Development and Optimization of a Biopreparedness Protocol for Extracting and Detecting Avian Influenza Virus in Broiler Chicken Meat

Abstract

Detection of avian influenza virus (AIV) in poultry meat is hampered by the lack of an efficient analytical method able to extract and concentrate viral RNA prior to PCR. In this study we developed a method for extracting and detecting AIV from poultry meat by a previously standardized 1-step real-time reverse transcriptase PCR (RRT-PCR) assay. In addition, a new process control, represented by feline calicivirus (FCV), was included in the original protocol, to evaluate all analytical steps from sample preparation to the detection phase. The detection limit was below 1×10⁻¹ TCID₅₀ of AIV per sample, and the quantification limit corresponded to 1×10¹ TCID₅₀ of AIV per sample. Moreover, the addition of 1×10² TCID₅₀/sample of FCV did not affect the quantification and detection limit of the reaction. These results show that the developed assay is suitable for detecting small amounts of AIV in poultry meat. In addition, the developed biopreparedness protocol can be applied to detect AIV in legal or illegal imported broiler chicken meat. The availability of a rapid and sensitive diagnostic method based on molecular identification of AIV in poultry meat provides an important tool in the prevention of AIV circulation.

A wide range of wild and domestic birds, including chickens, ducks, and turkeys, can be infected by avian influenza viruses (AIVs). AIV has become the subject of much attention in the past decade, owing to the emergence of a highly pathogenic avian influenza (HPAI) virus A/H5N1 circulating in Asia, Africa, and Europe and to the recent human fatalities caused by a low pathogenic (LP) AIV, A/H7N9, circulating in China.^1,2 The ability of HPAI A/H5N1 to jump from avian to mammalian species, including humans, pigs, horses, captive tigers, leopards, domestic cats, and dogs, represents a major concern for human health.^3-6

Pathogenicity of AIV in poultry ranges from asymptomatic, localized infections to the development of a systemic disease that can range from low to high mortality.⁷ A challenge with LPAIV is that birds do not always show clinical symptoms of the disease. In contrast to LPAIV strains, which have been recovered mainly from respiratory and gastrointestinal tracts of infected poultry, HPAI H5N1 viruses have been isolated from the brain, blood, bone, breast, and thigh meat.⁸

Consumption of uncooked meat could represent a potential risk for human health, as suggested by fatal AIV infections in domestic cats, tigers, and leopards fed uncooked poultry carcasses in laboratory tests and in zoos.^9-11

Given that AIVs are relatively heat sensitive^12,13 compared to other viral pathogens that commonly cause foodborne illness, the Word Health Organization (WHO) recommends cooking poultry meat to a core temperature of at a least 60°C.¹⁴ Moreover, international trade restrictions protect the food chain against dissemination of food pathogens, such as viruses, that originate in meat from infected animals.¹⁵ The European Food Safety Agency (EFSA) and the US Department of Agriculture's Animal and Plant Health Inspection Service (APHIS) have identified legal and illegal imported poultry commodities contaminated with AIV as the most important risk factors for the possible emergence of AIV in the European Union and the United States; in fact, this has already occurred.^16,17 These events emphasize the need for food controls to prevent accidental importation of poultry meat contaminated by AIVs.

The availability of a rapid and sensitive diagnostic method based on molecular identification of AIV in poultry meat would provide an important tool for the safety of international trade.¹ In previous work, we have standardized a 1-step reverse transcription real-time PCR (RRT-PCR) for AIV detection, including a commercial internal positive control (IPC).¹⁸ The inclusion of IPC using an exogenous target sequence of a nucleic acid is crucial to excluding false-negative results caused by the presence of PCR inhibitors.¹⁹ However, sample preparation and RNA purification steps are not monitored by the conventional RNA-based IPC methods.²⁰

The inclusion of a target monitoring not only the enzymatic amplification reaction, but also all analytical steps from sample preparation to the detection phase, would add much more certainty to the interpretation of diagnostic results for food pathogen detection. For this purpose, addition of feline calicivirus virus (FCV) as a process control (PC) has been recently described for the detection of RNA viruses in food.^21,22 The addition of a PC has also been described in the recently published ISO ISO/PRF TS 15216-2.²³ However, since an efficient extraction method from meat was not included in this document, this article represents an important contribution in the field of biopreparedness and food microbiology. In the current study, a complete procedure to extract, concentrate, and detect AIVs from poultry meat by a RRT-PCR method including a PC was developed and standardized.

