key: cord-0852361-g0zc23p6
authors: Zhu, Xiong; Wang, Xiaoxia; Han, Limei; Chen, Ting; Wang, Licheng; Li, Huan; Li, Sha; He, Lvfen; Fu, Xiaoying; Chen, Shaojin; Xing, Mei; Chen, Hai; Wang, Yi
title: Multiplex reverse transcription loop-mediated isothermal amplification combined with nanoparticle-based lateral flow biosensor for the diagnosis of COVID-19
date: 2020-07-15
journal: Biosens Bioelectron
DOI: 10.1016/j.bios.2020.112437
sha: 6d3dbf549db45defaad4114f9720c23a3e1e6937
doc_id: 852361
cord_uid: g0zc23p6

The ongoing global pandemic (COVID-19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has become a huge public health issue. Hence, we devised a multiplex reverse transcription loop-mediated isothermal amplification (mRT-LAMP) coupled with a nanoparticle-based lateral flow biosensor (LFB) assay (mRT-LAMP-LFB) for diagnosing COVID-19. Using two LAMP primer sets, the ORF1ab (opening reading frame 1a/b) and N (nucleoprotein) genes of SARS-CoV-2 were simultaneously amplified in a single-tube reaction, and detected with the diagnosis results easily interpreted by LFB. In presence of FITC (fluorescein)-/digoxin- and biotin-labeled primers, mRT-LAMP produced numerous FITC-/digoxin- and biotin-attached duplex amplicons, which were determined by LFB through immunoreactions (FITC/digoxin on the duplex and anti-FITC/digoxin on the test line of LFB) and biotin/treptavidin interaction (biotin on the duplex and strptavidin on the polymerase nanoparticle). The accumulation of nanoparticles leaded a characteristic crimson band, enabling multiplex analysis of ORF1ab and N gene without instrumentation. The limit of detection (LoD) of COVID-19 mRT-LAMP-LFB was 12 copies (for each detection target) per reaction, and no cross-reactivity was generated from non-SARS-CoV-2 templates. The analytical sensitivity of SARS-CoV-2 was 100% (33/33 oropharynx swab samples collected from COVID-19 patients), and the assay's specificity was also 100% (96/96 oropharynx swab samples collected from non-COVID-19 patients). The total diagnostic test can be completed within 1 h from sample collection to result interpretation. In sum, the COVID-19 mRT-LAMP-LFB assay is a promising tool for diagnosing SARS-CoV-2 infections in frontline public health field and clinical laboratories, especially from resource-poor regions.

In late December 2019, an unexpected outbreak of COVID-19 caused by a novel 59 coronavirus called severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2, 60 also known as 2019-nCov) emerged in Wuhan, Hubei province, China (Xu et al. 61 2020) . By 11 May 2020, the SARS-CoV-2 virus caused a huge epidemic in China and 62 spread to over 140 countries/territories with 7,145,539 globally confirmed COVID-19 63 cases and 408,025 deaths (World Health Organization, COVID-19 Situation Report-64 142) (Lu et al. 2020 ). The current COVID-19 pandemic is a serious global health 65 concern because of its possibly fatal disease progression and rapid escalation of new 66 cases ). Therefore, in order to control COVID-19, it is necessary to 67 urgently develop new detection assays, which can provide an early, rapid and reliable 68 diagnosis of COVID-19. 69

The diagnosis of COVID-19 based on clinical symptoms, especially in the early 70 stages of disease, is extremely difficult as there are no initial characteristic symptoms 71 of COVID-19 (Huang et al. 2020) . Although genome sequencing is highly accurate 72 for COVID-19 diagnosis, its use in rapid large scale diagnosis is not practical because 73 of its long run time and requirement for expensive equipment . In the 74 current COVID-19 outbreak, real-time reverse transcription polymerase chain 75 reaction (rRT-PCR) was used to detect SARS-CoV-2 in public health and clinical 76 laboratories because of its high specificity and sensitivity (Corman et al. 2020) . 77

