Food and Environmental Virology

, Volume 4, Issue 2, pp 68–72

Comparative Analysis of Viral Concentration Methods for Detecting the HAV Genome Using Real-Time RT-PCR Amplification


  • Kang Bum Lee
    • Department of Food and Nutrition, School of Food Science and TechnologyChung-Ang University
  • Hyeokjin Lee
    • Division of Hepatitis VirusKorea Center for Disease Control and Prevention
  • Sang-Do Ha
    • School of Food Science and TechnologyChung-Ang University
    • Division of Hepatitis VirusKorea Center for Disease Control and Prevention
    • Department of Food and Nutrition, School of Food Science and TechnologyChung-Ang University
    • School of Food Science and TechnologyChung-Ang University
Original Paper

DOI: 10.1007/s12560-012-9077-x

Cite this article as:
Lee, K.B., Lee, H., Ha, S. et al. Food Environ Virol (2012) 4: 68. doi:10.1007/s12560-012-9077-x


Hepatitis A is a major infectious disease epidemiologically associated with foodborne and waterborne outbreaks. Molecular detection using real-time RT-PCR to detect the hepatitis A virus (HAV) in contaminated vegetables can be hindered by low-virus recoveries during the concentration process and by natural PCR inhibitors in vegetables. This study evaluated three virus concentration methods from vegetables: polyethylene glycol (PEG) precipitation, ultrafiltration (UF), and immunomagnetic separation (IMS). UF was the most efficient concentration method, while PEG and IMS were very low for the recovery rate of HAV. These results demonstrate that UF is the most appropriate method for recovering HAV from contaminated vegetables and that this method combined with the real-time RT-PCR assay may be suitable for routine laboratory use.


Hepatitis A virus (HAV)ConcentrationDetectionUltrafiltration (UF)Immunomagnetic separation (IMS)


Hepatitis A virus (HAV) is a foodborne enteric virus that can cause acute hepatitis in humans (Boxman et al. 2007; Koopmans et al. 2002). HAV is a 27-nm diameter positive-sense, non-enveloped, single-stranded RNA genome picornavirus and one of the primary causes of foodborne disease worldwide. HAV is primarily transmitted by the fecal-oral route, either by direct person-to-person contact or by ingestion of contaminated food or water (Koopmans et al. 2002; Koopmans and Duizer 2004). Approximately 10 million people worldwide are infected with HAV every year and in about 50% of these cases the infectious source remains unidentified. Only 2–3% of reported cases are identified as part of a recognized foodborne outbreak, although a considerable percentage of sporadic cases might be foodborne (Fiore et al. 2006; Flore 2004; McCaustland et al. 1982).

The majority of foodborne outbreaks occur via fecal-contaminated food. Although this means that any food handled under poor hygienic practices could potentially be contaminated with foodborne pathogens including HAV, bivalve molluscan shellfish such as oysters, cockles, and mussels are the most common source of foodborne viruses (Kittigul et al. 2008).

Recently, fresh produce such as salads, fresh fruits, and vegetables have also been implicated in foodborne outbreaks of hepatitis A. These products are likely to be consumed raw or lightly cooked, and can become contaminated with fecal matter at almost any point during growing, harvesting, transport, or packing (Guillois-Becel et al. 2009; Hooper et al. 1977; Petriqnani et al. 2010; Pinto et al. 2009; Robesyn et al. 2009; Rowe et al. 2009).

PCR is the primary method of detecting enteric viruses in oysters or vegetables implicated in foodborne outbreaks, but PCR can be problematic due to insufficient viral recovery and the presence of PCR inhibitors (Gilpatrick et al. 2000). The most crucial step of viral detection in oysters or vegetables is the sample treatment process (Bidawid et al. 2000; Dubois et al. 2007; Papafragkou et al. 2008; Fino and Kniel 2008). This is because HAV and other enteric viruses may be found in large quantities in stool, but are usually present in low quantities in foods. Therefore, various methods have been developed to capture or concentrate these viruses prior to testing for their presence in foods (Moon et al. 2009; Choi 2005; Ha et al. 2009; Jothikumar et al. 2005; Butot et al. 2007). The objective of present study was to compare three virus recovery methods for HAV detection in leafy vegetables: polyethylene glycol (PEG) precipitation, ultrafiltration (UF), and immunomagnetic separation (IMS).