Materials and Methods

Viruses and Cells

LPAI A/Ck/It/9097/97 (H5N9) virus (from the repository of Istituto Superiore di Sanità laboratories) was propagated in the allantoic cavities of 11-day-old embryonated chicken eggs to produce working stocks of the virus. Virus stocks were titrated on Madin Darby canine kidney (MDCK) cells (ATCC CCL-34), and the 50% tissue culture infectious doses (TCID₅₀) calculated in a modified ELISA assay that identifies the expression of influenza A virus nucleoprotein in infected cells.¹⁸ Feline calicivirus virus (FCV, ATCC VR-782, strain F9) was propagated in Crandell-Reese feline kidney (CRFK) cells (ATCC CCL-94).²¹

The infectivity titer of FCV was determined as the 50% tissue culture infectious dose (TCID₅₀) per milliliter²⁴ using 10-fold serial dilutions in 24-well Multiwell™ (BD Falcon, Franklin Lakes, NJ, USA). CRFK cells were plated at a concentration of 10⁴ cells per well in Eagle's Minimum Essential Medium with 2% (v/v) bovine serum, and 100 μl of viral serial dilutions of virus were added per well, with 4 replicates per dilution. The plates were incubated for 7 days at 37°C with 5% CO₂ in air atmosphere and checked daily for characteristic cytopathic effects of FCV. Viruses were diluted to a final concentration of 10⁷ TCID₅₀/mL and stored at −80°C until use.

Meat Samples

Poultry breast, dissected by a commercial hybrid of heavy chicken (Gallus gallus) at about 50 days of age (broiler), was homogenized in a Pulse Matic blender (Osterizer, Ontario, Canada) with an equal amount of tri-distilled sterile water. The homogenized poultry meat was divided in portions of 900 μL each in 2 mL tubes and stored at −80°C until use.

Spiking of Meat Samples

Poultry meat samples were spiked with various AIV and FCV dilutions according to the following scheme:

• Seven samples of homogenized meat, each of 900 μL, were spiked with 100μL of 10-fold dilutions of AIV stock (in PBS), to reach final concentration titers ranging from 10⁶ to 10⁰ TCID₅₀ mL⁻¹.

• Seven samples of homogenized meat, each of 900 μL, were spiked with 100 μL of 10-fold dilutions of FCV viral stock to reach final concentrations ranging from 10⁶ to 10⁰ TCID₅₀ mL⁻¹.

• Seven samples of homogenized meat, each of 900 μL, were spiked with 100 μL of 10-fold dilutions of AIV viral stock to reach final concentrations ranging from 10⁶ to 10⁰ TCID₅₀ Ml⁻¹.

• One hundred microliters of FCV suspensions were added to each sample to reach a final concentration of 10³ TCID₅₀ mL⁻¹.

Each series of experiments, including a noninoculated homogenized meat sample as negative control, was repeated 3 times.

Viral RNA Extraction

One hundred microliters of spiked homogenized poultry meat samples were transferred into new tubes, and 750 μL of Trizol (Invitrogen, USA) and 150 μL of RNase-free water were added. After vortexing for 15 sec, 200 μL of chloroform was added, and samples were incubated at room temperature for 7 min.

Samples were then centrifuged in a microfuge at 13,000×rpm for 15 min at 4°C, and supernatant was transferred into new tubes containing 500 μL of isopropanol. One μg of glycogen (Fermentas) was added to each sample; the mixture was then incubated at room temperature for 10 min and stored at −20°C overnight. Each sample was then centrifuged in a microfuge at 11,600×rpm for 10 min at 4°C and the supernatant discarded. Each pellet was further washed in 500 μL of 80% ethanol. After spinning at 11,600×rpm in a microfuge for 5 min at 4°C, each supernatant was discarded; the pellet was dried by Vacuum Concentrator (Eppendorf) and finally resuspended in 50 μL of RNase-free water. After 1 hour at −20°C, 1 μL of RNase Inhibitor (Qiagen) was added to each sample. The samples were stored at −70°C until use.