However, the total run time for rRT-PCR is several hours from clinical sample 78 collection to result reporting. rRT-PCR also requires complex equipment, skilled 79 personnel and a stable power supply. Together, this limits their use in some COVID-80 19 outbreak regions where there is insufficient infrastructure to perform rRT-PCR 81 such as field laboratories and other resource-limited settings (Cui and Zhou 2020 to an unreliable diagnostic result because the ORF1ab gene alone cannot ensure the 104 sufficient sensitivity for SARS-CoV-2 test (Suo et al. 2020 ). These studies also 105 utilized traditional monitoring techniques, such as agarose gel electrophoresis, SYBR 106 dyes and pH indicators, to detect COVID-19 LAMP products. These techniques can 107 detect any nucleic acids and are not specific for COVID-19 LAMP products. 108

Additionally, electrophoresis is time-consuming and tedious, while for colored dyes 109 and indicators, the judgment of color change in tubes by eye is potentially subjective 110 (Lee et al. 2020). Therefore, there is an ongoing need to develop new LAMP-based 111 assays that are capable of simultaneously detecting multiple SARS-CoV-2 targets to 112 provide a rapid and more objective result, as well as to facilitate simpler diagnostic 113

Here, a mRT-LAMP coupled with a nanoparticle-based lateral flow biosensor (LFB) 115 assay (mRT-LAMP-LFB) was developed for the diagnosis of COVID-19. Two target 116 sequences, including ORF1ab and the nucleoprotein gene (N), were simultaneously 5 amplified in an isothermal reaction and detected in one test step. We described below 118 the basic COVID-19 RT-LAMP principle and the optimized reaction parameters (e.g. 119 amplification temperature) as well as demonstrate the feasibility of this new method. 120 121 2. Materials and Methods 122

As shown in Figure 1C , the LFB contains four components (a sample pad, a 124 Thus, the LFB devised here can detect three targets (a chromatography control and 134 two target amplicons). The assembled LFB were cut into 4-mm dipsticks, and was 135 packaged in a plastic cassette (Jie-Yi Biotech, Shanghai, Beijing) according to the 136 manual. The LFB was stored dry at room temperature until use. 137

For reporting the COVID-19 mRT-LAMP results, a 0.5 aliquot of reaction mixtures 138 was applied to the sample region of LFB. Then, a 80 µL aliquot of BF (running buffer, 139 10 mM PBS, PH 7.4 with 1% Tween 20) also was deposited to the same region of 140 LFB, the LFB is able to absorb the whole RF. As a result, the presence of the COVID-141 19 mRT-LAMP products was indicated by the appearance of red lines on the reaction 142 region (NC membrane) ( Figure 1C and Figure S1 ). Usually, only 1 min was required 143 for reporting the amplification products using LFB. To ensure the reliability of 144 diagnosis test, we recommended a time of 2 min for COVID-19 mRT-LAMP-LAMP 145 assay during the result indicating stage. 146

Two RT-LAMP primer sets (ORF1ab-RT-LAMP and N-RT-LAMP) were designed 149 according to the LAMP mechanism using a specialized software (PrimerExplore V5) 150 according to the manual (http://primerexplorer.jp/e/v5_manual/index.html), which 151 targeted the ORF1ab and N gene of SARS-CoV-2 (GenBank MN908947, Wuhan-Hu-152 1) (Figure 1D) . A BLAST analysis of the GenBank nucleotide database was 153 performed for the ORF1ab-and N-LAMP primer sets to validate sequence specificity. 154

More details of primer design, locations, sequences and modifications are shown in 155 Figure 1D and Table S1 . All of the oligomers were synthesized and purified by 156 

In the LAMP system (Figure 1A) , FIP (forward inner primer) initiates the isothermal 215 amplification, and the new strand derived from FIP primer is displaced by the F3 216 (forward primer) synthesis (Step 1). Then, 2 primers, including BIP (backward inner 217 prime) and B3 (backward primer), anneals to the newly produced strand (Step 2). The 218 displacement polymerase Bst 2.0, then extends the sequence in tandem generating a 219 dumbbell shaped product (Step 3) . This stem-loop product can then serve as the 220 template for the second stage of the LAMP reaction (exponential amplification). The 221 LB* primer (backward loop primer), which is labeled with biotin at the 5' end, can 222 anneal to a distinct product derived from the exponential LAMP reaction stage (Step 223 4). The LB* product also serves as the template for the next amplification step by LF* 224 (forward loop primer), which is modified at the 5' end with hapten (Step 5). As a 225 result, a double-labeled detectable product (LF*/LB* product) is formed with one end 226 of the LF*/LB* product labeled with hapten, and the other end with biotin (Step 6, 7) . 227 A hapten is assigned to one primer set which allows for multiplex LAMP detection. respectively ( Figure 1B, step 1) . With the assistance of AMV (avian myeloblastosis 259 virus) reverse transcriptase, the RNA (SARS-CoV-2 template) was converted to 260 cDNA, and acted as the initial template for subsequent LAMP amplification ( Figure  261   1B, Step 2). After 40 min at 63°C, ORF1ab-LAMP products were simultaneously 262 labeled with FITC and biotin, and N-LAMP products with Dig and biotin ( Figure 1B,  263 Step 3). 264 265

