In-house reverse transcriptase polymerase chain reaction for detection of SARS-CoV-2 with increased sensitivity

As the COVID-19 infection continues to ravage the world, the advent of an efficient as well as the economization of the existing RT-PCR based detection assay essentially can become a blessing in these testing times and significantly help in the management of the pandemic. This study demonstrated an innovative and rapid corroboration of COVID-19 test based on innovative multiplex PCR. An assessment of optimal PCR conditions to simultaneously amplify the SARS-CoV-2 genes E, S and RdRp has been made by fast-conventional and HRM coupled multiplex real-time PCR using the same sets of primers. All variables of practical value were studied by amplifying known target-sequences from ten-fold dilutions of archived positive samples of COVID-19 disease. The multiplexing with newly designed E, S and RdRp primers have shown an efficient amplification of the target region of SARS-CoV-2. A distinct amplification was observed in 37 min using thermal cycler while it took 96 min in HRM coupled real time detection using SYBR green over a wide range of template concentrations. Our findings revealed decent concordance with other commercially available detection kits. This fast HRM coupled multiplex real-time PCR with SYBR green approach offers rapid and sensitive detection of SARS-CoV-2 in a cost-effective manner apart from the added advantage of primer compatibility for use in conventional multiplex PCR. The highly reproducible novel approach can propel extended applicability for developing sustainable commercial product besides providing relief to a resource limited setting.


Introduction
The sudden onset of an acute respiratory syndrome of unknown etiology among the people from Wuhan City, of China in December 2019, linked to a near local seafood market, appearedto be something novel. Later on, it was discovered that the causal agent of the infection was a Beta-coronavirus related to previously known members of the family SARS and MERS. The initial report indicated that SARS-CoV-2 was more distant from previously known SARS-CoV and MERS-CoV than from two bat-derived SARS-like coronaviruses bat-SL-CoVZC45 (87.9% sequence identity) and bat-SL-CoVZXC21 (87.2% sequence identity). The virus belongs to the family Coronaviridaeand order Nidoviraleswith an immense ability to mutate and undergo recombination [1][2][3][4][5]. SARS-CoV-2 belongs to the genus β-coronavirus which is comprised of crownlike, enveloped, positive-sense single-stranded RNA (þssRNA) viruses. The genome sequence length of SARS-CoV-2 is about 30 kb, with a 50 -cap structure and 30 -poly-(A) tail enveloped by a complex of structural proteins to form a crown-like, enveloped virus [1,7,8].
The novel 2019-nCoV causes an outbreak witha lower respiratory tract disease called novel coronavirus pneumonia (NCP), create a large-scale epidemic in a short time that immediately received worldwide attention. Subsequently,2019-nCoV was renamed as SARS-CoV-2 by the International Committee on Taxonomy of Viruses and disease as COVID-19 [6].
Polymerase Chain Reaction (PCR) techniques for viral detection and quanti cation offer the advantages of high reproducibility and sensitivity [9]. Initially, for the detection of COVID-19, a reliable qRT-PCR assay was developed by the Centers for Disease Control and Prevention (CDC) (USA). With time, different laboratories across the world developed different primer/probe pairs for speci c detection of SARS-CoV2 with varying sensitivity and accuracy. However in at least three independent studies reported that the primer pairs suggested for N region by US CDC, namely 2019-nCoV_N2 showed signi cant background cross reactivity [10], along with non-speci c ampli cation. Another primer 2019-nCoV_N3 had been reported to give false negative result [11] in a study that includes SARS-CoV-2 positive patients. The same primer had been reported to give false positive result even in the absence of template [10]. Similarly the most preferable primer RdRp-P2, used in more than 30 European laboratories had been reported to have less sensitivity [12]. With the growing need to screen a large number of COVID-19 cases, the present circumstances, driven the need for more sensitive and speci c primers. We go through a literature survey by selecting those articles with primer pair suggestions along with their most probable sensitivity limit or limit of detection (LOD), reported based on the analysis. Selecting the literature with sensitivity analysis data for the primer pairs prescribed by different public health agencies, which revealed that there is a difference in detection sensitivity at lower template concentration besides producing the false-positive result, arises the need for primer/probes with more speci city sensitivity and accuracy to detect COVID-19 cases. Apart from this, the most notable limiting factors are the cost of tests, trained human resources and availability of reagents in the resource-limited facility.
At present, a rapid and accurate diagnosis of the disease is of utmost importance. As per the World Health Medical College and Hospital and were used as the positive control. As such a total of 100 known positives and 33 known negative samples were used in the study.