Primers and Probes

The primers and probes used to detect AIV and FCV were previously described by Di Trani et al. and Di Pasquale, respectively.^18,21

One-Step Real-Time Reverse Transcriptase PCR

AIV and FCV detections were performed separately in 2 different vials in a final volume of 25 μL, including 5 μL of viral RNA, 1 X Superscript III Platinum One-step qRT-PCR reaction mix (Superscript III Platinum One-step qRT-PCR kit, Invitrogen), 0.5 μL of ROX reference dye, 0.4 μM of primers, and 0.2 μM of probe for AIV RNA detection or 0.9 μM of primers and 0.1 μM of probe for FCV RNA amplification. Real-time PCR was performed in an ABI Prism 7000 SDS Real-Time apparatus (Applied BioSystems) using 1 cycle of reverse transcription at 45°C for 30 min and 95°C for 2 min, followed by 40 cycles at 95°C for 15 sec and 60°C for 1 min. Negative and positive controls were routinely included in each experiment.

Each fluorescent reporter signal was measured against the internal reference dye (ROX) signal to normalize for non-PCR-related fluorescence fluctuations between samples. The data were collected at the annealing step of each cycle, and the threshold cycle (Ct) for each sample was calculated by determining the point at which the fluorescence exceeded the threshold limit.

Standard Curves

Regression curves performed by testing different concentrations of AIV in spiked poultry meat samples were used to define the limits of detection (LOD) and quantification (LOQ) of the assay. The FCV regression curve obtained from spiked poultry meat samples was used to define the optimal concentration of PC. The regression curve obtained from poultry meat samples spiked with both viruses was used to evaluate the performance of the PC (FCV) at different concentration of the target virus (AIV). The regression curve was evaluated when the linearity of the curve was guaranteed. The amplification efficiency (E) was calculated using the equation: E=(10^−1/slope)−1; the recovery rates of AIV and FCV were evaluated using the equation R=2^−ΔCt.²⁵

Analysis of the Results

Analysis of variance (ANOVA), log-linear regression analysis, and tests for parallelism were undertaken using GraphPad Prism software version 5.0 (GraphPad Software, Inc., La Jolla, CA, USA) using default parameters. Regression line slopes, intercepts, and Ct values of different experiments were compared using the one-way ANOVA with Bonferroni post hoc comparison. The null hypothesis was rejected with a value lower than 0.05.

Results

The LOD of AIV in poultry meat samples, without the addition of FCV, was 1×10⁻¹ TCID₅₀/sample, while the LOQ was 1×10¹ TCID₅₀/sample. A good linearity (R²=0.9913) and a reaction efficiency of 100.9% of the curve was obtained using poultry meat samples spiked with logarithmic concentration of AIV, between 1×10¹ and 1×10⁵ TCID₅₀/sample (Table 1).

Table 1.

Regression curve performed by testing independently different concentrations of AIV or FCV in spiked poultry meat samples to detect the limit of detection and specificity of each assay

Concentration (virus TCID₅₀/sample)	Concentration (virus TCID₅₀/reaction)^*	AIV (Mean Ct±SD) ^**	FCV (Mean Ct±SD) ^**
1×10⁵	10⁴	22.62±0.27	18.26±0.69
1×10⁴	10³	25.01±0.37	22.12±0.26
1×10³	10²	28.17±0.10	26.37±0.40
1×10²	10	32.50±0.32	28.81±0.25
1×10¹	1	35.50±0.46	31.31±0.80
1×10⁰	0.1	36.35±0.48	34.87±0.19
1×10⁻¹	0.01	36.47±1.68	38.05±0.77
0 (no-spiked sample)	0	>45	>45
Curve parameters^a
Slope±SD		−3.299±0.178	−3.537±0.287
Y-intercept±SD		38.63±0.60	36.17±0.80
R²		0.9913	0.9682
Efficiency (%)		100.9	91.77

Estimated number of virus TCID₅₀ in each PCR run, assuming 100% nucleic acid extraction efficiency (each reaction contained 5 μL of a nucleic acid preparation of 50 μL extracted from 100 μL of meat).

