in the waste chambers before sealing the chips. The chip halves were then aligned using an aluminum guide and hand-pressed together. They were then further sealed using a 3,000 kgf vertical press, with flat rubber O-rings placed on top of the chips during compression to distribute force from the platens to the edges of the chips.

Imaging rig design and construction

The imaging rig was designed using various optical and electronic parts for fluorescence excitation and emission, and its construction is described in detail elsewhere.⁵⁸ Briefly, the optics are comprised of 465 nm LEDs, a 455/30 nm bandpass filter, 2 mm optical grade plastic light guides, a 520 nm long-pass filter, and a 10MP USB camera mounted with an M12 wide-angle lens. These components were positioned with respect to one another using parts printed with polylactic acid (PLA) and acrylonitrile butadiene styrene (ABS) filament on a 3D-printer. The electronics are comprised of an Arduino microcontroller which controls the LEDs and the motor used for actuating the chip, and thermally regulates a custom-milled aluminum heat block.

Viral RNA controls

Amplirun SARS-CoV-2 RNA controls (MBC137-R) were purchased from Vircell (Spain). Influenza A (No. 103001) and influenza B (No. 103003) RNA controls were purchased from Twist Biosciences (United States). All RNA controls were diluted into nuclease-free water (New England Biolabs, United States) according to the stock concentrations provided by the manufacturers (i.e., 13,000 copies/µL for SARS-CoV-2 RNA and 1,000,000 copies/µL for influenza A and influenza B).

For simulated saliva samples, approximately 1 mL of saliva was collected from healthy donors 30 minutes after oral rinsing with water. Saliva aliquots were then spiked with 100X inactivation solution and heated at 95°C for 5 minutes. Inactivation solution was made by mixing 2.5 mL of 0.5M TCEP-HCL, 1 mL of 0.5M EDTA (pH 8.0), 1.15 mL of 5 N NaOH, and 0.35 mL of water.⁸⁵ RNA controls were then diluted along with inactivated saliva into RT-LAMP reactions.

RT-LAMP Assays

RT-LAMP reactions performed in a 384-well plate contained 1X master mix (10 mM ammonium sulfate, 10 mM potassium chloride, 8 mM magnesium sulfate, 0.1% Tween-20, 200 mM betaine monohydrate, pH 8.8), 5.6 mM dNTP mix (Invitrogen, United States), 1X SARS-CoV-2 LAMP primer mix, 0.3 units/µL RTx reverse transcriptase (M0380L, NEB, United States), and 0.32 units/µL Bst 2.0 DNA polymerase (M0537M, NEB, United States). 15 µL reaction volumes were incubated at 55°C for 15 minutes and 65°C for 35 minutes in a LightCycler 480 Instrument II Realtime PCR System (Roche Diagnostics GmbH). The wells of the plate were coated with optical pH sensor by drop-casting 10 µL of the precursor solution per well and heating at 70°C for 10 minutes prior to the assays.

On-chip RT-LAMP reactions contained 1X chip master mix (10 mM ammonium sulfate, 50 mM potassium chloride, 8 mM magnesium sulfate, 0.1% Tween-20, pH 8.8), 5.6 mM dNTP mix, 0.8 M betaine monohydrate, 0.3 units/µL RTx reverse transcriptase, and 0.64 U/µL Bst 2.0 polymerase. All chemicals except where otherwise indicated were purchased from Macklin, China. Varying numbers of RNA copies were used per reaction. Each chip was loaded with 105 µL of reaction mix and then rotated clockwise at 3,400 RPM for 10 seconds, linearly ramped up to 11,500 RPM over 3 seconds, and then held at that angular frequency for 5 seconds. Microfluidic chips were actuated using a 3.3V DC motor with an ABS 3D-printed rotor fixed to the shaft using cyanoacrylate adhesive. The motor’s speed was regulated by pulse-width modulated (PWM) voltage from an Arduino microcontroller and measured with a digital tachometer (20713A, Neiko, China).

After loading and actuating the centrifugal microfluidic chips with reaction mixture, they were sealed with a die-cut circular piece of double-sided tape (93005LE, 3M, United States). They were then inserted into the heat block of the imaging rig. A small pin was inserted between the chip’s edge and the heat block to ensure a secure fit. RT-LAMP experiments were initially incubated at 55°C for 10 minutes followed by 65°C for 50 minutes. A Python (v2.7.16) script was used to control data acquisition on the imaging rig. At 30 second intervals, commands were sent to the microcontroller to switch on the LEDs and to the camera module to photograph the chip. The LEDs were then turned off and the image was processed within the script. 70 pixel-wide regions of interest (ROIs) were defined within the reaction chambers and the background-subtracted average fluorescence of the ROIs was calculated. The values were tabulated for each image and exported to a CSV file.

