key: cord-0999929-o225l0p4
authors: Chan, S. K.; Du, P.; Ignacio, K.; Metha, S.; Newton, I.; Steinmetz, N. F.
title: Biomimetic Virus-like Particles as SARS-CoV-2 Positive Controls for RT-PCR Diagnostics
date: 2020-10-20
journal: nan
DOI: 10.1101/2020.10.16.20213991
sha: 87b8672f33b2628387772ffbdf58238c842ca024
doc_id: 999929
cord_uid: o225l0p4

Coronavirus disease 2019 (COVID-19) is a highly transmissible disease that has affected more than 90% of the countries worldwide. At least 17 million individuals have been infected, and some countries are still battling first or second waves of the pandemic. Nucleic acid tests, especially reverse-transcription polymerase chain reaction (RT-PCR), have become the workhorse for early detection of COVID-19 infection. Positive controls for the molecular assays have been developed to validate each test and to provide high accuracy. However, most available positive controls require cold-chain distribution and cannot serve as full-process control. To overcome these shortcomings, we report the production of biomimetic virus-like particles (VLPs) as SARS-CoV-2 positive controls. A SARS-CoV-2 detection module for RT-PCR was encapsidated into VLPs from a bacteriophage and a plant virus. The chimeric VLPs were obtained either by in vivo reconstitution and co-expression of the target detection module and coat proteins or by in vitro assembly of purified detection module RNA sequences and coat proteins. These VLP-based positive controls mimic SARS-CoV-2 packaged RNA while being non-infectious. Most importantly, we demonstrated that the positive controls are scalable, stable, and can serve broadly as controls, from RNA extraction to PCR in clinical settings.

underserved areas or regions with tropical climates. While plasmid DNA is more stable than RNA, neither positive controls enables concurrent processing with patient samples, or full-process control. Because RNA/DNA samples are only added downstream in the testing process, they cannot control for whether the viral RNA was degraded during the initial extraction steps.

To overcome these shortcomings, we propose a biomimetic nanotechnology solution, i.e.

to package RNA transcripts containing the nucleic acid regions for binding of the SARS-CoV-2 primers and probes into a nanoparticle carrier. More specifically, we generated a biomimetic positive control by developing a virus-like particle (VLP) technology that takes advantage of the unique features of SARS-CoV-2 but is non-infectious and safe to use in diagnostic assays. While many nanotechnology platform technologies are available, including polymer and lipid nanoparticles, that could carry nucleic acids, the capsids from viruses naturally evolved to package nucleic acid and thus make a suitable choice. Specifically, biomimicry was achieved by packaging non-infectious, replication-deficient synthetic SARS-CoV-2 RNA target sequences into noninfectious VLPs from the bacteriophage Qbeta (Qβ) and the plant virus cowpea chlorotic mottle virus (CCMV). Both VLPs measure ~30 nm and have been extensively studied in nanotechnology. 20 Encapsidation into a viral capsid more closely mimics the conditions encountered by the RNA template of SARS-CoV-2 within clinical or environmental samples. This approach has been reported previously: for example, cowpea mosaic virus (CPMV) capsids were used to encapsidate target RNA for detection of foot-and-mouth disease virus (FMDV), 21 tobacco mosaic virus (TMV) was developed as a positive control for Ebola diagnostic assays, 22 and bacteriophage Qβ was explored as a control for foodborne virus detection. 23 In response to COVID-19, Asuragen and SeraCare have announced developments of SARS-CoV-2 positive controls in which RNA targets is encapsidated into bacteriophage VLPs 24 or a replication-deficient mammalian virus. 25 Of note is also the development of VLPs for vaccine development; [26] [27] research from academia and industry has led to development of more than 200 vaccine candidates in record time, many design rely of nanotechnology concepts and various VLP subunit vaccines are being developed. 28, 29 We developed a synthetic SARS-CoV-2 Detection Module (SDM). The positive controls contain synthetic SDM that contains all nucleic acid regions for binding of the CDC-designated RT-qPCR primers and probes. The SDM module was then was encapsidated into VLPs by in vitro or in vivo reconstitution of chimeric VLPs, yielding three SARS-CoV-2 positive controls termed Qβ 1P-C19, Qβ 2P-C19 (synthesized from a one or two plasmid system, respectively), and CCMV-C19. While in vitro reconstitution of viral capsids around a synthetic RNA template is well established (and was the method used for CCMV), the in vivo reconstitution is less frequently reported. However, the latter requires fewer processing steps and thus may be higher yielding and more economic to mass produce. To achieve efficient in vivo reconstitution, careful design considerations need to be taken in account, such as length of the target RNA and its molecular features, i.e. deletion of ribosome binding sites (for enhanced safety) and appendage of Qβ hairpin (for encapsidation). 30 Here we developed such methods for the Qβ VLP system. The structural integrity and stability of the resulting nanoparticle positive controls were assessed over time using a combination of transmission electron microscopy (TEM), size exclusion chromatography (SEC), dynamic light scattering (DLS) and electrophoretic mobility assays. A set of RT-PCR assays was performed to validate the biomimetic SARS-CoV-2 positive controls. Lastly, the positive controls were subjected to screening and used as external controls for patient sample testing.