Viral RNA extraction:
Viral RNA extraction was done using AuPreP Viral RNA Extraction Miniprep System (Life technologies, Cat no: RNV-52-906LT) following the manufacturer's protocol. The extracted RNA was then quanti ed using Nano Drop Spectrophotometer (NanoVue plus, Make: Invitrogen) and the nal RNA concentration along with A260/280 value was recorded. The RNA was converted into cDNA immediately after extraction.

Reverse Transcription
Tetro™ cDNA Synthesis Kit (Make: Meridian Bioscience, Cat. No: BIO-65043) was used for cDNA preparation using the manufacturer's protocol. 5µg of the extracted RNA was used for cDNA preparation in a nal volume of 20 µl reaction. Each reaction was incubated at 25 °C for 10 min followed by 45 °C for 30 min and nally terminated by incubating at 85 °C for 5 min and chilled on ice.

Primer design
The full sequences of SARS-CoV-2 were retrieved from the NCBI Reference Sequence Database. Alignment of the sequences was done in BioEdit software. Primer3 tool (https://bioinfo.ut.ee/primer3-0.4.0/) was used to design the three primer sets targeting the SARS-CoV-2-speci c E gene, S gene and RdRp gene. The selected primer pairs were analyzed using NCBI primer blast for the speci city of the primer pairs. Blast report showed that all the three primers have speci cally ampli ed the target region of SARS-CoV2 only.
Further, the primers pairs were analyzed for secondary structure, the amplicons as well as probable self and heterodimer formation tendencies using idtdna.com (https://sg.idtdna.com/pages/tools/oligoanalyzer). The primer sets were synthesized and delivered by Reprocell Brand: Bioserve (Hyderabad,Telangana. India).

Primer Testing through PCR
The accuracy and optimization of each primer sets were veri ed through PCR ampli cation. Gradient PCR of test was performed with an annealing temperature pro le ranging from 55°C to 68°C based on the melting temperature (Tm) of each of the SARS-CoV-2-speci c target primer set. The template was kept at a concentration up to a maximum of 10 ng of cDNAin a reaction volume of 20μlalong with forward and reverse primer sets at a nal concentration of 0.2-0.3 μM each using Emerald Amp GT PCR Master Mix (2x) from Takar Bio Inc (Cat No. RR310A). The cycling condition includes an initial denaturation step at 95 0 C for 5 min followed by 35 cycles of 95 0 C for 20 seconds, and annealing gradient for 30seconds, 72 0 C for 20seconds and a nal extension step at 72 0 C for 7 minutes.

Multiplex Traditional PCR ampli cation of E, S and RdRp gene
To minimize the amount of sample and reagent usage, preparation time, cost and labor, we developed an alternative protocol by adopting a multiplex PCR protocol for detection of SARS-CoV-2 in which all the primer sets were mixed into one reaction which reduced the total number of reactions reduced to 1(one) per sample instead of 3(three) reactions. When mixed, the nal concentration of the primer pairs was reduced proportionally to reduce the formation of primer-dimer. To visualize and separate each amplicon of each primer set, the designed primer pairs targeting the E, S and RdRP genes in the SARS-CoV-2 genome, produced an amplicon of different sizes (i.e., 101 bp, 103 bp, and 160 bp, respectively). The Applied Biosystems® Veriti® 96-Well Thermal Cycler instrument was used for multiplex PCR. EmeraldAmp® GT PCR Master Mix (Cat No: RR310A) was used and each 20 µl reaction mixture contained 10 µl of PCR Master Mix 0.5 µl of each primer and 1 µl of synthesized cDNA. The nal concentrations of primers were 0.2 µM (E,S and RdRp gene primers). The thermal cycler was set for initial denaturation at 95 °C for 5 min; followed by 40 cycles of PCR at 95 °C for 5 sec, 66 °C for 15 sec and 72 0 C for 5 minutes.