For determining the limits of detection (LOD95) and 95% confidence interval of each viral RNA titration, the probit analysis in SPSS Statistics (version 29.0.1.0(171), IBM, United States) was used.

Data plotting and processing

RT-LAMP fluorescence curves were normalized with respect to the 12-minute timepoint, by which time the optical pH sensors’ fluorescence had stabilized. These normalized curves were then plotted in GraphPad Prism (v9.4.1), which was also used for all other plotting and statistical analyses aside from the probit analysis. Additional fluorescence data processing was performed in Python (v3.10.4) to correct for discontinuities in relative fluorescence data due to bubble movement. Discontinuities were corrected as follows: first, each fluorescence data point was subtracted from the one before it, yielding stepwise difference data. Each of these stepwise differences (d_i) was then compared to the preceding 8 data points according to the following formula:

$$\left|{d}_{i}-{D}_{med}\right|> 3*\left(med\right\{\left|{d}_{i-8}-{D}_{med}\right|,\cdots ,\left|{d}_{i-1}-{D}_{med}\right|\left\}\right)$$

where ${D}_{med}=med\{{d}_{i-8},\cdots ,{d}_{i-1}\}$ and $med\left\{\right\}$ indicates the median of a set of points. If d_i satisfied the above inequality, then that stepwise difference was considered anomalous. Correlating these anomalies to discontinuities in the fluorescence data, the aberrant fluorescence values were then replaced with extrapolated values from the previous 10 data points using a linear regression. All subsequent data points were shifted by the same amount as the corrected value. This was performed for all fluorescence curves. Subsequently, the differential of the repaired fluorescence curves was calculated via the total variation regulation algorithm using a numeric differentiation library (PyNumDiff, v0.1.2.4).⁸⁶ A threshold could then be applied to the resulting differential fluorescence curves to identify positive reactions in a quantitative manner.

RT-LAMP Readouts Using Optical pH Sensors

LAMP reactions yield large quantities of amplified DNA products along with byproducts of nucleotide incorporation: magnesium pyrophosphate and hydrogen ions (Fig. 1a).⁸⁷ In minimally buffered reactions, hydrogen ion release decreases pH which evokes fluorescence changes in optical pH sensors. By conjugating fluorescein isothiocyanate (FITC), a pH responsive fluorescent indicator, to poly(2-hydroxyethyl methacrylate), a hydrogel forming molecule and casting onto a microfluidic chamber, FITC-pHEMA (Fig. 1b) optical pH sensing layers were created. Performing a titration of the optical pH sensors confirmed that their fluorescent response spans the range over which pH changes occur in RT-LAMP (Fig. 1c), with a pK_a of 6.4.⁵⁶ The optical pH sensors were drop-casted into RT-LAMP reaction microwells and incubated with SARS-CoV-2 RNA controls. It was observed that RNA-loaded reactions showed a substantial decrease in fluorescence (Fig. 1d), confirming the utility of the optical pH sensors.

We also investigated the benefits of bi-thermal incubation strategies towards RT-LAMP performance. RT-LAMP reactions are typically performed isothermally but we previously found that bi-thermal incubations enhanced their sensitivity.⁸⁸ Centrifugal microfluidic chips were made with SARS-CoV-2 primers in each reaction chamber, except for three no-primer control (NPC) reaction chambers without primers (one NPC chamber per microfluidic arm). These chips were then loaded with (i.e., 100 copies per reaction chamber) or without SARS-CoV-2 RNA and incubated. For the isothermal reaction, incubation at 65°C for one hour was performed, whereas bi-thermal reactions were first incubated at 55°C for 10 minutes followed by 65°C for 50 minutes. Whereas the isothermal incubation did not yield robust amplification in any of the reaction chambers (Fig. 2a), the bi-thermal incubation encouraged substantial amplification within all the reaction chambers (Fig. 2b).