We chose to develop VLPs from bacteriophage Qβ and plant virus CCMV as biomimetic nanotechnology for use as positive control probes for COVID-19 diagnostic assays. Qβ and CCMV offer several advantages over current approaches that use either MS2 VLPs (Asuragen) 24 and replication-defective mammalian virus (SeraCare). 31 Firstly, the Qβ capsid is 5Å larger and contains approximately 20% more genomic RNA compared to MS2. 32 Therefore, Qβ can accommodate a higher payload; and indeed Qβ can package 100 x copies of target RNA compared to MS2. 23 Furthermore, despite extensive sequence identity between Qβ and MS2, the coat protein subunits of Qβ offer higher thermal stability compared to MS2; the Qβ capsid gains thermal stability based on inter-subunit disulfide bonds. 33 Previous studies comparing Qβ and MS2 with comparable RNA cargo have shown that Qβ is more stable over a range of temperatures (-20°C, 4°C, and 45°C). 23 Taking together, data suggest that Qβ would be a better candidate for long term storage or ambient shipping.

To the best of our knowledge, CCMV has never been reported for development and application of positive control in molecular diagnostic assays. A particular advantage of the CCMV system is the straightforward in vitro reconstitution -hence offering a high degree of modularity. Purified coat proteins could be stored and reconstituted around a target RNA cassette as needed; for example, if mutants or new strains or emerge or to adapt the probe for use in other diagnostic assays. CCMV coat proteins could be obtained through heterologous expression 34 or through molecular farming in plants. The latter offers a high degree of scalability and speed, and All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020. 10.16.20213991 doi: medRxiv preprint can be implemented with relatively non-sophisticated infrastructure while keeping manufacturing costs low. 35 Furthermore, the bacteriophage and plant VLPs (Qβ and CCMV) offer advantages compared to a replication-deficient mammalian virus, which always carries a risk of residual activity or reversion, which may pose safety risks. [36] [37] Qβ and CCMV are non-infectious to mammals and unable to replicate in mammalian cells; 38 therefore offering another layer of safety in particular for use in low-resource settings where sterile and biosafety facilities are not always attainable. Moreover, production of Qβ and CCMV through fermentation or molecular farming is more cost effective and higher yielding compared to manufacture of mammalian virus in mammalian cell cultures. 39

The design of the SDM was based on the CDC-recommended detection regions. The 622 nt SDM consists of 4 segments with the first being a Qβ hairpin loop. This 29-nt Qβ hairpin has high affinity for the Qβ coat proteins 40 and is appended to the targeted RNA sequences to facilitate SDM encapsidation into Qβ VLPs. In previous work, the Qβ hairpin loop has been used to direct the encapsidation of protein cargos such as enzymes for biocatalytic applications 41 as well as target RNAs. 30, 42 The Qβ hairpin loop is followed by the 3 target regions: two SARS-CoV-2 N regions (Accession NC_045512.2 N1: gene location: 28271-28443; N2: gene location: 29091-29230) and the human RNase P region (or RP; Accession NM_006413: gene location: 1-280). The module is flanked by a T7 promoter and T7 terminator at 5' and 3' end, respectively, for RNA transcription to enable both in vitro and in vivo reconstitution of chimeric VLPs (Figure 1 and S1 ). SDM flanked with T7 promoter and T7 terminator (SDM + T7P/T) was cloned into pCDFDuet™-Qβ and pET-28a (+) to generate plasmid Qβ 1P-C19 and Qβ 2P-C19, respectively ( Figure S2 ). Plasmid Qβ 1P-C19 allows co-expression of the Qβ coat protein (CP) gene and SDM RNA from the same vector whereas plasmid Qβ 2P-C19 expresses only SDM RNA. pCDFDuet™-Qβ, which expresses the Qβ CP gene, was co-transformed into the same bacterial cell with plasmid Qβ 2P-C19 for in vivo encapsidation of SDM RNAs. The two-plasmid system has been used successfully in previous work for in vivo reconstitution of Qβ around target RNAs. 30, 42 Design of SDM from 5' to 3': T7 promoter (grey), Qβ hairpin (yellow), N1 (red), N2 (green), and RP (black), T7 terminator (grey). (B) Production of Qβ 1P-C19 and Qβ 2P-C19 VLPs via in vivo assembly. Qβ 1P-C19 VLPs were produced from a one-plasmid system, where the gene of the Qβ CPs and SDM RNAs were cloned in one vector (pCDFDuet™-1). Qβ 2P-C19 VLPs were produced from a two-plasmid system, where Qβ CP was derived from pCDFDuet™-Qβ and SDM RNA was produced from pET-28a (+). (C) CCMV-C19 VLPs were produced by in vitro reconstitution of the transcribed SDM RNAs with purified CCMV CPs.

Qβ coat protein SDM + T7P/T One-plasmid system Two-plasmid system All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020. 10.16.20213991 doi: medRxiv preprint Several safety measures were built into the synthesis of these positive controls. SDM + T7P/T was cloned out-of-frame with the open reading frame to avoid protein translation of truncated SARS-CoV-2 N1 and N2 gene segments. The ribosome binding site (RBS) upstream of SDM was removed for the same reasons. Furthermore, the presence of RBS upstream of target RNA has been shown to reduce packaging efficiency of RNAs into VLPs due to competitive binding of ribosomes versus Qβ CPs to the target RNA. 30 Therefore, for the one-and two-plasmid systems, the RBS was removed upstream of the SDM, but RBS was retained upstream of the Qβ CPs to enable protein translation. In vitro transcription of SDM RNAs yielded approximately 150 µg of SDM RNAs per reaction; denaturing urea polyacrylamide gel electrophoresis of the transcribed SDM RNA confirmed the ~622 nt RNA product ( Figure S3A ). Lastly, we used RT-qPCR assays to confirm that the SDM provided a template for the CDC primer/probe sets ( Figure S3C ).

