key: cord-1016868-8x2oritb authors: Yin, Hao; Wu, Zhenhua; Shi, Nan; Qi, Yong; Jian, Xiaoyu; Zhou, Lin; Tong, Yigang; Cheng, Zule; Zhao, Jianlong; Mao, Hongju title: Ultrafast multiplexed detection of SARS-CoV-2 RNA using a rapid droplet digital PCR system date: 2021-05-04 journal: Biosens Bioelectron DOI: 10.1016/j.bios.2021.113282 sha: f802e87cb0075301895c4e8f5a21d676797ff809 doc_id: 1016868 cord_uid: 8x2oritb We report the first combination of droplet digital and rapid PCR techniques for efficient, accurate, and quantitative detection of SARS-CoV-2 RNA. The presented rapid digital PCR system simultaneously detects two specific targets (ORF1ab and N genes) and one reference gene (RNase P) with a single PCR thermal cycling period around 7 s and the total running time less than 5 min. A clear positive signal could be identified within 115 s via the rapid digital RT-PCR, suggesting its efficiency for the end-point detection. In addition, benchmark tests with serial diluted reference samples of SARS-CoV-2 RNA reveal the excellent accuracy of our system (R(2)>0.99). More importantly, the rapid digital PCR system gives consistent and accurate detection of low-concentration reference samples, whereas qPCR yields Ct values with significant variations that could lead to false-negative results. Finally, we apply the rapid digital PCR system to analyze clinical samples with both positive and control cases, where results are consistent with qPCR test outcomes. By providing similar accuracy with qPCR while minimizing the detection time-consuming and the false-negative tendency, the presented rapid digital PCR system represents a promising improvement on the rapid diagnosis of COVID-19. surface to enhance the heat exchange (Neuzil et al., 2006; Sposito et al., 2016) . Besides, indirect 115 heating methods such as infrared laser (Kim et al., 2009 ) and plasmonic photothermal heating 116 (Son et al., 2015) have been developed in addition to conventional direct Joule heating (Cai et al., 117 2019). Particularly, microfluidic devices are fabricated to perform rapid RT-qPCR for detecting 118 viral RNA within 30 min (Powell et al., 2018) . 119 On the other hand, RT-qPCR could yield false negative results when the concentration of 120 target RNA sequences is low or interfered by other sequences Yu et al., 121 2020). Applying rapid PCR strategies with RT-qPCR would further reduce the amplification 122 efficiency and detection sensitivity of target sequences (Fernandez-Carballo et al., 2018) . 123 Moreover, RT-qPCR tests are semi-quantitative depending on specific reagent kits and 124 instruments so the diagnosis of COVID-19 cases could vary due to different lab facilities and 125 operations. Although this problem could be mitigated by the introduction of international 126 reference materials that acts as calibration standards, such standards are not yet available for all 127 clinically important viral pathogens (Nixon et al., 2014) . Therefore, based on this existing gold 128 standard, we can also introduce more fast and accurate technique to further improve the detection 129 performance. 2.1. Rapid ddPCR system fabrication 152 The digital PCR part of the system is based on the microfluidic technology. And we use 153 microfluidic chips to generate, split, and store droplets as independent reactors for digital PCR. 154 Besides, the rapid PCR thermal cycling is enabled by in-suit heater arrays. The electrical 155 elements (heaters, temperature sensors and scribe lines) of the heater array are fabricated by lift- 156 off process, and the heating discs are made in a second lithography step (details of the heater 157 array and the microfluidic chip fabrication process can be found in the Supplementary Material). 158 The resistance of heaters and temperature sensors are tested using a semiconductor parameter 159 analyzer (Keithley 4200, USA). Subsequently, the heater array is soldered onto an interface PCB 160 (printed circuit board) (Supplementary Figure. 2c) Owing to the big surface-to-volume ratio of the chambers and the small volumes of PCR mix, the 207 temperatures of the sensor can be seen as identical to that of the PCR mix. The temperature 208 measured by the gold sensor in the storage chamber closely followed that from the platinum 209 sensors next to the heater array with a dynamic offset of less than 0.5 s. In a 2-s denaturing step 210 programmed by the heater control, the duration of the effective denaturation temperature 211 (between 90°C and 95°C) is measured for approximately 1.2 s in the storage chamber ( Figure. 212 1d), which is sufficient for the denaturation of double DNA strands (Wittwer and Garling, 1991) . 213 Therefore, the delay in temperature change due to the thin glass slide would not have dramatic 214 effect on the rapid PCR process. Nanoliter high throughput 568 quantitative PCR Ultra fast miniaturized real-time 570 PCR: 40 cycles in less than six minutes Comparative 572 Study of Sensitivity, Linearity, and Resistance to Polymerase Chain Reaction and Loop Mediated Isothermal Amplification Assays for 574 Quantification of Human Cytomegalovirus Rapid and sensitive detection of viral nucleic acids using 577 silicon microchips Ultrafast 579 photonic PCR Rapid real-time PCR and high resolution melt 581 analysis in a self-filling thermoplastic chip Dynamic pattern formation in a 583 vesicle-generating microfluidic device Diagnosing COVID-19: The 586 Disease and Tools for Detection Combination of RT-qPCR testing and clinical 588 features for diagnosis of COVID-19 facilitates management of SARS-CoV-2 outbreak Comparison of microfluidic digital PCR and conventional quantitative PCR for 592 measuring copy number variation Minimizing the Time Required for DNA 594 Amplification by Efficient Heat-Transfer to Small Samples Rapid Cycle DNA Amplification -Time and 597 Temperature Optimization The 599 TM) a microvolume multisample fluorimeter with rapid temperature control Virological assessment of 604 hospitalized patients with COVID-2019 A new coronavirus associated with human respiratory disease in 608 China Analysis of SARS-CoV-2 in Infected Patients A pneumonia outbreak associated with a new 616 coronavirus of probable bat origin Digital PCR on an integrated self-priming compartmentalization chip CoV-2 Viral Load in Upper Respiratory Specimens of Infected Patients Positive signals of ORF1ab are missing when performing rapid digital RT-PCR for 641 30 cycles with a cycle time of 3s (b) and 25 cycles with a cycle time of 5s (c). (d) Positive signals 642 of all targets are detected after 28 cycles with a cycle time of 2.8s Testing serial diluted SARS-CoV-2 RNA Reference Sample by rapid digital RT-646 And the dilution factor of 10 -3 , 5 ×10 -3 , 10 -2 , 10 -649 1 , 5 ×10 -1 is used to mark the concentration of the reference gene in each sample (c). No positive 650 droplets are observed in negative control samples. Correlation between detected target 651 concentration and actual concentration calculated by the dilution factor Error bars represent the standard deviation based on 653 at least 3 replicates of each experiment 5000 copies/test. (b) & (c) Ten detection amplification curves and Ct values for low-copy 660 concentration samples with concentration of (b) 10 copies/test (target: N) and (c) 5 copies/test 661 (target: ORF1ab) by RT-qPCR. (d) Ct values from 10 replicated testing of low-concentration 662 reference samples using RT-qPCR. By definition, Ct scales inversely with the log-scale 663 concentration of the target gene in the sample. (e) Accuracy of testing the low-concentration 664 reference samples by rapid digital RT-PCR (10 replicated tests)