To verify that this enhanced amplification was not due to false positives, a bi-thermal incubation without template was also performed (Fig. 2c) which did not yield any amplification. Although isothermal amplification is a key advantage of LAMP, the inclusion of reverse transcriptase in RT-LAMP can hamper incubations at a single temperature which is optimized for Bst polymerase activity. Bi-thermal incubations are not substantially more technically complex to perform, since the setpoint of the incubator only needs to be programmatically adjusted. Given the improvements in sensitivity enabled by the bi-thermal incubation, we used this incubation scheme for all subsequent reactions.

Titration of Respiratory Virus RNA

For multiplexed detection of viral RNA, microfluidic chips were assembled by adhering an injection molded microfluidic half to a polystyrene substrate using a pressure sensitive adhesive (Fig. 3a). Primer sets were deposited into the separate reaction chambers of the microfluidic chips according to a specific layout (Fig. 3b) and allowed to dry before chip assembly. Each branch of the microfluidic chip targeted a separate viral RNA, with 4 replicates per virus. In addition, each branch contained one NPC chamber as a negative control. The primer sets were derived from other studies. The primer set for SARS-CoV-2 was obtained from Rabe and Cepko and targeted the ORF1a region.⁸⁵ The primer set for influenza A virus (IAV) was derived from one described by Jung et al., which targeted a conserved portion of the M gene.⁸⁰ The primer set for influenza B virus (IBV) was derived from a study by Ahn et al., which targeted a conserved portion of the NA gene.⁸⁹ The sequences for the influenza primers were adjusted slightly to match the strains used for the Twist RNA controls. The primer sets used for the RT-LAMP experiments are listed in supplementary table 1. Viral RNA controls were titrated to determine a limit of detection for the multiplexed chip assays. Fluorescence curves were thus obtained for chips incubated with SARS-CoV-2 (Supp. Figure 1), IAV (Supp. Figure 2), and IBV (Supp. Figure 3).

Positive reactions could be discerned visibly by their sharp decreases in fluorescence; however, it was difficult to apply fixed fluorescence cutoffs since the curves tended to differentially drift over time. To enable reliable determination of positive reactions, the fluorescence data was smoothed and differentiated using the total variation regulation (TVR) algorithm.^86,90 Differentiation mitigated the cumulative effects of fluorescence drifts and the fluorescence derivative plots maintained the qualitative outcomes of the original fluorescence plots. The fluorescence curves which visibly indicated a positive LAMP reaction by steep drops in fluorescence yielded differential curves which had sustained stretches of negative values. To quantitatively identify positive reactions, a horizontal threshold was set at -0.5×10^− 4 ΔRFU/second and to eliminate false positives, a temporal threshold was set at 44 minutes for SARS-CoV-2 (Fig. 4a, Supp. Figure 7). The same thresholds were then applied to IAV (Fig. 4b, Supp. Figure 8) and IBV reactions (Fig. 4c, Supp. Figure 9), with the temporal threshold being increased to 48 minutes for IBV reactions since they were generally somewhat slower. Beyond the temporal thresholds, any positive reactions were not considered true positives. These temporal thresholds compensated for spurious amplifications that occurred in some reactions (e.g., see Supp. Figures 8d, 9b), thereby improving accuracy. Using these thresholds, titration curves were generated, and 95% limits of detection (LOD95s) were determined with a probit analysis.

The LOD95 was 38 copies/reaction (CI 95%: 27–85 copies/reaction) for SARS-CoV-2 (Fig. 5a), 89 copies/reaction (CI 95%: 72–486 copies/reaction) for IAV (Fig. 5b), and 245 copies/reaction (CI 95%: 182–486 copies/reaction) for IBV (Fig. 5c). Assuming a purification strategy which enables five-fold concentration of viral nucleic acids and loading of 5 microliters of concentrate per reaction chamber, then detectable viral loads of 1.5x10³, 3.6x10³, and 9.8x10³ copies/mL could be achieved for SARS-CoV-2, IAV, and IBV, respectively. This is within the lower range of viral titers observed for these viruses⁹¹ and indicates that the multiplex assay has potential clinical utility.

