Development of a rapid technique for extraction of viral DNA/RNA for whole genome sequencing directly from clinical liver tissues

https://doi.org/10.1016/j.jviromet.2020.113907Get rights and content

Highlights

  • Whole genome sequencing of hepatic viral pathogens requires high quality nucleic acid which necessitates a time-consuming process involving propagation of the virus in vitro.

  • Extraction of viral nucleic acid directly from infected liver tissues is often unsatisfactory due to high liver fat content and renders low quantity and quantity viral nucleic acid.

  • This study presents a simple treatment procedure with kaolin hydrated aluminium silicate and subsequent enrichment to extract high quality DNA from livers affected with Fowl aviadenovirus.

  • Extracted viral DNA proved suitable for sensitive PCR detection of the virus as well whole genome sequencing.

Abstract

Characterisation of the entire genome of Fowl aviadenoviruses (FAdV) requires isolation and propagation of the virus in chicken embryo liver or kidney cells, a process which is not only time consuming but may occasionally fail to result in viral growth. Furthermore, in a mixed infection, isolation in cell culture may result in the loss of viral strains. In this study, we optimised a FAdV DNA extraction technique directly from affected liver tissues using kaolin hydrated aluminium silicate treatment. The whole genome of FAdV was sequenced directly from extracted DNA without any targetted PCR based enrichment. The extraction method was also tested on avian liver tissues affected with the RNA virus Avian hepatitis E virus and demonstrated to yield sequencing grade RNA. Therefore, the method described here is a simple technique which is potentially useful for the extraction of sequencing grade DNA/RNA from tissues with high fat content.

Introduction

Analysis of the whole genome of viral/bacterial pathogens is the gold standard for typing of strains (Bradley et al., 2015; Didelot et al., 2012; Qin et al., 2010). To circumvent the requirement to isolate and propagate viruses, recent techniques have focussed on isolation of viral DNA for sequencing purposes directly from clinical samples (Croville et al., 2018; Günther et al., 2017; Hasman et al., 2014). However, extraction of sequencing grade viral DNA from some clinical specimens such as blood, faeces and liver can be challenging. In the case of liver tissue this is primarily due to components such as fat that can interfere with the DNA/RNA extraction procedure. In this study, FAdV and Avian hepatitis E virus (AHEV) infected chicken livers were used as models to evaluate the suitability of clinical liver tissues for the extraction of viral DNA and RNA respectively.

FAdVs have double stranded linear DNA with a genome size of 37−45 kb. FAdVs have been grouped into 5 species A–E based on hexon gene restriction fragment length polymorphism (RFLP). Conventionally, FAdVs have been divided into 12 serotypes based on cross-neutralization assays (Hess, 2000). All twelve FAdV serotypes have been detected in association with inclusion body hepatitis (IBH), but members of species D and E are most prevalent (Hess, 2020). IBH, first described in 1963 (Helmboldt and Frazier, 1963), is a severe disease of chickens causing prominent lesions in the liver. IBH is most commonly seen in young broilers between 3 and 7 weeks of age (Hess, 2000), but sometimes in birds as old as three months of age (Howell et al., 1970) or as young as 9 days of age (Steer et al., 2011). It is characterised by a sudden spike in the flock mortality rate over a 3−4-day period, usually followed by gradual decline to the normal levels within 5 days. Occasionally mortality can continue for up to 3 weeks (Ana et al., 2013). Given that clinical signs are absent or mild in IBH, outbreaks that do not substantially increase mortality, can go unnoticed, suggesting that the incidence of IBH may be higher than that estimated through submissions to veterinary diagnostic laboratories.

Identification of the strain/serotype involved in the disease outbreak is imperative in epidemiological tracing and control of disease using vaccination (Steer et al., 2011). Strain typing using conventional microneutralisation assays is time consuming and needs extensive interpretation. Similarly, tests using polymerase chain reaction (PCR) and/or RFLP are cumbersome, and results are difficult to interpret (Meulemans, 2001; Raue and Hess, 1998). More recently, high resolution melt (HRM) curve analysis was developed and used for accurate typing of FAdV reference strains from each of the 12 serotypes (Steer et al., 2009) as well as field specimens (Steer et al., 2011). Also, nucleotide sequencing of the PCR products representing partial or complete Hexon gene loop 1 (L1) has been used for strain identification purposes (Kaján et al., 2012; Pan et al., 2017). However, HRM curve analysis and nucleotide sequencing of PCR products has its own limitations and may not be capable of differentiating FAdV strains belonging to the same serotype/genotype as only small segments of FAdV genome are targetted.