Cycle number at which fluorescence intensity equals a fixed threshold. Mean value±standard error of the mean.

Evaluated for AIV using the curve included between 1×10¹ and 1× 10⁵ TCID₅₀ mL⁻¹.

The value of the standard curve obtained using poultry meat samples spiked with 10-fold dilutions of FCV showed, in a range between 1×10⁰ and 1×10⁵ TCID₅₀/sample, an acceptable linearity (R²=0.9682) and a reaction efficiency of 91.77% (Table 1).

Table 2 shows the results of the experiments on poultry meat samples spiked with 10-fold dilutions of AIV and FCV at the most suitable concentration to function as PC (1×10² TCID₅₀/sample with an expected value of Ct between 28.56 and 29.05 with 95% of probability). The regression curve obtained showed a good linearity (R²=0.9832) and a reaction efficiency of 92.93% in the range included between 1×10¹ and 1×10⁵ TCID₅₀/sample of AIV (Table 2).

Table 2.

Ct values obtained by RRT- PCR from poultry meat samples spiked with 10-fold dilutions of AIV in the presence of FCV (ISPC) at fixed concentration 10² TCID₅₀/sample

Concentration (virus TCID₅₀/ sample)	Concentration (virus TCID₅₀/ reaction) ^*	AIV (Mean Ct±SD) ^**	FCV (Mean Ct±SD) ^**
1×10⁵	10⁴	22.10±0.24	29.08±0.53
1×10⁴	10³	25.00±0.24	28.54±0.32
1×10³	10²	27.30±0.15	28.75±0.37
1×10²	10	32.16±0.38	28.48±0.44
1×10¹	1	36.04±0. 40	28.69±0.29
1×10⁰	0.1	36.53±0. 28	29.16±0.23
1×10⁻¹	0.01	36.14±0.60	28.69±0.22
0 (no-spiked sample)	0	>45
Curve parameters^a
Slope±SD		−3.504±0.265
Y-intercept±SD		39.03±0.88
R²		0.9832
Efficency (%)		0.9293

Cycle number at which fluorescence intensity equals a fixed threshold. Mean value±standard error of the mean.

Evaluated for AIV using the curve included between 1×10¹ to 1×10⁵ TCID₅₀ mL⁻¹ and a fixed concentration of 10² TCID₅₀/sample.

The LOD of AIV in poultry meat, with the addition of PC, was 1×10⁻¹ TCID₅₀/sample, and the LOQ was 1×10¹ TCID₅₀/sample. The slope of linear regression curves of 10-fold dilutions of AIV obtained both without the PC (Tables 1 and 2) and with PC at a concentration of 1×10² TCID₅₀/samples did not show any statistically significant difference (p=0.8375).

All negative controls were constantly negative, giving no reaction with meat-derived components, thus demonstrating a clear specificity of the method for the detection of virus template. However, the statistical analysis showed a statistically significant difference between the elevators (p<0.01). The Y-intercept of linear regression curves expected values, without and with PC, ranged from 36.75 to 40.51 (95% of probability) and from 36.23 to 41.83 (95% of probability), respectively. Finally, no statistically significant difference was observed in the Ct values for the FCV, with different concentration of AIV (p=0.21) (Table 2).

Discussion and Conclusions

A major challenge in food molecular diagnostics is the availability of rapid and highly specific assays. Although viral foodborne disease is a significant problem, foods are rarely tested for viral contamination, and such testing is usually limited to shellfish. Frequently, the cause of an outbreak is suspected to be of viral origin, but because of the lack of sensitive and reliable methods, this hypothesis rarely can be confirmed by experimental results.

Typical viral outbreaks include a large number of cases and can create international and national crises in the food trade. Therefore, more sensitive molecular assays are required to detect viruses in food samples characterized by low viral loads; in this context, RT-PCR and RRT-PCR represent the methods of choice for detecting enteric viruses in food samples.

