Optimal Preparation of COVID-19 Viral Transport Medium for Culture


 IntroductionThe sudden arrival of the COVID-19 pandemic placed significant stresses on supply chains including viral transport medium (VTM). The VTM that was urgently required needed to support viral replication, as well as other routine diagnostic approaches. We describe the preparation and validation testing of VTM for rapidly expanding diagnostic testing, where the capacity of the VTM to preserve viral integrity, for culture, isolation and full sequence analysis, was maintained.MethodsVTM was prepared using different methods of sterilization then ‘spiked’ with virus. The VTM was investigated using viral culture in Vero cells, and for nucleic acid detection by quantitative PCR (qPCR).ResultsThe best results were obtained by filter and autoclave-based sterilization. The VTM proved robust for culture-based analyses provided the inoculated VTM was stored at 4oC, and tested within 48 hours. The filtered VTM also supported PCR-based diagnosis for at least 5 days when the mock inoculated VTM was held at room temperature (RT).DiscussionThe manual handling of VTM production, including filling and sterilization, was optimized. SARS-CoV-2 was spiked into VTM to assess different sterilization methods and measure the effects of storage time and temperature upon VTM performance. While most diagnostic protocols will not require replication competent virus, the use of high quality VTM will allow for the next phase of laboratory analysis in the COVID-19 pandemic, including drug and antibody susceptibility analysis of re-isolated SARS--CoV-2, and for the testing of vaccine escape mutants.


Introduction
The COVID-19 global pandemic has tested the supply chains that support the diagnosis of infections, including testing equipment and assay components and the availability of swabs and viral transport medium (VTM), in some areas, became limiting for COVID-19 testing.
The CDC has a draft Standard Operating Procedure (SOP) for the preparation of VTM (SOP#:DSR-052-04) and VTM must sustain viral integrity and suppress contaminating microorganisms that might interfere with diagnosis (1). In Australia, the Therapeutic Goods Administration requests VTM sterility and a composition consistent with the CDC SOP.
The Media Production Unit (MPU) at the Peter Doherty Institute for Infection and Immunity, a small accredited (ISO/IEC17025:2017), near at-cost, media provider for local and regional diagnostic laboratories using largely manual processes, was asked to supply VTM at 20,000 units per week.
Production of VTM at the MPU expanded in parallel with the local production of 3D-printed nasopharyngeal swabs to address increased testing (2). This study analyses sterilization and VTM ability to support to the growth of SARS-CoV-2 in permissive cells.
Antimicrobial properties of VTM prepared by different means Bacterial or yeast stationary phase cultures were diluted and inoculated into VTM at 10 3 -10 4 CFU. The VTM was incubated for 3 days at RT (18-20°C), whereafter 10 µL was removed and plated onto HBA.

Virus 'spiking' experiments
The 'spiking' of VTM with SARS-CoV-2 was conducted in the Doherty Institute PC3 Laboratory. Each 3 mL VTM sample (n = 3 per preparation condition) was incubated with 300 µL of SARS-CoV-2 stock (at 10 5.25 TCID 50 /mL) to yield 60,000 infectious viral particles per 3.3 mL. Samples were incubated at RT (18-19 o C, 40% relative humidity) or at 4 o C, for various periods before 200 µL was harvested and a TCID 50 performed.
50% Tissue Culture Infectious Dose assay (TCID 50 ) Tissue culture plates (24 well) were seeded with Vero cells 24 hours prior to use and were used at ~ 95% con uency. The samples were serially diluted and inoculated (4 replicates/sample and up to 4 serial dilutions) and incubated for 45 min to enable virus infection of monolayers, then infection media (without FBS) but with TPCK trypsin (1 µg/mL) was added. The plates were incubated (37 o C, 5% CO 2 ) and examined 3 d post-inoculation for virus-induced cytopathic effect (CPE). The TCID 50 /mL was calculated (5) and plotted as the mean ± standard error of repeated experiments. A two-way ANOVA with Tukey posttests and p < 0.05 was considered to indicate statistical signi cance.

Quantitative PCR assay of virus in VTM
Inactivated SARS-CoV-2 virus was inoculated into 3 mL of ltered sterilized VTM or Kang-Jian Viral transport media (catalogue no KJ502-19) and tested at day 0 then 5 days post inoculation after storage at RT or 4°C. A 300 µl sample was extracted on the Chemagic360 platform using a Chemagic Viral Nucleic Acid Kit (Perkin Elmer) and eluted into 60 µL. PCR targeting the SARS-CoV-2 E gene comprised 5 µL nucleic acid extract, 1 x qScript XLT One-Step RT-qPCR ToughMix Low ROX (Gene Target Solutions), 400 nM of primers E_Sarbeco_F and E_Sarbeco_R and 200 nM probe E_Sarbeco_P1 (6) using conditions described by Corman et al. 2020 (6).