Materials and Methods

Virus Origin and Sample Inoculation

A HAV-positive stool sample from a patient was provided by Korea Centers for Disease Control and Prevention (KCDC). Based on the sequence analysis of VP3-VP1 region of detected HAV genome, HAV-positive sample was clustered in HAV genotype IIIA of genus Hepatovirus. HAV-positive feces was suspended with phosphate-buffered saline (1:10 ratio) and centrifuged at 6,000×g for 1 min. The supernatant was transferred to new stock tube. The amount of HAV RNA was determined as 1.2 × 107 copies/ml by real-time RT-PCR in used fecal suspension.

The virus inoculation was performed according to the procedure described by Moon et al. (2009) with a few modifications. Briefly, 5 g of fresh lettuce and perilla leafs, obtained from commercial sources, were inoculated by direct application of 50 μl of an HAV fecal suspension and allowed to dry for 4 h in a laminar box.

Virus Concentration

The artificially contaminated vegetable was washed with 50 ml of 0.25 M threonine–0.3 M NaCl (pH 9.5) in 3D-Shaker (VWR international, Radnor, PA, USA) for 5 h. The wash-off was used to compare three methods of virus concentration from the above-mentioned lettuce and perilla leafs: PEG, UF, and IMS (Moon et al. 2009). As negative control of virus concentration, non-seeded fresh lettuce and perilla leafs were used. HAV-positive stool sample originating from patient was used as positive control of whole procedure. Positive and negative controls underwent the same treatment as artificially contaminated samples. All samples were analyzed at least three times.

PEG Precipitation

The vegetable wash-off from lettuce or perilla leaf was separated by centrifugation (6,000×g, 20 min, 4°C), and 25 ml of PEG 8,000 (40%) and 10 ml of 3 M NaCl were added to the reaction before incubating at 4°C for 16 h. The precipitated virus particles were separated by centrifugation (16,000×g, 20 min, 4°C). The pellet was resuspended with 5 ml of supernatant and mixed with 5 ml chloroform. After centrifugation at 10,000×g for 30 min at 4°C, the supernatant was precipitated again by adding 5 ml of PEG 8,000 (40%) and 2 ml of 3 M NaCl, then incubated at 4°C for 3 h. The precipitated virus particles were separated by centrifugation at 35,000×g for 30 min, and the final pellet was resuspended in 500 μl of diethyl pyrocarbonate (DEPC)-treated water (Moon et al. 2009; Ha et al. 2009).

UF Membrane

Ultrafiltration was used to concentrate the viruses with Vivaspin 20 filters (100,000 nominal molecular weight limit; Sartorius stedim, French). Vegetable wash-off containing HAV in a 50-ml Vivaspin 20 tube was concentrated to a volume of 250–500 μl by centrifugation at a maximum speed of 10,000×g at 4°C for 10 to 40 min, depending on the viscosity of the sample. The virus concentrate was stored at 4°C until viral RNA was extracted.

Immunomagnetic Separation (IMS)

Thirty microliter of magnetic beads (Dynabeads M-280, tosyl activated; Dynal, Norway) was washed three times with 0.1 M sodium phosphate buffer (pH 7.4) at room temperature. The washed beads were suspended gently in 10 μg goat anti-HAV polyclonal antibody (GenWay Biotech, San Diego, CA, USA) and mixed with vegetable wash-off. Two hours incubation at room temperature was performed with slow tilt rotation on an MX4 sample mixer (Dynal, Norway). In order to remove unbound polyclonal antibody, the virus-bound immunomagnetic beads were washed four times with PBS (pH 7.4) containing 0.1% (wt/vol) bovine serum albumin (BSA; Sigma) at room temperature. Antibody-bound magnetic beads were collected gently with 500 μl of NuPAGE buffer (Invitrogen, Carlsbad, CA, USA) with rotation on MCB1200 processing system (Sigris research, Brea, CA, USA) for 10 min at room temperature. The total antibody-bound beads complex was submitted to RNA extraction and realtime RT-PCR.

Viral RNA Extraction

Viral RNA extraction using the QIAamp viral RNA mini kit (Qiagen, Hilden, Germany) was performed as described by the manufacturer. Viral RNA was eluted in 50 μl of AVE buffer and stored at −80°C until use.

Real-Time RT-PCR

For the RT-PCR control, the PCR product from HAV HM175 was purified with a PCR purification kit (QIAGEN, Valencia, CA, USA) and cloned into TA cloning kit (QIAGEN, Valencia, CA, USA) which have SP6 and T7 promoter. HAV RNA was transcribed with T7 RNA polymerase as described in the manufacturers’ protocol (Ambion Inc., Austin, TX, USA). HAV RNA transcripts was adjusted to 2 × 1010 copy/ml and used to prepare quantitative RT-PCR standard.

