Effects of swab pool size and transport medium on the detection and isolation of avian influenza viruses in ostriches

Abstract

Background Rigorous testing is a prerequisite to prove freedom of notifiable influenza A virus infections in commercially farmed ostriches, as is the isolation and identification of circulating strains. Pooling 5 ostrich tracheal swabs in a 50 % v/v phosphate-buffered saline (PBS): glycerol transport medium (without antibiotics) is the current standard practice to increase reverse transcription real time PCR (RT-rtPCR) testing throughput and simultaneously reduce the test costs. In this study we investigated whether doubling ostrich tracheal swabs to 10 per pool would affect the sensitivity of detection of H5N8 high pathogenicity avian influenza virus (HPAIV) and H7N1 low pathogenicity avian influenza virus (LPAIV) by quantitative RT-rtPCR, and we also compared the effect of a protein-rich, brain heart infusion broth (BHI) virus transport media containing broad spectrum antimicrobials (VTM) on the efficacy of isolating the H5N8 and H7N1 viruses from ostrich tracheas, since the historical isolation success rate from these birds has been poor. Results Increasing the ostrich swabs from 5 to 10 per pool in 3 mls of transport medium had no detrimental effect on the sensitivity of the RT-rtPCR assay in detecting H5N8 HPAIV or H7N1 LPAIV; and doubling of the swab pool size even seemed to improve the sensitivity of virus detection at levels that were statistically significant (p less than or equal to 0.05) in medium and low doses of spiked H5N8 HPAIV and at high levels of spiked H7N1 LPAIV. On virus isolation, more samples were positive when swabs were stored in a protein-rich viral transport medium supplemented with antimicrobials in PBS: glycerol (10/18 vs. 7/18 for H5N8 HPAI); although the differences were not statistically significant, overall higher virus titres were detected (106.7 – 103.0 vs. 106.6 - 103.1 EID50 for H5N8 HPAIV and 105.5 - 101.4 vs. 105.1 – 101.3 EID50 for H7N1 LPAIV); and fewer passages were required with less filtration for both H5N8 HPAI and H7N1 LPAI strains. Conclusion Ostrich tracheal swab pool size could be increased from 5 to 10 in 3mls of VTM with no loss in sensitivity of the RT-rtPCR assay in detecting HPAI or LPAI viruses, and HPAI virus could be isolated from a greater proportion of swabs stored in VTM compared to PBS: glycerol without antibiotics.

Wild Bird Surveillance in the Gauteng Province of South Africa during the High-Risk Period for Highly Pathogenic Avian Influenza Virus Introduction

“…VetMAX AI detection kits demonstrated superior analytical and diagnostic performance compared to the Spackman test method and others, for example, the limit of viral RNA detection of VetMAX AI detection kits is one to two logs lower [28], and in our experience some of the provincial veterinary laboratories as well as the national reference laboratory that still use the Spackman oligonucleotides consistently fail to detect lower levels of IAV in inter-laboratory comparisons with wild bird samples (unpublished laboratory data). The application of next generation sequencing to swab samples where rRT-PCR Ct values are <30 (i.e., a high viral content) have greatly aided in the identification and genetic characterization of IAVs in wild bird faecal samples and we were able to obtain full genome sequences for an H5N1 HPAI virus and five H9N2 viruses, but a Ct <30 often also results in high rates of successful cultivation from faecal swabs stored in viral transport medium with antibiotics [26,31], and we successfully isolated representative H5N1 HPAI and H9N2 strains that are valuable as antigens for diagnostic testing, vaccine development and other research.…”

“…Sample pools with Ct values ≤30 [26,31] were either inoculated into 9-to-11-day-old embryonated Specific Pathogen Free (SPF) hen's eggs (AviFarms, Pretoria, South Africa) for virus isolation according to the WOAH-recommended method (2019), or in a confluent QH9-2/1 Quail cells (Nuvonis, Vienna, Austria), according to the supplier's recommended procedures.…”

Experimental infection of ostriches with H7N1 low pathogenic and H5N8 clade 2.3.4.4B highly pathogenic influenza A viruses

“…Virus isolation is still considered the "gold standard" for agent identification, where it is used primarily for diagnosis of the first clinical case and to obtain antigen for subsequent testing (OIE, 2019). Virus isolates from RT-qPCR positive ostrich field samples have been rare (Abolnik et al, 2016), but recent studies aimed at investigating the possible reasons for this determined that the standard PBS: glycerol transport medium (50 %, v/v, without antimicrobials) negatively affects AIV viability in ostrich tracheal swabs (Pieterse et al 2021) and that excluding antimicrobials from VTM encourages the growth of specific bacteria that produce metabolites with anti-viral properties (Abolnik et al, 2021). The ostrich tracheal swabs in the present study had been stored in an appropriate protein-rich VTM that included antimicrobials; therefore we used them to determine the limits of detection for virus isolation from ostrich tracheal swabs.…”

“…Sampling to identify the agent by real-time quantitative RT-PCR (RT-qPCR) usually initiates only after antibodies are detected (DAFF, 2012) and only rarely after observation of clinical signs or increased mortality rate. Serologic surveillance indicates that AIV infection with multiple subtypes are common in ostriches and occur seasonally (Abolnik et al, 2016), but when the flock is revisited for sampling, the agent is often not detected, or present in too low quantities for identification using molecular methods and/ or virus isolation (Pieterse et al, 2021). This has led to speculation that AIVs circulate at very low levels in ostriches (below the limit of current detection targets), or for short periods of time and/ or that transmission within the flock is poor; but this may not necessarily be the case.…”