In vivo encapsidation of SDM RNAs in Qβ VLPs was achieved by co-expression of Qβ CPs and SDM RNAs in E. coli using the aforementioned one-plasmid system and two-plasmid system to produce Qβ 1P-C19 VLPs and Qβ 2P-C19 VLPs, respectively. Production of Qβ 2P-C19 VLPs using the two-plasmid system was performed by transforming plasmid pCDFDuet™-Qβ and plasmid Qβ 2P-C19 into the same bacterial cells for co-expression of Qβ CPs and SDM RNAs. We also developed the one-plasmid system Qβ 1P-C19 from a single plasmid using the pCDFDuet™-1 vector.

For CCMV-C19, native CCMV was obtained from infected black-eyed pea No. 5 plants and disassembled to obtain purified CPs. The purified CPs were then reassembled with in vitro All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. Characterization of the VLP-based SARS-CoV-2 RT-PCR positive controls revealed production of intact, pure, and monodisperse chimeric VLPs (Figure 2) . Separation of intact VLPs on native agarose gels indicated successful RNA encapsidation, as the RNA and protein comigrate, yielding overlapping and discrete bands when stained with GelRed™ (RNA stain) and

Coomassie blue (protein stain). The VLPs have an overall negative charge and therefore migrate toward the anode. 45, 46 The size of the VLPs was determined by dynamic light scattering (DLS) and transmission electron microscopy (TEM). DLS of Qβ 1P-C19 and Qβ 2P-C19 revealed the presence of monodisperse nanoparticles with an average diameter of ~32 nm (PD ~0.12-0.13). For CCMV-C19, DLS also revealed monodisperse nanoparticles with a hydrodynamic radius of ~33 nm (PDI 0.128). In all cases, broken particles, free CPs, or aggregates were not detected. These findings are in agreement with TEM imaging showing intact and uniform VLPs with Qβ measuring All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint ~ 26-27 nm and CCMV measuring ~29 nm (Figure 2) ; the differences between DLS and TEMspecifically, the increased size as measured by DLS is explained by the hydration status of the sample; while 'dry' samples are imaged in TEM, DLS reports the hydrodynamic particle size of VLPs in solution. 47 Size-exclusion chromatography (SEC) for all positive controls showed typical elution profiles for VLPs, where nucleic acid (260 nm) and protein (280 nm) were eluted concurrently at around 12 mL from a Superose 6 column. 46, 48 Also, the SEC results confirmed the structural integrity of particles, showing no free RNAs or CPs in the VLP preparations. Lastly, the particle properties of the chimeric VLPs packaging the SDM matched data of Qβ VLPs devoid of the SDM ( Figure S4 ) as well as native CCMV particles ( Figure S5 ). All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint of negatively stained VLPs packaging the SDM using transmission electron microscope (TEM). Average size of 20 particles tabulated by ImageJ software was stated in inset box. SEC: Analysis of positive controls by size exclusion chromatography (SEC) using a Superose 6 column and GE Healthcare Äkta Purifier chromatography system; protein was detected at 280 nm and RNA was detected at 260 nm.

As a benchmark, the encapsidated RNAs within the positive controls were extracted using QIAGEN QIAamp Viral RNA mini kit; the procedure followed CDC recommendations. One difference to note between SARS-CoV-2 and our control probes is that SARS-CoV-2 is an enveloped virus, 49 but our control probes based of Qβ and CCMV are non-enveloped viruses.

Nevertheless, the extraction protocol is highly suitable for RNA extraction from viruses and the lysis buffer (Buffer AVL) used in this study has been optimized by the manufacturer to isolate RNA from a wide variety of viruses, enveloped and non-enveloped. Moreover, QIAGEN QIAamp® Viral RNA Mini Kit (RNA isolation kit used in this study) is recommended by CDC and has previously been reported to isolate RNA from both enveloped and non-enveloped viruses. [50] [51] For these reasons, the structural difference between SARS-CoV-2 and the VLP control probes should not affect the results of the diagnostic assay. RNAs were extracted and then eluted in RNase-free water instead of the CDC-recommended AVE buffer for quantification and purity check. Sodium azide in the AVE buffer interferes with absorbance readings between 220 and 280 nm and was thus avoided for the quality control studies. Total nucleic acids extracted include carrier RNAs (poly A, as per manufacturer's instructions), the target SDM RNAs, and, in the case of Qβ, also random host RNAs that are packed in vivo. Here, we assume that the amount of carrier RNAs is constant for each sample. When Qβ 1P-C19 and Qβ 2P-C19 were compared to Qβ VLPs, All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint Qβ 1P-C19 and Qβ 2P-C19 packaged ~30% more total nucleic acids compared to Qβ VLPs devoid of the SDM ( Table 1) . The higher nucleic acid content for the designer VLPs Qβ 1P-C19 and Qβ 2P-C19 may be attributed to the in vivo transcribed SDM RNAs with the Qβ hairpin loop. While Qβ can package RNA non-selectively based on electrostatic interactions of negatively charged cellular RNAs to the positively charged EF-loop on Qβ coat proteins, 52 the 29-nt Qβ hairpin loop added to the SDM confers high affinity for Qβ CP 40,53 and therefore enhances packaging and encapsidation efficiency of the SDM vs. random host RNA. While there was no significant difference in the mean of total nucleic acids extracted from Qβ 1P-C19 and Qβ 2P-C19, Qβ 2P-C19 showed higher batch-to-batch variability with the amount of total nucleic acid extracted from Qβ 2P-C19 ranging from 230 ng/µL to 360 ng/µL. This indicates that the two-plasmid system may be less reproducible and that the one-plasmid system would be the preferred system based on the higher quality control and assurance provided. Extracted RNA was analyzed by denaturing urea polyacrylamide gel electrophoresis and staining of the nucleic acids and imaging under UV light revealed presence of the SDM (622 nt). In addition, background RNAs were observed, which can be attributed to the packaging of host RNA as described above ( Figure S6A , Lane 1-3). CCMV-C19 showed the lowest amount of total nucleic acids encapsidated, or about 50% of the total RNA extracted from Qβ positive controls. However, in stark contrast to Qβ, where the SDM only makes a fraction of the total RNA packaged ( Table 1) , for CCMV-C19, 100% of the RNA encapsidated into CCMV-C19 is the target SDM RNA ( Figure S6A ) -this is an advantage of the in vitro assembly system. Based on BCA assay and RiboGreen nucleic acid assay (Table S3) , about 3 SDM RNA molecules were packaged per CCMV-C19; and this is consistent with previous reports, e.g. reporting that 4 copies of 500-nt long RNAs could be encapsidated per CCMV. 54 In contrast, one SDM RNA molecule was encapsidated in every 17 th Qβ 1P-C19 particle and 30 th Qβ 2P-C19 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint particle; the inefficiency in packaging can be explained by competition of the SDM target cargo with host E. coli RNAs. The encapsulation of SDM into VLPs was further validated by investigating the release kinetics of SDM from the Qβ and CCMV-based probes. VLPs (Qβ 1P-C19, Qβ 2P-C19, CCMV-C19) were heated at 75°C and sampled; first samples were analyzed by agarose gel electrophoresis;