Results
The antimicrobial properties of the VTMs produced by the three sterilization methods is shown in Table 1. The antifungal activity of the VTM was decreased by autoclaving or gamma irradiation, whereas the antibacterial activity was not affected by the method of preparation.  (Fig. 1). Samples stored at RT or 4 o C showed relatively similar infectious viral loads over 24 hours, where after the viral load in the VTM prepared by gamma irradiation declined signi cantly.
Samples stored at RT revealed limited impact on recovery of replicative virus with storage over the rst 24 hours. After 24 hours, there was a signi cant decline in recovered virus from VTM that was prepared by gamma-irradiation, and by 48 hours at RT, no replicative virus was recovered from VTM that was prepared by gamma-irradiation.
To con rm that the ltered VTM did not interfere with PCR-based diagnosis of COVID-19, inactivated SARS-CoV-2 was added to VTM (MPU or Kan-Jian (control)) and incubated for 5 days at RT, or at or 4°C, prior to qPCR testing (  Quantitative PCR (qPCR) was used to analyse inactivated SARS-CoV-2 was 'spiked' into commercially source viral transport media (Kang-Jian) or VTM produced by the Media Production Unit. The transport media were held at 4°C or RT (18-20°C) for 5 days, because nucleic acid was extracted and the numbers of genomes determined by cycle threshold (Ct) number in the PCR. The values were within normal variation (i.e. one cycle, 27.5-28.5 cycles) for the amount of nucleic acid inoculated into the transport media. ND -not done.

Discussion
COVID-19 drove the parallel development of diagnostic regimes on different platforms, using assays that detect viral nucleic acid, viral proteins and antibodies induced by infection. As therapies and vaccines are introduced, it will be important to isolate infection-competent virus from diagnostic samples to test for e.g. virus 'escape mutants' to complement qPCR testing. This study explores the ability of VTM to support SARS-CoV-2 culture, produced using different sterilization options -lter sterilization, autoclave and gamma-irradiation.
The Australian Department of Health asked that VTM comply with CDC speci cations (7), despite recent evidence that, for short transits, saline performs as well at VTM (8,9) for standard qPCR-based diagnostic procedures. We expected that culture-based analysis would require greater preservation of the virus present in diagnostic samples. Although not previously a VTM provider, the MPU was asked to provide 20,000 vials of VTM per week and sought to optimize the production so as to minimize its impact on staff and ongoing media supply operations.
Gamma-irradiated VTM was less supportive of viral culture, resulting in a drop of viral titre of between 100 − 10,000-fold, 2-3 days after inoculation, which was more marked if the samples were stored at RT. After 72 hours at RT, no infectious virus was recovered from VTM prepared by gamma-irradiation. The reason why irradiation would destroy the preservative properties of VTM were not explored but irradiation can denature bovine serum proteins (10). VTM that was lter-sterilized, or autoclaved, preserved SARS-CoV-2 replication, however the antifungal present in the VTM was inactivated by autoclaving, suggesting that VTM prepared by autoclaving might be less useful where diagnostic samples are not refrigerated, or where transit time from sampling to diagnostic lab is longer than 1-2 days.
The next phase of the global response to the COVID-19 pandemic will be to more routinely isolate and sequence SARS-CoV-2 found in clinical samples, and to test whether they remain sensitive to the COVID-19 antivirals or plasma-derived (11) or laboratory-generated (12,13) antibodies. After vaccination is implemented, SARS-CoV-2 that escape vaccine-induced immunity (14) will need to be studied and this will require viral culture. The data presented here suggests viral recovery will be optimized by the use of lter sterilized VTM.

Conclusions
Page 7/10 The conservative preservation of SARS-CoV-2 in patient samples to enable viral culture will be required to fully understand why the growing list of COVID-19 vaccines or therapies fail. We showed that the method of CDC VTM preparation is critical to preserving the capacity of the virus to replicate. The optimal method to preserve viral culture used lter sterilization, while other methods supported nucleic acid-based diagnosis.

Consent to publication
All the authors consent to the publication of the manuscript.

Availability of Data
The data contained within the manuscript is available for electronic transmission via the corresponding author.

Competing interests
There are no competing nancial interests.
Funding Figure 1 Recovery of replicative virus from VTM. VTM was prepared using three methods -lter sterilization, sterilization by autoclaving, or by gamma-irradiation. The VTM was inoculated with replicative virus then stored for various periods of time at 4°C (A) or RT (18-20°C, (B)) before viable virus was assayed using Vero cells, serially diluted VTM, where viral cytopathic effect (CPE) was examined. The virus recovered was expressed as TCID50/mL. At each time point, the data for each method of preparation was compared using a Two ANOVA and Tukey post-test correction. A p<0.05 was considered signi cant (ns = not signi cant).