Sensitivity and Speci city of Multiplex PCR:
The sensitivity of the primer pairs was tested using multiplex PCR with optimized primer pairs for E,S and RdRp gene on archived known SARS-CoV2 positive and negative samples. Variable C t value of the samples ranging from 19 to 35 makes it possible to analyze the sensitivity of the primer pairs. Apart from this the template was diluted 10 times serially and tested for sensitivity and speci city of the primer pairs using multiplex PCR.

2.7.1Cross Reactivity analysis:
To analyse the cross reactivity of the assay procedure with other respiratory viruses, the HRM based multiplex real time PCR was tested against archived In uenza A positive, In uenza B positive, and H1N1positive samples. However, as an e cient alternative cross reactivity studies can be performed insilico where cross reactivity can be de ned as 80% or more sequence similarity between a primer and any nucleotide sequence in the target organism. [13] 2.8 Multiplex High Resolution Melting (HRM) based Real Time PCR for Detection of SARS-CoV2: 100 nanogram of previously synthesized cDNA from archived known SARS-CoV2 positive samples was used with TB Green® Premix Ex Taq™ II (Takara, RR820B) using Rotorgene Q (Qiagen) 5plex realtime PCR machine following cycling condition of an initial hold at 95°C for 5min; followed by a 45 repeat cycle of hold at 95°C for 20s, annealing at 66°C for 30s, nal extension at 72°C for 20s. High Resolution Melt pro le was set for a range of temperature from 55°C to 95°C with 0.1 0 increments at each step. Initial pre-melt for 90 second at rst step and hold at each temperature for 2 seconds was set. Standardized primer pairs for simultaneous detection of E, S and RdRP genes in the SARS-CoV-2 genome were used with a nal primer concentration of 200 nM.

Sequencing and Phylogenetic analysis:
PCR amplicons of E, S and RdRp gene were sequenced through Sanger sequencing (Bioserve sequencing service from Reprocell USA, inc, Hyderabad,Telangana. India). The obtained sequence was aligned separately for the three different gene targets against the SARS-CoV-2 reference sequence (NC_045512) and other related coronavirus sequences obtained from the NCBI database to analyse the percent identity.

Primer Optimization:
In-silico validation of primer pairs and amplicon sequences showed no secondary structure and also the possibility of self or heterodimer formation was not observed. In silico PCR tool nullify the possibility of nonspeci c reactions in the same genome as well as the genomes of different species. The optimized annealing temperature of primer pairs was found to be 66 0 C. The optimum primer concentration for multiplex PCR was found to be 0.2μM nal concentration of each primer. With this primer concentration, the resulting Agarose gel image showed no primer dimer formation (Fig: 1).

Traditional PCR protocol for SARS-CoV2 detection:
Following primer optimization, all three primers were tested separately using a single known positive sample with our optimized PCR cycling condition to observe the ampli cation e ciency of the protocol. Agarose gel electrophoresis showed a distinct band of size 101bp,103bp, and 160bp for E gene, S gene and RdRp gene respectively without any primer dimer formation (Fig: 2).

Development of a traditional Multiplex PCR protocol for SARS-CoV2 detection:
For fast detection and easy implementation in any biological laboratory in the world, we developed a traditional PCR protocol for SARS-CoV-2 detection. Archived positive sample from State Level Viral Research and Diagnostics Laboratory (VRDL) Gauhati Medical College and Hospital was used for cDNA preparation which was used as a template for multiplex PCR optimization. The PCR was performed for 40 cycles following the cycling condition described earlier. The gel electrophoresis showed two sharp bands of variable size i.e. 160 bp for the targeted RdRp gene and another two overlapping bands of 101 bp and 103 bp for the E gene and S gene of SARS-CoV2 respectively (Fig: 3).