As shown in Figure 1C , the result of COVID-19 mRT-LAMP assay was read out 267 using LFB. The details of LFB is shown in Figure 1C 

The optimal temperature for COVID-19 RT-LAMP was tested with 63°C shown to be 301 the best for COVID-19 RT-LAMP amplification (Figure S3 and S4) . Then, the 302 optimal isothermal amplification time for the COVID-19 mRT-LAMP-LFB assay was 303 also determined. It was found that the plasmid template at LoD (limit of detection) 304 level (12 copies), an incubation time of 30 min at 63°C produced positive results with 305 three red lines (TL1, TL2 and CL) (Figure S5) . For the RNA template detection, a 306 reverse transcription process (approximate 10 min) was essential, thus a COVID-19 12 mRT-LAMP reaction time of 40 min was recommended for the detection of clinical 308

samples. Therefore, the entire diagnostic process for COVID-19 mRT-LAMP-LFB 309 can be completed within 1 h with sample collection (3 min), rapid RNA extraction (15 310 min), mRT-LAMP reaction (40 min) and result interpretation (<2 min) (Figure 2) . 

Dilutions of templates (ORF1ab-plasmid and N-plasmid constructs) from 1.2×10 4 to 320 1.2×10 -2 copies were used to determine the sensitivity of COVID-19 mRT-LAMP-321 LFB. As shown in Figure 3 , the limit of detection (LoD) for our assay was 12 copies 322 per reaction for both the ORF1ab-plasmid and N-plasmid (Figure 3 ). The two target 323 genes were able to be detected and identified in a single-tube reaction ( Figure 3A) . 324

The COVID-19 mRT-LAMP results using LFB were consistent with turbidity and 325 VDR detection (Figure 3B and 3C) , however, it should be noted that traditional 326 visual monitoring techniques (VDR and turbidity) cannot differentiate target genes for 327 multiplex analysis. The sensitivity of COVID-19 mRT-LAMP-LFB assay was also 328 consistent with singleplex ORF1ab-and N-RT-LAMP assays (Figure 3, Figure S6  329 and Figure S7) . 

The COVID-19 mRT-LAMP-LFB showed 100% specificity with positive results 341 observed in all positive controls (120 copies each of ORF1ab-plasmid and N-plasmid) 342 (Table S2) . Similarly, a negative result was observed in all samples which contained 343 non-SARS-CoV-2 templates (synthetic nucleic acid sequences, virus, bacteria and 344 fungi). In these samples, only a single red band in the CL region was present. This 345 indicates no cross-reaction with non-SARS-CoV-2 templates (Table S2) . 346 347

Of the total of 129 respiratory samples, which were initially analyzed using rRT-PCR 349 

Transparency declaration: The lead author and guarantor affirms that the manuscript is an honest, 471 accurate, and transparent account of the study being reported; that no important aspects of the study 472 have been omitted; and that any discrepancies from the study as planned and registered have been 473 explained.