Whole genome sequencing (WGS) is now a popular technology for characterisation of microbial pathogens in research settings. Transferring this technology from research to routine diagnostics will be important for rapid identification and characterisation of pathogens for the selection of control and prevention strategies. Sequencing of whole genomes of FAdV strains/isolates is the most reliable technique for their characterisation and epidemiological tracing but it requires isolation and propagation of the virus in vitro (McFerran and Adair, 1977). For this purpose, FAdV needs to be propagated in primary cell culture (e.g. chicken embryo liver cells) (Hess et al., 1998) followed by nucleic acid extraction (Absalón et al., 2017; Marek et al., 2016). However, viral culture is time-consuming and may occasionally fail to result in viral growth. It can also overlook mixed infections due to preferential growth of one strain over the others (Hess, 2000). Therefore, WGS directly from the clinical isolates can reduce the processing time and circumvent the need to isolate the virus.

This study aims to optimise viral DNA extraction technique from IBH affected liver tissues and to characterise FAdV isolates using WGS directly from clinical samples. The extraction technique was also examined using liver tissues affected by the RNA virus AHEV. To the best of the authors’ knowledge, this study is the first to attempt whole genome sequencing of FAdV isolates directly from clinical tissues.

Section snippets

Origin of FAdV and AHEV samples

Fresh liver tissues from 13-day old broiler flocks in Victoria, Australia with spiking mortality of 0.5 %, gross and microscopic hepatic lesions, and suspected IBH, were submitted to the Asia Pacific Centre for Animal Health laboratories at the University of Melbourne in March 2018. These livers were subjected to FAdV PCR-HRM, as per the method described by Steer et al. (2009) and found to contain FAdV-11. For AHEV, multiple liver tissues were submitted with spotty liver disease in 33–60 weeks

High quality DNA/RNA extracted from liver tissues treated with kaolin hydrated aluminium silicate

The supernatant obtained from kaolin treatment of liver homogenate was clear with lipid tissue precipitated at the bottom of tube (Fig. 1). Supernatant was distinctly separated from cell debris and easily collected for downstream processing. The supernatant obtained from kaolin untreated liver homogenate was turbid due to its high fat content. Additionally there was no visible distinction between pellet and supernatant (Fig. 1). The quantity and quality of DNA and RNA samples were measured by

Discussion

The full genome sequence of FAdV-11 was readily determined in this study directly from liver tissues, without targetted PCR-based enrichment. To the best of authors’ knowledge this is the first study using kaolin for DNA extraction directly from liver tissues for whole genome sequencing. Kaolin has been used to remove non-specific inhibitors from serum samples (Almasi et al., 2011; Nir et al., 1969) and was shown here that it is also useful for liver samples, given the high lipid content of the

CRediT authorship contribution statement

Kinza Asif: Validation, Formal analysis, Investigation, Data curation, Visualization, Writing - original draft. Denise O’Rourke: Methodology. Alistair R. Legione: Software, Writing - review & editing. Penelope A. Steer-Cope: Supervision, Writing - review & editing. Pollob Shil: Supervision, Writing - review & editing. Marc S. Marenda: Software, Supervision, Writing - review & editing. Amir H. Noormohammadi: Conceptualization, Supervision, Writing - review & editing, Project administration,

Declaration of Competing Interest

We have no conflicts of interest to disclose.

Acknowledgments

First author is a recipient of Melbourne research scholarship (MRS) funded by The University of Melbourne, Melbourne, Australia. Internal funding for this study was provided by Asia Pacific Centre for Animal Health (APCAH), The University of Melbourne.

References (35)

  • B. Adair et al.

    Development of a microtitre fluorescent antibody test for serological detection of adenovirus infection in birds

    Avian Pathol.

    (1980)
  • M. Ana et al.

    The first whole genome sequence of a Fowl adenovirus B strain enables interspecies comparisons within the genus Aviadenovirus

    Vet. Microbiol.

    (2013)
  • P. Bradley et al.

    Rapid antibiotic-resistance predictions from genome sequence data for Staphylococcus aureus and Mycobacterium tuberculosis

    Nat. Commun.

    (2015)
  • B.W. Calnek et al.

    Adenoviruses of chickens: serologic groups

    Avian Dis.

    (1975)
  • X. Didelot et al.

    Transforming clinical microbiology with bacterial genome sequencing

    Nat. Rev. Genet.

    (2012)
  • T. Günther et al.

    Recovery of the first full-length genome sequence of a parapoxvirus directly from a clinical sample

    Sci. Rep.

    (2017)
  • H. Hasman et al.

    Rapid whole-genome sequencing for detection and characterization of microorganisms directly from clinical samples

    J. Clin. Microbiol.

    (2014)
  • Cited by (1)

    • Characterisation of the whole genome sequence of an avian hepatitis E virus directly from clinical specimens reveals possible recombination events between European and USA strains

      2021, Infection, Genetics and Evolution
      Citation Excerpt :

      The supernatants were collected and processed for viral RNA purification. The supernatants obtained from centrifugation of liver homogenates were processed for RNA extraction using ultracentrifugation as described previously (Asif et al., 2020). Briefly, the supernatant was layered onto a 30% w/v ice-cold sucrose cushion in an open top tube and virions were pelleted at 98,000 ×g for 90 min at 4 °C.

    View full text