HRM based detection of SARS-CoV2:
As an alternative to our traditional Multiplex PCR protocol for SARS-CoV2 detection, we further developed a multiplex Real Time PCR protocol with implementation of High Resolution Melting pro le in addition to the thermal cycling pro le for obtaining an enhanced separation of melt curve. A very similar results was obtained for multiplex HRM based RT PCR protocol. The ampli cation plot showed one combined reaction curve with signi cantly lower C t value for all the three primers put together apart from the ampli cation curve for the individual primer put in separate reaction tube (Fig:4). Further the accuracy of the multiplex reaction in real time was con rmed by analyzing the high resolution melt curve with three distinct peaks, (Fig:5) which was further compared with the melt curve peak obtained for the three primers when used separately in distinct tubes (Fig:6-9). Based on this observation we can conclude that the three distinct melt curve peak were of E, S and RdRp gene. Thus this result demonstrates that the newly developed multiplex HRM based real-time PCR protocol can be used for the fast and accurate detection of SARS-CoV-2 without the need of costly probe based approach of detection.

Sensitivity and Speci city:
Upon testing the primer pairs with known positive archived samples of SARS-CoV2 having Ct value ranging from 19 to 35 obtained with other commercial kits, our primer pairs able to detect all the samples without any false result. Further upon 10-fold serial dilution of the template, the primer pairs showed e cient ampli cation. Thus our HRM based Real Time Multiplex PCR protocol, besides being highly sensitive, the assay protocol is speci c (100%) and reproducible (100%) also. L.O.D. was determined by using seven 10-fold serial dilution of the clinical sample SC005G (4.71 x 10 5 RNA copies/ml), which was previously reported to have a low Ct value (Ct-ORF1b = 19, Ct-N gene =21; Meril COVID-19 detection kit). All the dilutions were replicated 5 times and the average Ct value obtained for individual target region for each dilution was shown in Table:1 All the three target were detected 100% of the replicates for dilution range from 10 -1 (4.71 x 10 4 RNA copies/ml)to 10 -5 (4.71RNA copies/ml) with a corresponding average Ct value of 34.28 for RdRp gene, Ct value of 33.77for E gene, and Ct value of 34.30 for S gene (Figure 10). At 10 -6 dilution, Target 1 and Target 2 were detected in 40% of the replicates with average Ct value of 37.42 for Target 1(RdRp gene) and 38.21 for Target 2 (E-gene). Target 3 (S-gene) was detected in 60% of replicates with average Ct value of 39.53 as shown in table. However, none of the target was detected in any of the replicates for 10 -7 dilution (Table :1)  (Table:2) ( Figure:11), which is representative of high viral load in this age group. Our 10-fold serial dilution study showed the assay LoD to be 10 -5 dilution (4.71 RNA copies) with corresponding Ct value of 34.28 for Target 1, 33.72 for Target 2, and 34.3 for Target 3. At 10 -1 dilution 100% of the specimens were detected for the entire three targets. As such, the risk of false negative result increases from 0% at 10 -1 dilution (approximately 11 Ct lower than LoD) to 95% at LoD, which increases further with increase in Ct value after LoD. In the studied cohort of SARS-CoV2 positive samples Mean Ct value for Target 1(24.53) & Target 2 (23.96) was 10 Ct bellow the LoD, and for Target 3,mean Ct (25.37) value was 9 Ct lower than the LoD value. Only 2/100 (2%) positive sample showed a slightly higher Ct value for only Target 3 (approx. 35) than LoD, whereas for Target 1& 2, the obtained Ct value were near the assay LoD, and as such would be at risk of false negative result. However, the assay e ciently detects 98/100 (98%) samples for Target 1 and Target 2 and as such supporting the enhanced sensitivity for RdRp gene (Target 1) and E gene (Target 2) detection (Figure 12:).

Comparison Of Ct Value
We further analysed the sensitivity and speci city of our own primer pairs by comparing the C t value obtained with our optimized detection protocol with the C t value obtained for the same set of sample using commercially available detection kits used in diagnostic setup (Meril COVID-19 One-Step RT-PCR Kit ; Cat no: NCVPCR-02). All the ve symptomatic con rmed SARS-CoV2 positive samples detected using Meril COVID-19 detection kit were also detected positive with our optimized protocol. Besides showing 100% speci city, the protocol also showed considerable decrease in overall Ct value for the multiplex PCR protocol as well as for individual primer when used in separate tube (Table: 3), which is representative of enhanced sensitivity of the economized SARS-CoV2 detection protocol.