RNA bands were then extracted and subjected to RT-qPCR analysis; a time course study was performed and released RNA was samples over a 5-min time course. Stable encapsulation is confirmed at t=0 mins with the protein and RNA stain co-localizing after electrophoretic separation ( Figure 3 ). Upon heating of the samples, released RNA was also apparent and the intensity increased with increasing incubation time. The heating also reduced the mobility of the coat proteins, which further suggest disassembly and RNA release. 55, 56 To confirm that SDM were indeed encapsidated into the VLPs, RNAs from time point 0 (intact particles) was extracted from gels and validated by RT-qPCR. Presence of N1, N2, and RP fragments was confirmed (Figure 3 ; RT-qPCR panel). We also carried out RT-qPCR analysis on All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint RNA released from the Qβ 1P-C19 and Qβ 2P-C19 VLPs upon heating at 75°C and presence of SDM RNAs and its N1, N2, and RP was confirmed. Similar observations were made for the CCMV-C19 construct; however, to achieve effective release longer incubation was needed to release the RNA. Also, for the CCMV-C19 probe, RT-qPCR confirmed presence of N1, N2, and RP, the components of the SDM RNA. The amount of SDM RNAs quantified by RT-qPCR matched the RNA band intensity measured. It was interesting to note that RNA release kinetics were slower for faster for Qβ vs. CCMV, because Qβ possess higher thermal stability. 57 The porous structure of Qβ likely is the underlying reason for the faster RNA release kinetics observed. All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint RT-qPCR was performed using all three sets of primer/probe sets (N1, N2, RP). NTC represents no-template control in RT-qPCR.

Amplification efficiency is one of the most important factors in qPCR. An ideal amplification efficiency of 100% corresponds to exponential doubling of the PCR product during every cycle to give an amplification factor of 2. 58 The slope of standard curve is be used to tabulate the amplification efficiency, with the equation: E = (10 -1/slope -1) x 100%. 59 Amplification efficiency of the CDC primer/probe set has been validated with synthetic SARS-CoV-2 RNA transcripts as well as clinical samples. 12,60 N1 and N2 primer sets have reported amplification efficiencies of more than 90%, which validates that the primers are optimized for SARS-CoV-2 RNA binding. Here, we confirmed the amplification efficiency of the CDC primers on our SDM RNA. All three regions (N1, N2, RP) showed amplification efficiency above 90%, with correlation coefficient (R 2 ) >0.99 (Figure 4) . N1 and RP primer/probe sets demonstrated higher sensitivity as compared to N2 primer/probe set and enabled detection of our SARS-CoV-2 positive controls at 10 1 copies/µL for Cq <40. For the N2 primer/probe set, the detection limit was 10 2 copies/µL. This result aligned with a previous report when using real patient samples, 12 indicating that our SARS-CoV-2 positive controls are able to function as a SARS-CoV-2 mimic for RT-qPCR detection.

All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Using the RT-qPCR method, we determined the SDM RNA copy number per nanogram of total nucleic acids extracted (after normalization of the Cq value, see Figure S6B ). The SDM RNA copy number was tabulated from mean copy number of N1, N2, and RP based on the standard curves shown in Figure 4 . Proper baseline settings are critical for accuracy. 61 Therefore, to standardize the baseline setting in RT-qPCR, we first normalized the quantification cycle (Cq) value of CCMV-C19 to in vitro transcribed SDM RNA by assuming that total RNAs extracted from CCMV-C19 contain only SDM RNA. The normalized ratio is then applied to Cq values of Qβ 1P-C19 and Qβ 2P-C19, as mentioned in Figure S6B , to obtain the percentage of SDM RNAs in total RNA extracted. CCMV-C19 has the highest SDM RNA copy number, with >10 9 copies per nanogram of total nucleic acids ( Table 1) . This is as expected, because CCMV-C19 was assembled in vitro using only SDM target RNAs. Qβ 1P-C19 and Qβ 2P-C19 have similar copy number of SDM RNA, or around 10 8 for every nanogram of total nucleic acids ( Table 1) . The lower copy number is owing to encapsidation of cellular E. coli RNAs, which has been shown in (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint Figure S6A . All CDC primer/probe sets demonstrated no binding to random cellular E. coli RNAs that encapsidated in Qβ.