Real-time PCR for HAV detection was performed with the MX3005p model using previously described methods (Jothikumar et al. 2005). HAV amplifications were performed with a one-step real-time RT-PCR kit (Agilent; Santa Clara, CA, USA). The real-time RT-PCR reaction mixture contained 5 μl of extracted RNA, 12.5 μl of 2× Mastermix (Agilent; Santa Clara, CA, USA), 1 μl of each primer (10 pmol), 0.5 μl of each fluorescent probes (10 pmol), 1 μl of RT/Rnase block mixture, and 4 μl of diethyl pyrocarbonate-treated distilled water to make 25 μl reaction volume for the reaction. The primers and probe were manufactured from applied biosystems Inc. (ABI; Foster city, CA, USA). The temperature and time parameters were as follows: RT for 30 min at 50°C, denaturation for 10 min at 95°C, and amplification (40 cycles of 15 s at 95°C and one min at 60°C). Real-time RT-PCR was performed three times.

Results and Discussion

Control HAV RNA was extracted from a stool sample collected from an HAV-positive patient and detected with real-time RT-PCR for quantitative analysis. The copies number was 1.2 × 107 copies/ml. Tenfold diluted stool samples were used for comparing three concentration methods.

The ability of each of the three virus concentration methods to recover HAV from a contaminated food item is shown in Table 1. The loss of virus particles during virus concentration and viral RNA extraction procedures was estimated by real-time RT-PCR. For the lettuce samples, depending on the method, a loss of 103 to 105 real-time RT-PCR detectable HAV genome copies was observed. Based on these results, the UF appears to be the most efficient method for HAV recovery and is simple, less time-consuming and relatively easy to perform. In contrast, the PEG method is more time-consuming and requires a large number of organic reagents.
Table 1

Real-time RT-PCR results of three hepatitis A virus concentration methods in virus-seeded lettuce and perrila leaf

Food matrix


Mean of HAV genome copies (copies/ml)

Virus recovery rate (%)


Polyethylene glycol precipitation

6.8 × 103



7.2 × 104


Immunomagnetic separation

7.8 × 102


Perrila leaf

Polyethylene glycol precipitation

6.7 × 103



4.9 × 104


Immunomagnetic separation

8.0 × 102


Polyethylene glycol (PEG) precipitation, ultrafiltration (UF), and immunomagnetic separation (IMS) are all applicable in detecting viruses in naturally contaminated food using conventional or nested RT-PCR (Ha et al. 2009; Fino and Kniel 2008; Moon et al. 2009; Myrmel et al. 2000). Recently, IMS was successfully applied to HAV RNA extraction from contaminated food prior to a real-time RT-PCR-based HAV detection (Jothikumar et al. 1998). UF has not yet been applied to naturally contaminated food but the results of this study show that the method is promising.

Hepatitis A viruses are highly infectious and it is important to be able to detect even low levels of the virus in vegetables and other food items. Conventional and real-time RT-PCR have become favored detection tools when investigating outbreaks of gastroenteritis with suspected viral origin (Love et al. 2008; Jothikumar et al. 2005). Unfortunately, in nested RT-PCR, there is a high risk of contaminating samples after the 1st amplification. SYBR Green and Taqman-based RT-PCR assays can reduce the risk of contamination and provide faster results, because no gel electrophoresis is required and additional confirmation of amplicons is not necessary. Therefore, real-time RT-PCR is faster, less laborious, and lowers the risk of cross-contamination during detection.

The possibility of improving the sensitivity and simplicity of IMS was initially investigated. IMS has been shown to concentrate HAV and minimize inhibitors (Rzezutka et al. 2006). The inhibitors were removed by multiple wash steps during the IMS procedure. The major drawback of this method, however, is that multiple specific antibodies are required to bind HAV antigens.

The amount of virus present in HAV-contaminated ready-to-eat food is usually lower than the concentrations of the virus that can be efficiently detected by current RT-PCR methods. Additionally, the inhibitors present in many food samples must be removed prior to RT-PCR, thus necessitating complicated concentration and purification procedures. These procedures could help health organizations trace the contamination route along the food consumption chain and identify possible sources of HAV contamination.


This study was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2009-0085532).

Copyright information

© Springer Science + Business Media, LLC 2012