The Lancet

Available at SSRN

Infection and Drug Resistance 13

A multiplex reverse transcription loop-mediated isothermal amplification (mRT-LAMP) coupled with a nanoparticle-based lateral flow biosensor (LFB) assay (mRT-LAMP-LFB) was successfully established to rapidly and accurately diagnose COVID-19. 2. mRT-LAMP-LFB assy only requires simple heating equipment to maintain a constant temperature of 63°C for 40 min

The total diagnostic test can be completed within 1 h from sample collection to result interpretation

We would like to thank Prof. Ruiting Lan, Laurence Luu and Michael Payne (University of New South 476 Wales, Sydney, Australia) for linguistic assistance during the preparation of this manuscript.

tests to detect SARS-CoV-2 infections. Such detection techniques are required not 378 only in countries where SARS-CoV-2 infection is already spreading but also in 379 countries/regions where has not yet emerged. 380In this report, we developed a novel LAMP-based test that provides a simple, rapid 381 and reliable diagnosis for COVID-19, named COVID-19 mRT-LAMP-LFB. Our 382 assay integrates LAMP amplification, reverse transcription and multiplex analysis 383 with a nanoparticles-based biosensor, to facilitate the diagnosis of COVID-19 in a 384 one-step, single-tube reaction. Our test only requires a simple apparatus (e.g. a water 385 bath or a heat block) to maintain a constant temperature (63°C) for 40 min. Thus, 386 various portable user-friendly apparatus adapted for RT-LAMP amplification exist, 387 designed in our protocol is estimated to be $2 USD per test. Herein, we deem that a 393 COVID-19 RT-LAMP-LFB test would cost ~$5.5 USD per disposable, and only $500 394 USD or less for a dedicated instrument. In addition, compared with the published 395 COVID-19 RT-LAMP assays, our protocol directly analyzed the mRT-LAMP results 396 using LFB, which is a simple, objective, less error-prone and easy-to-use platform, 397 and avoids the requirements of complex processes (e.g. electrophoresis), special 398 reagents (e.g. pH indicators) and expensive instruments (e.g. real-time turbidity) 399 were specifically designed and recognize eight regions within the target genes 408 ( Figure 1D ). Our specificity tests revealed that no false-positive results were 409 produced from non-SARS-CoV-2 templates, and positive results were obtained from 410 positive control and SARS-CoV-2 templates (Table S2) . Moreover, two targets 411 (ORF1ab and N genes) could be simultaneously amplified and detected in a 'one-step' 412 RT-LAMP reaction, which further increased the assay's reliability. Our data also 413 demonstrated that 27 out of 33 COVID-19 patients were diagnosed as positive when 414only ORF1ab gene was employed as diagnostic marker, and 28 out of 33 patient 415 samples were detection with SARS-CoV-2 infection when only N gene was used as 416 diagnostic sequence (Table S3) . Particularly, all 33 COVID-19 patients were 417 diagnosed with SARS-CoV-2 infection by COVID-19 RT-LAMP-LFB assay, in 418 which ORF1ab and N genes were simultaneously used as the diagnostic markers 419 (Table S3) . Thus, our approach could effectively avoid any undesirable false negative 420 results from current COVID-19 RT-LAMP assays that can only amplify and detect a 421 single gene target (e.g. ORF1ab) (Lamb et al. 2020; Yu et al. 2020) . 422The sensitivity of our RT-LAMP-LFB assay is sufficient for the diagnosis of 423 COVID-19. Using our protocol, the limit of detection for COVID-19 RT-LAMP-LFB 424 assay was 12 copies each of the ORF1ab-plasmid and N-plasmid constructs, which is 425 consistent with the sensitivity test results generated from ORF1ab-RT-LAMP-LFB 426 and N-RT-LAMP-LFB singleplex detection (Figure 3 , Figure S6 and Figure S7) . 427The COVID-19 RT-LAMP-LFB duplex assay did not improve or decrease the 428 sensitivity when compared with the simplex assays (ORF1ab-RT-LAMP and N-RT-429 LAMP assays). For the detection of RNA templates extracted from respiratory 430 samples, our COVID-19 RT-LAMP-LFB assay was highly sensitive and specific and 431 able to correctly diagnose 100% (33/33) of SARS-CoV-2 clinical samples examined 432 by rRT-PCR and 100% (96/96) of samples from non-SARS-CoV-2 infected patients. 433The COVID-19 RT-LAMP LFB assay designed in this report has advantages, some 434 of which are shared with other LAMP-based diagnosis techniques. For example, the 435 COVID-19 RT-LAMP-LFB assay was less sensitive to various inhibitors, or was less 436 affected by the presence of various salts, or could tolerate the inhibitory effect of large

The authors declare that they have no competing interests.