There was no significant difference in SDM RNA copy number between Qβ 1P-C19 and Qβ 2P-C19 (unpaired t-test, p > 0.05), suggesting that the one-plasmid system with indirect upstream RBS has similar packaging efficiency as the two-plasmid system (though the two- (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint ( Figure 5A ). All three detection regions recommended by CDC (N1, N2, RP) in all three SARS-CoV-2 positive controls, Qβ 1P-C19, Qβ 2P-C19, and CCMV-C19, were successfully detected.

Importantly, it was possible to process the positive controls alongside the clinical samples throughout every step, starting from RNA extraction though amplification, providing seamless full-process control.

Consistent with the above-described findings, CCMV-C19 probes were the most sensitive, as reflected by the high copy numbers detected (> 2,000 copies/µL, Figure 5B) ; this is consistent with CCMV packaging the highest copy number of SDM. Fewer copies of SDM were packaged in Qβ 1P-C19 and Qβ 2P-C19 (due to encapsidation of random host RNAs) and this resulted in lower SDM copies detected (<1,000 copies/µL and <500 copies/µL, respectively, Figure 5B ).

Overall, the data were aligned with the results obtained from RT-qPCR (Table 2) .

When comparing the performance of Qβ 1P-C19 vs. Qβ 2P-C19, it was interesting to note the differences comparing ddPCR vs. RT-qPCR: while RT-qPCR showed no apparent differences in copy number of SDM (Table 1) , ddPCR indicated higher SDM copy for Qβ 1P-C19 vs. Qβ 2P-C19, with 1.53 x10 9 SDM/µg VLP vs. 4.48 x10 8 SDM/µg VLP ( Figure 5C ). The introduction of PCR inhibitors at the RNA extraction stage may account for the reduced sensitivity of the RT-qPCR method. 62 Previous studies have shown that ddPCR method exhibits higher tolerance to PCR inhibitors compared to conventional RT-qPCR and this is explained by reaction partitioning. 63, 64 Also, droplet-based reaction partitioning in ddPCR will lead to increased variance in SDM copy number when comparing the positive controls. Regardless of the differences in performance observed, each of the proposed SARS-CoV-2 positive controls -Qβ 1P-C19, Qβ 2P-C19, and CCMV-C19 -provided robust positive controls with encapsidate at least 10 8 SDM copies from every microgram of VLP ( Figure 5C ). All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. 

Most of the commercially available positive controls require cold-chain distribution in order to maintain product quality. However, the pipeline of cold-chain distribution starting from packaging to logistics increases the cost of the product and limits their widespread distribution.

Cold-chain infrastructure is not available in low-resource settings and is especially problematic in All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint countries with tropical climates. Bacteriophages and plant viruses evolved as highly stable capsids that retain their structural properties under various environmental conditions, making them suitable candidates for development of positive controls that do not require a cold-chain infrastructure. To investigate the stability of our in-house SARS-CoV-2 positive controls under conditions that simulate shipping conditions, samples were left at room temperature (approximately 20-25°C) for a month. Structural integrity and SDM functionality were then assessed using agarose gel electrophoresis, DLS, TEM, and SEC, as well as RT-qPCR. Qβ 1P-C19 and Qβ 2P-C19 exposed to these conditions remained intact. There was no aggregation and there were no free CPs to indicate disassembly of the Qβ positive controls, and the particle characteristics matched those obtained of freshly prepared particles (Figure 6 vs. Figure 2) . The excellent stability of Qβ can be attributed to the two cysteine residues on the Qβ coat protein forming inter-subunit disulfide bonds. 53 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint The CCMV-C19 positive control was found to be less stable, however still yielding a recovery of ~70% intact CCMV-C19 positive control after 1-month storage at ambient conditions ( Figure 6 , DLS panel). The CCMV assembly is stabilized by electrostatic and hydrophobic interactions. 65, 66 However, the lack of disulfide bonds in CCMV reduces longitudinal particle stability. 67 Data are consistent and indicate partial disassembly leading to some degree of aggregation: native gel electrophoresis revealed intact particles with RNA and CPs having matched electrophoretic mobility; however a second protein band appeared near the well, indicating aggregation of CPs and RNA (Figure 6 , Lane 2). While the RNA stain was less obvious, this pattern may indicate a mixture of disassembled CPs associated and/or aggregated with RNA via their positively-charged N-terminal tails. 68 VLP aggregation was not apparent from TEM imaging, therefore we attribute the agarose gel electrophoretic pattern as well as the observed aggregates in the DLS measurement to partial disassembly of CPs and their aggregation as CP-RNA complexes. Indeed, SEC of CCMV-C19 indicates a second broadened elution peak at [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] mLs. For application in the clinic, it may be desired to stabilize the CCMV capsid through All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint introduction of covalent bonds between CPs; this concept was recently demonstrated using a homobifunctional cross-linker 3, 3'-dithiobis (sulfosuccinimidylpropionate) (DTSSP) to form intra-particle cross-links between lysine side chains of adjacent CPs. 43 Next, and most importantly, we assayed for SDM stability. While no structural changes where observed for the Qβ 1P-C19 and Qβ 2P-C19 positive controls, RT-qPCR data indicate that the SDM copy number was reduced by 30-35% ( Table 2) . This observation is aligned with the reduction of total nucleic acid extracted from one-month-old positive controls (Table S1). We deduced the reduction of RNAs is due to hydrolysis contributed by water molecules in buffered solution providing hydroxyl or hydronium ions for proton transfer. 69 The Qβ capsid is a porous structure and has pores measuring 1.3-1.4 nm at its 3-fold axis as well as 0.7 nm-sized pores at its 5-fold axis. These pores facilitate diffusion of water (causing RNA hydrolysis), ions, as well as degraded RNAs (leading to loss of RNA content). 32,55 A possible strategy to stabilize the Qβ positive control for future applications may be to plug the pores through appropriate chemistry or through freeze-drying in appropriate excipients. 70 Lastly, an expected reduction in SDM copy number was also apparent for CCMV-C19. A 30% reduction of SDM was estimated from the dissociation observed (Figure 6) , and this is consistent with reduction of total nucleic acids (Table S1) after the 1-month storage period. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. probes also harbor a significant amount of 'junk' RNAs that may impact the quality of the RT-qPCR. Moreover, the increased variation in Cq values for the Qβ 1P-C19, and Qβ 2P-C19, with ~25 % fluctuation, may also be explained by the porous nature of the capsid which may be more prone to RNA hydrolysis. 69 Nevertheless, overall stability of the VLP constructs is maintained and demonstrated over the 1-week time course at various temperatures; even incubation at 40°C for one week did not significantly impact the SDM packaged in the VLP probes; therefore our proposed biomimetic control probes may be suitable for use also in low-resource settings.

