Uncoupling Molecular Testing for SARS-CoV-2 From International Supply Chains

Abstract

The rapid global rise of COVID-19 from late 2019 caught major manufacturers of RT-qPCR reagents by surprise and threw into sharp focus the heavy reliance of molecular diagnostic providers on a handful of reagent suppliers. In addition, lockdown and transport bans, necessarily imposed to contain disease spread, put pressure on global supply lines with freight volumes severely restricted. These issues were acutely felt in New Zealand, an island nation located at the end of most supply lines. This led New Zealand scientists to pose the hypothetical question: in a doomsday scenario where access to COVID-19 RT-qPCR reagents became unavailable, would New Zealand possess the expertise and infrastructure to make its own reagents onshore? In this work we describe a review of New Zealand's COVID-19 test requirements, bring together local experts and resources to make all reagents for the RT-qPCR process, and create a COVID-19 diagnostic assay referred to as HomeBrew (HB) RT-qPCR from onshore synthesized components. This one-step RT-qPCR assay was evaluated using clinical samples and shown to be comparable to a commercial COVID-19 assay. Through this work we show New Zealand has both the expertise and, with sufficient lead time and forward planning, infrastructure capacity to meet reagent supply challenges if they were ever to emerge.

Keywords: COVID-19; HomeBrew; RT-qPCR; molecular reagents; supply chain.

Figure 1

Interactive reagent calculator used to determine what, and in what quantity, reagents were required for a successful onshore production scheme. Not all components used in this reagent calculator contributed to the final HB RT-qPCR assay.

Figure 2

Overview of the synthesis of the dNTPs. *Purity was determined by UV peak area at 260 nm by reverse phase HPLC (details in text).

Figure 3

dNTP purity analysis by HPLC UV peak area at 260 nm. Method: C18 column (Agilent Poroshell 120 EC-C18, 2.7 μM, 100 ×4.6 mm), linear gradient of 0–15% MeCN in 50 mM aqueous triethylammonium acetate (pH 7.0) with 2 mM EDTA over 10 min at a flow rate of 1.0 ml min⁻¹. The retention times of the peaks (in minutes) are shown on the x axes, and the peak intensities in milli-absorbance units (mAU) at 260 nm are shown on the y axes.

Figure 4

DNA monomers used for the synthesis of DNA primers and probes. (A) Structure of a 5′-O-DMT protected nucleoside phosphoramidite used for an automated DNA synthesis. (B) Structure of a 5′-O-DMT protected nucleoside bound to the CPG-support and used as the first nucleotide at the 3′-end of the DNA sequence. (C) Structure of a DMT-protected BHQ-1 bound to the CPG support and used for the synthesis of DNA probes with BHQ-1 present at 3′-end. (D) Structure of fluorescein containing phosphoramidite used for installation of fluorescein at the 5'end of the DNA probe.

Figure 5

(A) SDS Page of Reverse Transcriptase enzyme: Lanes (1) abTES RT (2) abTES RT + protease digest (3) abTES RT + digest with inactivated protease (4) MashUp RT (5) MashUp RT + protease digest; (B) Cp in triplicate for serial dilution of MashUp RT compared to BioLine RT.

Figure 6

Purification and enzyme activity for HB Taq. (A) SDS Page showing 0.16 mg protein for MonoS fractionation prior to sample fractionation (Lane 1), the FT fraction (Lane 2), F1 (Lane 3) and F2 (Lane 4) fractions. (B) mAU readings for fractions FT, F1 and F2 from MonoS purification column. (C) Titration of HB Taq activity relative to a control enzyme using end-point PCR and gel image intensity: 0.2 μl enzyme (lane 1 and 5); 0.1 μl enzyme (lane 2 and 6); 0.05 μl enzyme (lane 3 and 7); 0.025 μl enzyme (lane 4 and 8).

Figure 7

(A) PCR performance comparison of HB dNTP product relative to a commercially manufactured equivalent over three log dilutions of template. (B) Effect of inclusion of HB RNase Inhibitor product on SARS-CoV-2 RT-qPCR for CDC N-gene at varying concentrations. (C) Effect of adding sacrificial DNA to the RT-qPCR reaction mixture to mitigate non-specific and premature reporter moiety cleavage. (D) A primer/probe (CDC N-gene) concentration matrix was used to determine optimal reagent concentrations.

Figure 8

PCR performance comparison of HB E-gene (A) and HB CDC N-gene (B) with a commercially synthesized hydrolysis probes.

Figure 9

Quality control assay for RNase inhibitor, silver staining (~2 μg protein per lane) of SDS-PAGE gel (A) and RNase inhibitor assay (showing % activity of RnaseA) of final product 1 month after purification (B). Lane 1: 1st re-suspended isoelectric precipitate with arrow indicating possible RNase inhibitor. Lane 2: 2nd re-suspended isoelectric precipitate. Lane 3: post-Sepharose binding sample. Possibly not all RNase inhibitor was captured. Lane 4: main peak of elution fraction 10–15. Lane 5: tail of elution (fraction 16–30). Lane 6: final product concentrated from the main fractions. Commercial RNase Inhibitor (Roche) was used as a positive control for the RNase inhibitor assay. By definition one unit of RNase inhibitor inhibits 5 ng of RNaseA activity by around 50%. In our case the amount of RNase A for the assay may have been overestimated.

Figure 10

Comparative amplification curves for 14 clinical samples using the Quanta commercial RT-qPCR reagents and the HB RT-qPCR kit to detect the SARS-CoV-2 N gene.