Compatibility of RT-LAMP Reactions with Inactivated Saliva

To test the practicality of the on-chip viral RNA detection methodology, the compatibility of the RT-LAMP reactions with inactivated saliva samples was determined. Saliva is easy to collect and useful for viral diagnostics, making it suitable for POC diagnostics.⁹² However, in POC settings it is ideal to avoid extraction procedures, since this can make tests more complicated and less scalable.⁹³ In another work, saliva samples were directly used in RT-LAMP reactions by first mixing with a solution of TCEP and EDTA, and incubating at 95°C.⁸⁵ To test if this saliva inactivation procedure would work for reactions performed on the chips, saliva samples from three donors were collected 30 minutes after an oral rinse with water. These samples were then aliquoted into tubes, mixed with inactivation solution, and heat-treated. These treated samples were loaded into reactions along with quantities of each viral RNA that had previously yielded 100% sensitivity.

For each saliva sample, three plots were generated, one for each viral RNA. Both normalized fluorescence curves (Supp. Figure 10) and the numerical derivatives of these curves (Supp. Figure 11) were plotted to thoroughly assess outcomes. In contrast to the results obtained in the study where this methodology was tested, it was found that inactivated saliva was not always compatible with the RT-LAMP reactions.⁸⁵ Although one of the samples yielded results that were reasonably consistent with the reactions without saliva (Supp. Figure 11a), another sample completely inhibited the reactions within the incubation timeframe (Supp. Figure 11c). Comparing the saliva reactions against control reactions which did not use saliva (Supp. Figure 12), a statistically significant decrease in sensitivity was observed for all viral RNA’s (Fig. 6). In real-world settings this would entail false negatives and it can be concluded that inactivated saliva samples are not ideal for RT-LAMP reactions. Samples should be made consistent before loading into chips and extraction procedures should be incorporated into POC technologies. Nucleic acid isolation is therefore an important consideration since the benefits are twofold: it ensures samples do not inhibit reactions and can further improve sensitivities by concentrating low viral loads.

In this work, multiplexed detection of viral RNAs via RT-LAMP was achieved using optical pH sensors integrated with centrifugal microfluidic chips. The physical separation of primer sets into different reaction chambers enabled the discrimination of influenzas A and B and SARS-CoV-2 RNAs. RT-LAMP reactions elicited decreases in optical pH sensor fluorescence allowing reaction outcomes to be determined by setting thresholds for the numerical derivatives of the real-time fluorescence curves. We showed that saliva inactivation does not guarantee its compatibility with RT-LAMP reactions; technologies have been developed for POC nucleic acid purification and it would be worthwhile to adapt these strategies in the future.⁹⁴ The limits of detection obtained for the primer sets, in principle, enable clinical applications and testing of clinical samples will be performed to validate the utility of these microchips. In sum, the applicability of fluorescent optical pH sensors for infectious disease diagnostic applications was established and the usage of injection molded chips promotes scalability.

Acknowledgements

We would like to acknowledge the Research Grants Council and University Grants Committee for their support with funding through the Hong Kong PhD Fellowship Scheme (PF18-14492) as well as Grant R9375. We would also like to acknowledge the Hong Kong University of Science and Technology Entrepreneurship Center and the Hong Kong Science and Technology Park for their funding and support through the HKUST-Sino One Million Dollar Competition and STEP Programme, respectively.

Author Details

1: Department of Chemistry, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong

2:Quommni Technologies Limited, Tsuen Wan, New Territories, Hong Kong

3:Department of Chemical Pathology, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong

Author Contributions

G. D. S., P. P. C. and S. N. conceived and initiated the project and designed experiments, G. D. S., Y. Y. K. T. and S. B. performed experiments, G. D. S. and S. N. wrote the manuscript.

Conflicts of Interest

G. D. S., P. P. C. and Y. Y. K. T. co-founded a start-up company, Quommni Technologies Limited, to commercialize technologies involving centrifugal microfluidics. There are no other conflicts of interest.

Ethics Approval and Consent to Participate

The authors herewith confirm that the explicit consent of all parties was obtained for all saliva collected from human subjects for this study.