All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. 

In conclusion, both plant virus-derived CCMV and bacteriophage-derived Qβ VLPs offer promising platforms for the encapsidation of RNA modules and application as stable, widely applicable positive controls for RT-qPCR or ddPCR detection of infectious agents, such as SARS-CoV-2. We developed in vitro reconstitution protocols and in vivo expression systems, yielding CCMV-and Qβ-based nanoparticles encapsidating SARS-CoV-2 detection modules that are compatible with the CDC primer/probe sets. Because the RNA is stabilized inside the VLP particle, the positive control mimics the conditions encountered by the RNA template of SARS-CoV-2 within clinical or environmental samples. The increased stability also enables for these positive controls to be applied as full-process controls, as demonstrated in the clinical assays performed. Both, the CCMV and Qβ platforms are massively scalable through manufacture via plant molecular farming and bacterial fermentation. The in vivo expression of Qβ VLPs offers the advantage of fewer processing steps compared to in vitro dis-assembly and reassembly used to All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint obtain the chimeric CCMV VLPs. The latter method does offer control of target molecule encapsidation: while the payload of CCMV is solely SDM (~ 10 9 SDM copies/µg VLP), only a fraction of the Qβ payload is SDM (~ 10 8 SDM copies/µg VLP), with a large portion of the cargo being non-target host RNAs. Nevertheless, both VLP systems were robust in clinical assays.

Importantly, the developed positive controls are safe and avoid the risks of using RNA extracted from infected patients. The positive controls demonstrated considerable stability over 1-month at ambient conditions; further, the probes offered excellent stability at temperatures as high as 40°C over one-week (longer time periods were not tested). Together, these attributes made the handling of SARS-CoV-2 positive controls safe and accessible for clinical personnel in a wide range of settings; given the ease of manufacture and stability conferred over a range of environmental setting, the proposed designs may aid diagnostic testing not just in testing facilities, but also at the point of entry (i.e. at airports or border crossings) and in low-resource areas. The potential to make these SARS-CoV-2 positive controls widely available at ambient conditions could help to alleviate some of the disparities in testing that are contributing to the increased COVID-19-related deaths in underserved populations in the US and across the world.

Qβ 1P-C19: SARS-CoV-2 detection module (SDM) was synthesized and cloned into pCDFDuet™-Qβ between restriction site NotI and NdeI to generate Qβ 1P-C19 plasmid (GenScript, Figure S1 ). The gene was cloned not in frame with the open reading frame (for safety All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. 

Plasmids were transformed into BL21 (DE3) competent E. coli cells (New England Biolabs ® ) and plated out on antibiotic containing plate. Selection of Qβ 1P-C19 transformants was based on streptomycin resistance (100 µg/mL); while Qβ 2P-C19 transformants was based on streptomycin resistance (100 µg/mL) and kanamycin (50 µg/mL). The E. coli was inoculated in Luria-Bertani (LB) media supplemented with antibiotic and incubated overnight at 37°C with shaking at 250 rpm. For VLP expression, the overnight culture was diluted 1: 100 in Thermo Fisher Scientific's MagicMedia™ E. coli Expression Medium and incubated overnight at 30°C with shaking at 300 rpm. The culture was centrifuged at 9800 x g for 10 min at 4°C. The cell pellet was resuspended with 0.1 volume of 1 X phosphate buffered saline (PBS), pH 7.4 supplemented with 0.2 mg/mL lysozyme and sonicated for 10 min with 5 sec on/off cycle. The lysate was then centrifuged at All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint 9800 x g for 15 min to collect the supernatant. Crude VLPs in supernatant were pelleted down with 10% (w/v) PEG 8000 and 0.2 M NaCl followed by resuspension in 1 X PBS (pH 7.4). Crude VLPs were treated with 0.7 volumes of 1:1 chloroform: butanol mixture and the upper aqueous layer was purified on 10-40% (w/v) sucrose gradient by centrifugation at 133,000 x g for 3 h, 4ºC.

The band with VLPs was collected and centrifuged at 210,000g for 3 h, 4ºC. The final pellet was resuspended in 1 X PBS (pH 7.4) and the VLPs were stored at -20°C. VLP concentration was measured with Pierce™ BCA protein assay kit.

Plasmid Qβ 1P-C19 was digested with NotI and NdeI to obtain the linearized SDM. 