Among multiple RRT-PCR formats described for the identification of AIV in field specimens, some of them use a one-step RRT-PCR procedure.²⁶ The performance of RRT-PCR to detect viral contamination in food is dramatically affected by an efficient RNA extraction method; in fact, several substances of the meat matrix can be released in the extracting solution, thus decreasing either the efficiency of virus recovery or inhibiting nucleic acid amplification. So far, few protocols have reported efficient viral extraction methods specific to food matrices, including poultry meat. The Trizol procedure of vRNA extraction proposed in this article is a modification of the classic single-step acid guanidium-thiocyanate-phenol-chloroform extraction technique,²⁷ allowing the recovery of a vRNA free of DNA.^28,29

Analysis of the linear regression curves of AIV showed a weak interference of the chosen PC amount with the detection of AIV, since the difference between the elevators of the 2 curves (without and with the addition of PC) are statistically significant (p<0.01). This difference, however, does not modify the performance of the RRT-PCR assay, since the range of quantification and the LOD (below 1×10⁻¹ TCID₅₀/sample) are not affected.

Our results show that the developed assay is suitable for the detection of low AIV concentration in muscle from domestic poultry, providing a useful protocol for the vRNA extraction from meat samples. In this respect the knowledge is lacking, since the majority of studies take into consideration mainly mussels and vegetables as food matrices. In particular, the standard for the quantitative and qualitative detection of enteric viruses (noroviruses and HAV) produced by the European Committee on Standardisation (CEN)^23,20 includes extraction from only a few food matrices (mussels, vegetables, and bottled water).

This study provides a useful biopreparedness contribution: the development of a new standard method to detect AIV in poultry meat matrices. The circulation of A/H7N9 LPAI virus in China, associated with a large number of human infections and deaths and in the absence of large avian influenza outbreaks in poultry farms, underlines the relevance of this study, which may be useful in the control of poultry meat.

Footnotes

Acknowledgments

This research was supported by the framework of the EU project AniBioThreat (Grant Agreement: Home/2009/ISEC/AG/191) with financial support from the Prevention of and Fight against Crime Programme of the European Union, European Commission—Directorate General Home Affairs. This publication reflects the views only of the authors, and the European Commission cannot be held responsible for any use that may be made of the information contained therein.

References

Beato

, Capua

, Alexander

. Avian influenza viruses in poultry products: a review. Avian Pathol, 2009 Jun 38,3:193–200.

Liu

, Shi

et al. Origin and diversity of novel avian influenza A H7N9 viruses causing human infection: phylogenetic, structural, and coalescent analyses. Lancet, 2013 June 1,38 9881:1926–1932.

te Beest

, Stegeman

, Mulder

, van Boven

, Koopmans

. Exposure of uninfected poultry farms to HPAI (H7N7) virus by professionals during outbreak control activities. Zoonoses Public Health, 2011 Nov 58,7:493–499.

Paul

, Wongnarkpet

, Gasqui

et al. Risk factors for highly pathogenic avian influenza (HPAI) H5N1 infection in backyard chicken farms, Thailand. Acta Trop, 2011 Jun 118,3:209–216.

Smith

. Predictability and preparedness in influenza control. Science, 2006 Apr 21,312 5772:392–394.

Rimmelzwaan

, van Riel

, Baars

et al. Influenza A virus (H5N1) infection in cats causes systemic disease with potential novel routes of virus spread within and between hosts. Am J Pathol, 2006 Jan 168,1:176–183quiz 364.

Swayne

, Slemons

. Using mean infectious dose of high- and low-pathogenicity avian influenza viruses originating from wild duck and poultry as one measure of infectivity and adaptation to poultry. Avian Dis, 2008 Sep 52,3:455–460.

Swayne

, Beck

. Experimental study to determine if low-pathogenicity and high-pathogenicity avian influenza viruses can be present in chicken breast and thigh meat following intranasal virus inoculation. Avian Dis, 2005 Mar 49,1:81–85.

Keawcharoen

, Oraveerakul

, Kuiken

et al. Avian influenza H5N1 in tigers and leopards. Emerg Infect Dis, 2004 Dec 10,12:2189–2191.

10.

Kuiken

, Rimmelzwaan

, van Riel

et al. Avian H5N1 influenza in cats. Science, 2004 Oct 8,306 5694:241.