M. L. Barreto, M. G. Teixeira and E. H. Carmo, J. Epidemiol. Community Health, 2006, 60, 192–195.
GBD 2019 Collaborators, Global Burden of Disease Study 2019 (GBD 2019) Data Resources, 2019.
D. M. Morens, G. K. Folkers and A. S. Fauci, Nature, 2004, 430, 242–249.
B. Aslam, W. Wang, M. I. Arshad, M. Khurshid, S. Muzammil, M. H. Rasool, M. A. Nisar, R. F. Alvi, M. A. Aslam and M. U. Qamar, Infect. Drug Resist., 2018, 11, 1645.
J. R. Rohr, C. B. Barrett, D. J. Civitello, M. E. Craft, B. Delius, G. A. DeLeo, P. J. Hudson, N. Jouanard, K. H. Nguyen, R. S. Ostfeld, J. V. Remais, G. Riveau, S. H. Sokolow and D. Tilman, Nat. Sustain., 2019, 2, 445–456.
K. E. Jones, N. G. Patel, M. A. Levy, A. Storeygard, D. Balk, J. L. Gittleman and P. Daszak, Nature, 2008, 451, 990–993.
D. Singh and S. V Yi, Exp. Mol. Med., 2021, 53, 537–547.
E. C. Holmes, S. A. Goldstein, A. L. Rasmussen, D. L. Robertson, A. Crits-Christoph, J. O. Wertheim, S. J. Anthony, W. S. Barclay, M. F. Boni and P. C. Doherty, Cell, 2021, 184, 4848–4856.
M. Ciotti, S. Angeletti, M. Minieri, M. Giovannetti, D. Benvenuto, S. Pascarella, C. Sagnelli, M. Bianchi, S. Bernardini and M. Ciccozzi, Chemotherapy, 2020, 64, 215–223.
F. Wu, S. Zhao, B. Yu, Y.-M. Chen, W. Wang, Z.-G. Song, Y. Hu, Z.-W. Tao, J.-H. Tian and Y.-Y. Pei, Nature, 2020, 579, 265–269.
M. T. Adil, R. Rahman, D. Whitelaw, V. Jain, O. Al-Taan, F. Rashid, A. Munasinghe and P. Jambulingam, Postgrad. Med. J., 2021, 97, 110–116.
World Health Organization, COVID-19 weekly epidemiological update, edition 115, 26 October 2022, World Health Organization, Geneva PP - Geneva.
A. Nishi, G. Dewey, A. Endo, S. Neman, S. K. Iwamoto, M. Y. Ni, Y. Tsugawa, G. Iosifidis, J. D. Smith and S. D. Young, Proc. Natl. Acad. Sci. U. S. A., 2020, 117, 30285–30294.
T. Chen, G. Gozgor and C. K. Koo, Front. Public Heal., 2021, 9, 674729.
K. Messacar, S. K. Parker, J. K. Todd and S. R. Dominguez, J. Clin. Microbiol., 2017, 55, 715–723.
A. C. H. Yu, G. Vatcher, X. Yue, Y. Dong, M. H. Li, P. H. K. Tam, P. Y. L. Tsang, A. K. Y. Wong, M. H. K. Hui, B. Yang, H. Tang and L. T. Lau, Front. Med. China, 2012, 6, 173–186.
C. A. Omolo, N. Soni, V. O. Fasiku, I. Mackraj and T. Govender, Eur. J. Pharmacol., 2020, 883, 173348.
I. de A. Santos, V. R. Grosche, F. R. G. Bergamini, R. Sabino-Silva and A. C. G. Jardim, Front. Microbiol., 2020, 11, 1818.
S. H. Safiabadi Tali, J. J. LeBlanc, Z. Sadiq, O. D. Oyewunmi, C. Camargo, B. Nikpour, N. Armanfard, S. M. Sagan and S. Jahanshahi-Anbuhi, Clin. Microbiol. Rev., 2021, 34, e00228-20.
P. Khan, L. M. Aufdembrink and A. E. Engelhart, ACS Synth. Biol., 2020, 9, 2861–2880.
L. Magro, C. Escadafal, P. Garneret, B. Jacquelin, A. Kwasiborski, J.-C. Manuguerra, F. Monti, A. Sakuntabhai, J. Vanhomwegen and P. Lafaye, Lab Chip, 2017, 17, 2347–2371.
T. Carvalho, F. Krammer and A. Iwasaki, Nat. Rev. Immunol., 2021, 21, 245–256.
J. W. Tang, S. Bialasiewicz, D. E. Dwyer, M. Dilcher, R. Tellier, J. Taylor, H. Hua, L. Jennings, J. Kok and A. Levy, J. Med. Virol., 2021, 93, 4099.