Primary leaves of Vigna unguiculate, California black-eyed peas No. 5 were mechanically infected with CCMV after growing for 12 days (these protocols are carried out under USDA-approved P526 permits). Plants were grown for another 12 days before leaves were harvested. CCMV was purified using established procedures. 72 In brief, harvested CCMV infected leaves were homogenized with Preparation Buffer (0.2 M NaOAc, 1 mM EDTA, PH 4.8) and filtered through All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint cheesecloth. Filtrate was later centrifuged at 15,000 x g for 15 min. The supernatant was precipitated by 0.02 M NaCl, and 8% (w/v) PEG 8000 and stirred overnight at 4°C. The solution was centrifuged at 15,000 x g for 10 min at 4°C. The pellet was resuspended in 20 mL CCMV Buffer (0.1 M NaOAc, 1 mM EDTA, PH 4.8) followed by centrifugation at 8000 x g for 10 min at 4°C. The supernatant was then centrifuged over a 5 mL 20% (w/v) sucrose cushion in water at 148,000 x g for 2 h at 4°C. The pellet was resuspended in CCMV Buffer. Purified CCMV was stored as intact virion until further use. Purified coat proteins (CPs) were obtained using established disassembly protocols. 54, 73 Disassembled CCMV CPs were reassembled with purified SDM obtained through in vitro transcription from plasmids; the SDM-to-CP ratio was 6:1 as previously described. 54 The reconstituted CCMV was stored in CCMV Buffer at -80°C. VLP concentration was measured with Pierce™ BCA protein assay kit.

Transmission electron microscopy (TEM): Positive controls (Qβ 1P-C19, Qβ Β 2P-C19, CCMV-C19) were diluted to 0.2 mg/mL in Milli-Q water and 4 µL was adsorbed to Formvar/carbon-coated 400 mesh copper grids (Electron Microscopy Science) for 2 min. The grid was washed with 4 µL of water for 2 min followed by adsorption of 4 µL of 2% (w/v) uranyl acetate (Fisher Scientific) for 2 min. Solution was removed from grid by blotting with filter paper.

TEM grids were imaged with FEI Tecnai G2 Spirit transmission microscope at 80 kV. The size of particles was analyzed using ImageJ software. 20 particles were randomly selected and their diameters measured.

Qβ 1P-C19 and Qβ 2P-C19 were diluted to 0.2 mg/mL in 1 X phosphate buffered saline (PBS, pH 7.4) Qβ 1P-C19 and CCMV-C19 samples were diluted to 0.2 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 

SARS-CoV-2 positive control (10 µg) in a total volume of 20 µL was incubated at 75°C for 0 min, 0.5 min, 1 min, 2 min, 3 min, 4 min, and 5 min. The solution was later analyzed with native agarose gel electrophoresis as mentioned previously and band intensity was analyzed by ImageJ software.

The RNA bands were excised by scalpel and soaked in 500 µL of 1 x TE buffer, Molecular Biology Grade (Promega) supplemented with 40 U RNase inhibitor (Applied Biosystem™). The mixture was incubated at room temperature for 25 min with gentle shaking. The gel was later removed from TE buffer and RNA was extracted using Thermo Scientific™ GeneJET Gel Extraction Kit according to manufacturer's protocol with slight modification. Briefly, 1 volume of Binding buffer was added to 1 volume of gel and incubated at 56°C until dissolved (6-8 min). The solution was transferred to column and spun 1 min at 14,550 x g. The column was washed twice with 700 µL of Wash buffer by spinning at 14,550 x g for 1 min. An additional spin at similar condition was performed to remove residual ethanol. Then, 30 µL of nuclease-free water was added to the center of membrane and incubated at 56°C for 5 min. Lastly the column was spun at 14,550 x g for 1 min to elute the RNA. 1 µL of eluted RNA was quantified with RT-qPCR using N1, N2, and RP primers/probe, respectively.

SARS-CoV-2 positive controls (10 µg) were extracted using QIAGEN's QIAamp Viral RNA mini kit (cat. no: 52904) according to manufacturer's protocol and eluted in 140 µL of AVE buffer. 10 µL of eluted RNA sample was diluted to 10 -6 with RNase-free water. Clinical samples extracted All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint from a COVID-19 case (positive control patient) and a healthy patient (negative control). Clinical samples were diluted 100 times with RNase-free water. 10 µL of the diluted RNA sample was used to set up singleplex ddPCR in a 20 µL reaction with Bio-Rad One-step RT-ddPCR Advanced Kit (cat. no: 1864022) according to the manufacturer's protocol. Briefly, the reaction mixture consisted of 5 µL of 4 X One-step RT-ddPCR supermix for probes, 10 µL of RNA, 2 µL of reverse transcriptase, 1 µL of 300 nM DTT, 900 of nM each forward and reverse primer and 250 of nM probe. N1, N2 and RP primer/probes were synthesized from Integrated DNA Technologies (Table   S2 ). The mixture was then used for droplet generation by adding 70 μL of Bio-Rad droplet generation oil (cat. no: 1864007). Droplets were generated with Bio-Rad QX200™ Droplet

Generator. The droplets were incubated at 25°C for 3 min, 45°C for 60 min, 95°C for 10 min and then cycled at 95°C for 30 sec and at 55°C (N1, N2,) or 60°C (RP) for 60 sec. Amplification was performed for 45 cycles using Applied Biosystems ® Veriti ® 96 well Thermal Cycler. The droplets from each sample were read on QX200™ droplet reader machine. The data was processed using QuantSoft™ version 1.7.4 software.