Discussion
Nucleic acid testing is considered the gold standard for the detection of active infection. PCR-based viral detection techniques offer the advantages of high reproducibility and sensitivity apart from viral quanti cation [9]. In this study, we developed three sets of primers targeting E gene, RdRp gene and S gene of SARS-CoV2 with enhanced sensitivity and speci city apart from their applicability in both traditional as well as Real time based detection. Based on optimized primer sets, a cost-effective detection protocol for the virus was designed and developed. The primers were designed in such a way so that it becomes feasible to further extend the protocol for multiplex PCR and thus become applicable for both gel-based multiplex PCR assay and real time based detection. Owing to its highly sensitive primer pairs, the assay procedure offers versatility in its applicability in variable resource settings for e cient and cost-effective detection of SARS-CoV2. In India, various kits are being used for the detection of SARS-CoV-2. The rst detection kit was supplied by the National Institute of Virology (NIV), Pune, which used the SuperScript™ III One-Step RT-PCR System with Platinum™ Taq DNA Polymerase (make Invitrogen Catalogue number: 12574026) that takes about 5-6 hours for con rmation of the result. In due course of time, many COVID-19 RT-PCR kits have become commercially available targeting different major targets of theSARS-CoV-2genome. The time required for completion of tests of these commercially available kits varied from 1.5 to 3 hours after sample preparation and RNA extraction. Our optimized assay procedure offers simultaneous detection of three genes (E, S and RdRp gene) from SARS-CoV2 within a single tube reaction. Cost of test is a signi cant limiting factor while considering implementation of Nucleic Acid Testing for the detection of SARS-CoV2 apart from trained manpower [13]. With most of the presently available assay procedure, the estimated cost is around $ 51 (Centres for Medicare and Medicaid Services, COVID-19 test pricing. 2020) [14] is equivalent to INR 3,794.00. As such these detection assays possess several constraints both in terms of reagent limitation and cost involvement. However, with our detection assay involving implementation of high resolution melting analysis following PCR ampli cation using Sybr green master mix further reduces the cost signi cantly by eliminating the need of costly probe based detection. The present study protocol reduces the cost of tests per sample to almost INR 900. Besidesthe adoption of the sample pooling protocol further, contributes towards the reduced cost of test by mitigating the limitation of the shortage of chemical.
However, the main disadvantage in this process is the probability of reduced ampli cation e ciency due to the unavailability of PCR components for multiple primer sets used together in a single tube. In the present study, PCR protocol is optimized to minimize the competition among the primers sets for limited resources and achieved enhanced ampli cation e ciency. The current study revealed an extremely simple method of accurate detection of the SARS CoV-2viral entity, applicable even with resource limited laboratory settings and could be immensely helpful in the management of a robust pandemic like COVID-19 and lends us aware of a future emergency.   Ampli cation graph of E, S and RdRp gene of SARS-CoV2 using SYBR green master mix (TB Green® Premix Ex Taq™ II, Takara, Cat. No: RR820B). All the three primers were tested individually as well as in multiplexing reaction in single tube. Ampli cation plot showed e cient ampli cation of the three gene individually and also upon multiplexing of the three primers.

Declarations
Page 16/19   Individual melt curve peak for RdRp gene obtained upon single-plex RT-PCR reaction using primer pairs for RdRp gene.

Figure 8
Individual melt curve peak for S gene obtained upon single-plex RT-PCR reaction using primer pairs for S gene.

Figure 9
Individual melt curve peak for E gene obtained upon single-plex RT-PCR reaction using primer pairs for E gene.

Figure 11
Ct value comparison on positive sample strati ed by age.Age group 80-89 showed considerably lower mean Ct value for all the three target in comparison to overall Mean Ct which (p<0.05) is representative of high viral load.

Figure 12
Distribution of experimentally obtained Ct values for the 100 positive specimen for Target1, Target 2,& Target 3 are plotted. Only 2/100 (2%) positive sample were found with Ct values slightly higher than LoD or near the assay LoD and as such the sensitivity of detection may be less than 100%. Remaining 98 specimens showed e cient detection of the entire three targets.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.