O. Vandenberg, D. Martiny, O. Rochas, A. van Belkum and Z. Kozlakidis, Nat. Rev. Microbiol., 2021, 19, 171–183.
Y.-W. Tang, G. W. Procop and D. H. Persing, Clin. Chem., 1997, 43, 2021–2038.
J. E. Schmitz, C. W. Stratton, D. H. Persing and Y.-W. Tang, J. Clin. Microbiol., 2022, 60, e02446-21.
B. Clerici, A. Muscatello, F. Bai, D. Pavanello, M. Orlandi, G. C. Marchetti, V. Castelli, G. Casazza, G. Costantino and G. M. Podda, Front. Public Heal., 2021, 8, 593491.
T. C. Williams, E. Wastnedge, G. McAllister, R. Bhatia, K. Cuschieri, K. Kefala, F. Hamilton, I. Johannessen, I. F. Laurenson, J. Shepherd, A. Stewart, D. Waters, H. Wise and K. E. Templeton, Wellcome Open Res., 2021, 5, 1–12.
G. C. K. Mak, S. S. Y. Lau, K. K. Y. Wong, N. L. S. Chow, C. S. Lau, E. T. K. Lam, R. C. W. Chan and D. N. C. Tsang, J. Clin. Virol., 2020, 133, 104684.
P. Kralik and M. Ricchi, Front. Microbiol., 2017, 8, 108.
P. A. Zimmerman, C. L. King, M. Ghannoum, R. A. Bonomo and G. W. Procop, Pathog. Immun., 2021, 6, 135.
T. R. Kozel and A. R. Burnham-Marusich, J. Clin. Microbiol., 2017, 55, 2313–2320.
A. Chen and S. Yang, Biosens. Bioelectron., 2015, 71, 230–242.
D. Dutta, S. Naiyer, S. Mansuri, N. Soni, V. Singh, K. H. Bhat, N. Singh, G. Arora and M. S. Mansuri, Diagnostics, 2022, 12, 1503.
H. Harpaldas, S. Arumugam, C. Campillo Rodriguez, B. A. Kumar, V. Shi and S. K. Sia, Lab Chip, 2021, 21, 4517–4548.
G. M. Whitesides, Nature, 2006, 442, 368–373.
E. K. Sackmann, A. L. Fulton and D. J. Beebe, Nature, 2014, 507, 181–189.
C. D. Chin, T. Laksanasopin, Y. K. Cheung, D. Steinmiller, V. Linder, H. Parsa, J. Wang, H. Moore, R. Rouse, G. Umviligihozo, E. Karita, L. Mwambarangwe, S. L. Braunstein, J. van de Wijgert, R. Sahabo, J. E. Justman, W. El-Sadr and S. K. Sia, Nat. Med., 2011, 17, 1015–1019.
C. Rivet, H. Lee, A. Hirsch, S. Hamilton and H. Lu, Chem. Eng. Sci., 2011, 66, 1490–1507.
F. B. Myers and L. P. Lee, Lab Chip, 2008, 8, 2015.
S. K. Sia and L. J. Kricka, Lab Chip, 2008, 8, 1982–1983.
C. M. Pandey, S. Augustine, S. Kumar, S. Kumar, S. Nara, S. Srivastava and B. D. Malhotra, Biotechnol. J., 2018, 13, 1700047.
R. Gorkin, J. Park, J. Siegrist, M. Amasia, B. S. Lee, J. M. Park, J. Kim, H. Kim, M. Madou and Y. K. Cho, Lab Chip, 2010, 10, 1758–1773.
O. Strohmeier, M. Keller, F. Schwemmer, S. Zehnle, D. Mark, F. Von Stetten, R. Zengerle and N. Paust, Chem. Soc. Rev., 2015, 44, 6187–6229.
K. Abi-Samra, T.-H. Kim, D.-K. Park, N. Kim, J. Kim, H. Kim, Y.-K. Cho and M. Madou, Lab Chip, 2013, 13, 3253–3260.
R. Burger, L. Amato and A. Boisen, Biosens. Bioelectron., 2016, 76, 54–67.
X. Fang, H. Chen, S. Yu, X. Jiang and J. Kong, Anal. Chem., 2011, 83, 690–695.
X. Mao and T. J. Huang, Lab Chip, 2012, 12, 1412–1416.
T. Notomi, H. Okayama, H. Masubuchi, T. Yonekawa, K. Watanabe, N. Amino and T. Hase, Nucleic Acids Res., 2000, 28, 63e – 63.
K. Nagamine, T. Hase and T. Notomi, Mol. Cell. Probes, 2002, 16, 223–229.
T. J. Moehling, G. Choi, L. C. Dugan, M. Salit and R. J. Meagher, Expert Rev. Mol. Diagn., 2021, 21, 43–61.
H. Zhang, Y. Xu, Z. Fohlerova, H. Chang, C. Iliescu and P. Neuzil, TrAC Trends Anal. Chem., 2019, 113, 44–53.
R. Augustine, A. Hasan, S. Das, R. Ahmed, Y. Mori, T. Notomi, B. D. Kevadiya and A. S. Thakor, Biology (Basel)., 2020, 9, 182.
T. Iwamoto, T. Sonobe and K. Hayashi, J. Clin. Microbiol., 2003, 41, 2616–2622.
Y. Mori, K. Nagamine, N. Tomita and T. Notomi, Biochem. Biophys. Res. Commun., 2001, 289, 150–4.
N. A. Tanner, Y. Zhang and T. C. Evans, Biotechniques, 2015, 58, 59–68.
D. Yuan, J. Kong, X. Li, X. Fang and Q. Chen, Sci. Rep., 2018, 8, 1–8.
G. D. Suarez, S. Bayer, Y. Y. K. Tang, D. A. Suarez, P. P.-H. Cheung and S. Nagl, Lab Chip, 2023, 23, 3850–3861.
G. Orellana, Anal. Bioanal. Chem., 2004, 379, 344–346.
M. Shamsipur, A. Barati and Z. Nematifar, J. Photochem. Photobiol. C Photochem. Rev., 2019, 39, 76–141.
V. L. Dao Thi, K. Herbst, K. Boerner, M. Meurer, L. P. M. Kremer, D. Kirrmaier, A. Freistaedter, D. Papagiannidis, C. Galmozzi, M. L. Stanifer, S. Boulant, S. Klein, P. Chlanda, D. Khalid, I. Barreto Miranda, P. Schnitzler, H.-G. Kräusslich, M. Knop and S. Anders, Sci. Transl. Med., 2020, 12, eabc7075.
C. Uribe-Alvarez, Q. Lam, D. A. Baldwin and J. Chernoff, PLoS One, 2021, 16, e0250202.
D. Wencel, T. Abel and C. McDonagh, Anal. Chem., 2014, 86, 15–29.
A. Zanzotto, N. Szita, P. Boccazzi, P. Lessard, A. J. Sinskey and K. F. Jensen, Biotechnol. Bioeng., 2004, 87, 243–254.
K. S. Lee, P. Boccazzi, A. J. Sinskey and R. J. Ram, Lab Chip, 2011, 11, 1730–1739.
E. Poehler, S. A. Pfeiffer, M. Herm, M. Gaebler, B. Busse and S. Nagl, Anal. Bioanal. Chem., 2016, 408, 2927–2935.
S. Lladó Maldonado, P. Panjan, S. Sun, D. Rasch, A. M. Sesay, T. Mayr and R. Krull, Biotechnol. Bioeng., 2019, 116, 65–75.
H. Zirath, S. Spitz, D. Roth, T. Schellhorn, M. Rothbauer, B. Müller, M. Walch, J. Kaur, A. Wörle and Y. Kohl, Lab Chip, 2021, 21, 4237–4248.
B. Müller, P. Sulzer, M. Walch, H. Zirath, T. Buryška, M. Rothbauer, P. Ertl and T. Mayr, Sensors Actuators B Chem., 2021, 334, 129664.
X. Lin, W. Qiu, G. D. Suarez and S. Nagl, Chemosensors, 2022, 10, 379.
W. Zhan, G. H. Seong and R. M. Crooks, Anal. Chem., 2002, 74, 4647–4652.
W.-G. Koh and M. Pishko, Sensors Actuators B Chem., 2005, 106, 335–342.
S. A. Pfeiffer, S. M. Borisov and S. Nagl, Microchim. Acta, 2017, 184, 621–626.
C. Herzog, E. Beckert and S. Nagl, Anal. Chem., 2014, 86, 9533–9539.
S. Jezierski, D. Belder and S. Nagl, Chem. Commun., 2013, 49, 904–906.
E. Poehler, C. Herzog, C. Lotter, S. A. Pfeiffer, D. Aigner, T. Mayr and S. Nagl, Analyst, 2015, 140, 7496–7502.
C. Herzog, E. Poehler, A. J. Peretzki, S. M. Borisov, D. Aigner, T. Mayr and S. Nagl, Lab Chip, 2016, 16, 1565–1572.
M. P. Gashti, J. Asselin, J. Barbeau, D. Boudreau and J. Greener, Lab Chip, 2016, 16, 1412–1419.