SARS-CoV-2 positive controls (10 µg) were aliquoted to a total volume of 20 µL and stored at four different temperature (-20°C, 4°C, room temperature: 20°C-25°C, 40°C) for 1 h, 1 day and 1

week, respectively. RNA was extracted from VLPs and 1 µL of eluted RNA was used in RT-qPCR as mentioned previously using N1 primers/probe. All samples were assayed in triplicate on BioRad CFX96 Touch™ Real-Time PCR Detection System. Quantification cycle (Cq) values were tabulated by CFX™ Maestro Software.

All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. https://coronavirus.jhu.edu/testing/testing-positivity (accessed Jul 13, 2020).

All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. (24) Asuragen Develops Armored RNA Quant® SARS-CoV-2 Control All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint

In vivo encapsidation In vitro encapsidation Virus-like particles All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 20, 2020. . https://doi.org/10.1101/2020.10.16.20213991 doi: medRxiv preprint

The Mathematical Strategy That Could Transform Coronavirus Testing

Comparison of National RT-PCR Primers, Probes, and Protocols for SARS-CoV-2 Diagnostics

Analytical Sensitivity and Efficiency Comparisons of SARS-CoV-2 RT-QPCR Primer-Probe Sets

Laboratory Readiness and Response for Novel Coronavirus (2019-NCoV) in Expert Laboratories in 30 EU/EEA Countries

Minimal System for Assembly of SARS-CoV-2 Virus like Particles

Production of SARS Coronavirus-Like Particles That Bind Host Cells and Serve as Vaccine Antigen Plasmodium Falciparum CS Protein Mediated by

COVID-19 Vaccine Development and a Potential Nanomaterial Path Forward

RNA: Packaged and Protected by VLPs

The Crystal All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Structure of Bacteriophage Qβ at 3.5 Å Resolution

Engineering Thermal Stability in RNA Phage Capsids via Disulphide Bonds

Highly Efficient Strategy for the Heterologous Expression and Purification of Soluble Cowpea Chlorotic Mottle Virus Capsid Protein and in Vitro PH-Dependent Assembly of Virus-like Particles

Plant Molecular Farming of Virus-like Nanoparticles as Vaccines and Reagents. WIREs Nanomedicine and Nanobiotechnology

Two-Stepping through Time: Mammals and Viruses

The Double-Edged Sword: How Evolution Can Make or Break a Live-Attenuated Virus Vaccine

Design of Virus-Based Nanomaterials for Medicine

Plant Viruses in Plant Molecular Pharming: Toward the Use of Enveloped Viruses. Front

Specific RNA Binding by Q

No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity

RNA-directed Packaging of Enzymes within Virus-like Particles

Functional RNAs: Combined Assembly and Packaging in VLPs

Versatile Reversible Cross-Linking Strategy to Stabilize CCMV Virus like Particles for Efficient SiRNA Delivery

Deletion of Highly Conserved Arginine-Rich RNA Binding Motif in Cowpea Chlorotic Mottle Virus Capsid Protein Results in Virion Structural Alterations and RNA Packaging Constraints

Role of Electrostatics in the Assembly Pathway of a Single-Stranded RNA Virus

Preclinical Development of a Vaccine against Oligomeric Alpha-Synuclein Based on Virus-like Particles

No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity

The Antitumor Efficacy of CpG Oligonucleotides Is Improved by Encapsulation in Plant Virus-Like Particles

The Role of Lipid Metabolism in COVID-19 Virus Infection and as a Drug Target

Optimization and Validation of Sample Preparation for Metagenomic Sequencing of Viruses in Clinical Samples

Metagenomic Analysis of Viral Nucleic Acid Extraction Methods in Respiratory Clinical Samples 11 Medical and Health Sciences 1108 Medical Microbiology

Structures of Qβ Virions, Virus-like Particles, and the Qβ-MurA Complex Reveal Internal Coat Proteins and the Mechanism of Host Lysis

Engineering Thermal Stability in RNA Phage Capsids via Disulphide Bonds

Self-Assembly of Viral Capsid Protein and RNA Molecules of Different Sizes: Requirement for a Specific High Protein/RNA Mass Ratio

Encapsidated Atom-Transfer Radical Polymerization in Qβ Virus-like Nanoparticles

Polymorphic Assembly of Virus-Capsid Proteins around DNA and the Cellular Uptake of the Resulting Particles

Engineered Mutations Change the Structure and Stability of a Virus-Like Particle

A Novel Method for Real Time Quantitative RT-PCR

Rapid Cycle Real-Time PCR: Methods and Applications

Comparative Analysis of Primer-Probe Sets for the Laboratory Confirmation of SARS-CoV-2

No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity

Amplification Efficiency: Linking Baseline and Bias in the Analysis of Quantitative PCR Data

PCR Inhibitors-Occurrence, Properties and Removal

Tolerance of Droplet-Digital PCR vs Real-Time Quantitative PCR to Inhibitory Substances

Accurate Digital Polymerase Chain Reaction Quantification of Challenging Samples Applying Inhibitor-Tolerant DNA Polymerases

Structures of the Native and Swollen Forms of Cowpea Chlorotic Mottle Virus Determined by X-Ray Crystallography and Cryo-Electron Microscopy

The Mechanism and Pathway of PH Induced Swelling in Cowpea Chlorotic Mottle Virus

Virus Engineering: Functionalization and Stabilization

200-All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted

Mg2+-Independent Hairpin Ribozyme Catalysis in Hydrated RNA Films

Rational Design of a Stable, Freeze-Dried Virus-like Particle-Based Vaccine Formulation

Construction of SARS-CoV-2 Virus-Like Particles by Mammalian Expression System

The Self-Assembly of Spherical Plant Viruses

Dispensability of 3′ TRNA-like Sequence for Packaging Cowpea Chlorotic Mottle Virus Genomic RNAs

No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted