key: cord-0702090-95bwh7gw
authors: Ahmad, Mohd; Sharma, Pooja; Kamai, Asangla; Agrawal, Anurag; Faruq, Mohammed; Kulshreshtha, Ankur
title: HRPZyme Assisted Recognition of SARS-CoV-2 Infection by Optical Measurement (HARIOM)
date: 2021-05-01
journal: Biosens Bioelectron
DOI: 10.1016/j.bios.2021.113280
sha: 3df8ce854da7997453e914fbf3f7fdb87c4cdad1
doc_id: 702090
cord_uid: 95bwh7gw

In order to define public health policies, simple, inexpensive and robust detection methods for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are vital for mass-testing in resource limited settings. The current choice of molecular methods for identification of SARS-CoV-2 infection includes nucleic acid-based testing (NAT) for viral genetic material and antigen-based testing for viral protein identification. Host exposure is detected using antibody detection assays. While NATs require sophisticated instrument and trained manpower, antigen tests are plagued by their low sensitivity and specificity. Thus, a test offering sensitive detection for presence of infection as a colorimetric readout holds promise to enable mass testing in resource constrained environments by minimally trained personnel. Here we present a novel HRPZyme Assisted Recognition of Infection by Optical Measurement (HARIOM) assay which combines specificity of NATs with sensitivity of enzymatic assays resulting in enhanced signal to noise ratios in an easily interpretable colorimetric readout. Using this assay, we could detect up to 10(2) copies of synthetic viral RNA spiked in saliva as a detection matrix. Validating our assay on suspected human subjects, we found concordance with PCR based readouts with visible colorimetric distinction between positive and negative samples in less than an hour. We believe that this assay holds the potential to aid in mass screening to detect SARS-CoV-2 infection by facilitating colorimetric detection with minimal resources and trained personnel.

transcripts in the same sample, hence enhancing sensitivity further. We demonstrate 96 the same by using primers targeting ORF-1ab and N-gene in a single-pot reaction. 97

The assay was able to detect 10 2 copies of viral RNA as determined by spiking copies 98 of viral RNA in saliva samples. Furthermore, we validated our method on COVID-19 99 patient samples from nasopharyngeal swabs or oral swab. We found matching 100 concordance with 2-gene RT-PCR and TrueNAT results and, thus we believe that 

For biosafety considerations, sample availability and ease of handling, we 182 decided to first test our reaction conditions for optimal peroxidase activity using 183 corresponding to beta-actin gene from human cDNA prepared from HEK 293T RNA 185 as a template using normal forward primer and reverse primer with or without 186 HRPZyme sequence. The PCR product was incubated first with hemin for 30 minutes 187 at room temperature in accordance with various published protocols followed by 188 incubation with HRP substrate, TMB and H 2 O 2 . Though in our hands, we couldn't 189 find any significant colorimetric difference between various test groups until 120 containing HRPZyme sequence, the color development process was very slow, as it 207 took around 3 hours for clear differences to emerge, rendering it ineffective for rapid 208 adapted from existing literature, (Li et al. 2016) where HRPZyme oligo was first 210 incubated with hemin for 30 minutes at room temperature to form a stable HRPZyme 211 complex followed by incubation with TMB substrate. We thought if we could 212 expedite the process by coincubation of hemin and TMB substrate with the PCR 213 product to take advantage of the peroxidase activity as soon as the HRPZyme starts 214 maturing. Upon incorporation of these changes in the protocol we found that 215 coincubation indeed accelerated the process significantly as we could now clearly see 216 stark colorimetric difference between the groups in as early as 120 minutes (Figure 217 

matrix 288

As the power of HARIOM lies in oligo-embedded enzymatic activity, we 289 wondered if we could further enhance the sensitivity by amplifying two genes instead 290 of just one. To test this, we used primers against ORF-1ab as well as N-gene. We 291 tested two different PCR protocols for this using cDNA from nasopharyngeal swab 292 preparation of SARS-CoV-2 positive patient, SARS-CoV-2 negative patient and non-293 template control (NTC). In one strategy, we used H-inv containing primers for both 294 genes together in one-step PCR for 35 cycles, in another strategy, we used non-H-inv 295 primers for first 10 cycles, followed by another PCR with this product using H-inv 296 containing primers for next 30 cycles. We observed a clean single band corresponding 297 to ORF-1ab as well as N gene in first strategy (both genes yielding almost similar 298 product size 106 and 108 bps) with no amplification either in SARS-CoV-2 negative 299 subject or NTC. These results were reflected in clean colorimetric distinction across 300 samples and in absorption spectra at 450 nm ( Figure 5A) . However, in two-step PCR 301 we observed a lot of background in SARS-CoV-2 negative and NTC samples, which 302 was also reflected in colorimetric readouts and absorption spectra at 450nm. (Figure  303 5B). Thus, we decided to follow the one-step PCR strategy moving forward. 304 305 During initial phases of testing, NP/oral swab suspended in VTM was 306 standard matrix that was used for viral testing using RT-PCR. However, the sample 307 collection method was inconvenient for the sample provider and at the same time 308 posed exposure risk to health personnel. It has previously been reported that saliva 309 can serve as a sensitive specimen for detection of respiratory viruses, (Kim et al. 310 biofluid for detection of SARS-CoV-2 viral RNA, and it was found to be a better and 312 more sensitive alternative to NP swabs. (Wyllie et al. 2020) 313 314 As salivary components have been shown to severely inhibit other colorimetric NAT 315 detection assay, such as RT-LAMP, we sought to explore if we could use saliva as a 316 specimen for our assay. We spiked serially diluted copies of viral synthetic RNA from 317 10 6 to 10 0 copies in inactivated saliva, performed cDNA synthesis and PCR, followed 318 by detection using HARIOM. While using N-gene alone, we were able to detect 10 3 319 viral copies (Figure 5C, 5D) . Using the protocol optimized above of using both ORF-320 1ab and N-gene in a single-pot reaction, we could bring down our detection limit to 321

10 2 viral copies, which corresponds to 2 viral copies/l of saliva, which is well within 322 the clinical limit requirements for detection ( Figure 5E, 5F) . (Supplementary Figure 3) . 352 353 J o u r n a l P r e -p r o o f spiked in saliva matrix, but for clinical SARS-CoV-2 detection, viral isolation was 355 still required which is technically challenging and time consuming. We sought to test 356 a recently developed RNA extraction free protocol for detection of nucleic acids from 357 saliva, that has been recently approved by FDA. We first tested the detection of beta-358 actin from saliva of four healthy individuals for presence of beta-actin RNA using 359 previously used primer set. We used 5 l of proteinase K treated saliva from healthy 360 individuals for cDNA synthesis followed by amplification of beta-actin genomic 361 region using reverse primer with or without HRPZyme sequence. We subjected the 362 PCR product to HARIOM and could observe a clear signal in samples containing 363

HRPZyme sequence, thus establishing that saliva can directly be used for RNA 364 extraction free detection of nucleic acids (Figure 6C, 6D) . In conclusion, we demonstrate that the modified HRPZyme sequence we used in 377 conjunction with our optimized protocol presented here holds great promise for 378 detection of nucleic-acids. HARIOM offers a powerful and generalized method for 379 detection of nucleic-acids from a variety of sample matrices, ranging from 380 nasopharyngeal swab, oral swab and direct saliva without the need for RNA isolation. 381

We also demonstrated its capability in detection of SARS-CoV-2 infection from 382 clinical specimen. We do acknowledge the limitation that the method presented here 383 still is dependent on availability of PCR instrument, but we believe that in future 384 iterations this could be taken to instrument-independent platforms, such as RPA, 385 further enhancing the sensitivity of reaction while marinating the simplicity of 386 colorimetric readout. One major limitation of the study that we feel is not being able 387 to test on enough clinical samples so as to accurately gauge sensitivity and specificity 388 as per clinical requirements, but as this is just a proof of principle demonstration of 389 the technique, we hope to address this in future studies. We believe that the biosensor 390 

International journal 415 of infectious diseases : IJID : official publication of the International Society 416 for Infectious Diseases 94

Department of Pathology

Nucleic 431

The authors gratefully acknowledge the financial support from Department of 396