11.

Simmerman

, Embarek

PKB

. Avian influenza A (H5N1): potential threat to food safety. Heredia

, Wesley

, Garcia

. Microbiologically Safe Foods. New York: Wiley, 2009.

12.

Thomas

, Swayne

. Thermal inactivation of H5N1 high pathogenicity avian influenza virus in naturally infected chicken meat. J Food Prot, 2007 Mar 70,3:674–680.

13.

Swayne

. Microassay for measuring thermal inactivation of H5N1 high pathogenicity avian influenza virus in naturally infected chicken meat. Int J Food Microbiol, 2006 Apr 25,108 2:268–271.

14.

WHO. Highly pathogenic H5N1 avian influenza outbreaks in poultry and in humans: food safety implications. November 4, 2005. http://www.who.int/foodsafety/fs_management/No_07_AI_Nov05_en.pdf. 2013 July 15.

15.

Arzt

, White

, Thomsen

, Brown

. Agricultural diseases on the move early in the third millennium. Vet Pathol, 2010 Jan 47,1:15–27.

16.

Beato

, Capua

. Transboundary spread of highly pathogenic avian influenza through poultry commodities and wild birds: a review. Rev Sci Tech, 2011 Apr 30,1:51–61.

17.

van den Berg

. The role of the legal and illegal trade of live birds and avian products in the spread of avian influenza. Rev Sci Tech, 2009 Apr 28,1:93–111.

18.

Di Trani

, Bedini

, Donatelli

et al. A sensitive one-step real-time PCR for detection of avian influenza viruses using a MGB probe and an internal positive control. BMC Infect Dis, 2006; 6:87.

19.

Hoorfar

, Cook

, Malorny

et al. Making internal amplification control mandatory for diagnostic PCR. J Clin Microbiol, 2003 Dec 41,12:5835.

20.

Rossmanith

, Wagner

. The challenge to quantify Listeria monocytogenes—a model leading to new aspects in molecular biological food pathogen detection. J Appl Microbiol, 2010 Mar 110,3:605–617.

21.

Di Pasquale

, Paniconi

, De Medici

, Suffredini

, Croci

. Duplex real time PCR for the detection of hepatitis A virus in shellfish using feline calicivirus as a process control. J Virol Methods, 2011 Jan 163,1:96–100.

22.

Rossmanith

, Mester

, Fruhwirth

, Fuchs

, Wagner

. Proof of concept for recombinant cellular controls in quantitative molecular pathogen detection. Appl Environ Microbiol, 2011 Apr 77,7:2531–2533.

23.

International Organization for Standardization. Microbiology of food and animal feed—Horizontal method for determination of hepatitis A virus and norovirus in food using real-time RT-PCR—Part 2: Method for qualitative detectionISO/TS 15216-2:2013http://www.iso.org/iso/home/store/catalogue_tc/catalogue_detail.htm?csnumber=60297. 2013 July 15.

24.

Reed

, Muench

. A simple method of estimating fifty percent endpoints. Am J Hygiene, 1938; 27:493–497.

25.

Fleige

, Walf

, Huch

, Prgomet

, Sehm

, Pfaffl

. Comparison of relative mRNA quantification models and the impact of RNA integrity in quantitative real-time RT-PCR. Biotechnol Lett, 2006 Oct 28,19:1601–1613.

26.

Pasick

. Advances in the molecular based techniques for the diagnosis and characterization of avian influenza virus infections. Transbound Emerg Dis, 2008 Oct 55,8:329–338.

27.

Chomczynski

, Sacchi

. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem, 1987 Apr 162,1:156–159.

28.

Xiang

, Qiu

, Hegele

, Tan

. Comparison of different methods of total RNA extraction for viral detection in sputum. J Virol Methods, 2001 May 94,1-2:129–135.

29.

Chomczynski

, Sacchi

. The single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction: twenty-something years on. Nat Protoc, 2006; 1,2:581–585.

30.

Lees

. International standardisation of a method for detection of human pathogenic viruses in molluscan shellfish. Food Environ Virol, 2010; 2,3:146–155.