K. G. de Oliveira, P. F. N. Estrela, G. de Melo Mendes, C. A. Dos Santos, E. de Paula Silveira-Lacerda and G. R. M. Duarte, Analyst, 2021, 146, 1178–1187.
J. H. Jung, B. H. Park, S. J. Oh, G. Choi and T. S. Seo, Biosens. Bioelectron., 2015, 68, 218–224.
L. Malic, D. Brassard, D. Da Fonte, C. Nassif, M. Mounier, A. Ponton, M. Geissler, M. Shiu, K. J. Morton and T. Veres, Lab Chip, 2022, 22, 3157–3171.
F. Tian, C. Liu, J. Deng, Z. Han, L. Zhang, Q. Chen and J. Sun, Sci. China Chem., 2020, 63, 1498–1506.
R. R. G. Soares, A. S. Akhtar, I. F. Pinto, N. Lapins, D. Barrett, G. Sandh, X. Yin, V. Pelechano and A. Russom, Lab Chip, 2021, 21, 2932–2944.
Y. Yao, X. Chen, X. Zhang, Q. Liu, J. Zhu, W. Zhao, S. Liu and G. Sui, ACS Sensors, 2020, 5, 1354–1362.
B. A. Rabe and C. Cepko, Proc. Natl. Acad. Sci. U. S. A., 2020, 117, 24450–24458.
F. Van Breugel, J. N. Kutz and B. W. Brunton, IEEE Access, 2020, 8, 196865–196877.
S. Purushothaman, C. Toumazou and C.-P. P. Ou, Sensors Actuators B Chem., 2006, 114, 964–968.
G. D. Suarez, D. A. Suarez, Y. Y. K. Tang, J.-X. Zhang, J. Li, S. Nagl and P. P.-H. Cheung, Anal. Methods, 2022, 14, 378–382.
S. J. Ahn, Y. H. Baek, K. K. S. Lloren, W.-S. Choi, J. H. Jeong, K. J. C. Antigua, H. Kwon, S.-J. Park, E.-H. Kim and Y. Kim, BMC Infect. Dis., 2019, 19, 1–12.
R. Chartrand, ISRN Appl. Math., 2011, 2011, 1–11.
D. Jacot, G. Greub, K. Jaton and O. Opota, Microbes Infect., 2020, 22, 617–621.
K. E. Kaczor-Urbanowicz, C. Martin Carreras-Presas, K. Aro, M. Tu, F. Garcia-Godoy and D. T. W. Wong, Exp. Biol. Med., 2017, 242, 459–472.
C. Myhrvold, C. A. Freije, J. S. Gootenberg, O. O. Abudayyeh, H. C. Metsky, A. F. Durbin, M. J. Kellner, A. L. Tan, L. M. Paul, L. A. Parham, K. F. Garcia, K. G. Barnes, B. Chak, A. Mondini, M. L. Nogueira, S. Isern, S. F. Michael, I. Lorenzana, N. L. Yozwiak, B. L. MacInnis, I. Bosch, L. Gehrke, F. Zhang and P. C. Sabeti, Science, 2018, 360, 444–448.
R. Paul, E. Ostermann and Q. Wei, Biosens. Bioelectron., 2020, 169, 112592.

There is a conflict of interest G. D. S., P. P. C. and Y. Y. K. T. co-founded a start-up company, Quommni Technologies Limited, to commercialize technologies involving centrifugal microfluidics. There are no other conflicts of interest.

ViralDiagnosticsSupplementaryVideoV3.mp4
Fluorescence Image Sequences of Viral RNA LAMP Assays on Centrifugal Microfluidic Chips
GDSuarezViralDiagnosticsSuppInfoSN11.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Multiplexed Detection of Respiratory Virus RNA Using Optical pH Sensors and Injection Molded Centrifugal Microfluidics

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

Oligonucleotides

Optical pH Sensor Titration

Microfluidic Chip Fabrication

Imaging rig design and construction

Viral RNA controls

RT-LAMP Assays

Data plotting and processing

Results and Discussion

RT-LAMP Readouts Using Optical pH Sensors

Titration of Respiratory Virus RNA

Compatibility of RT-LAMP Reactions with Inactivated Saliva

Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1