Capture and concentration of viral and bacterial foodborne pathogens using apolipoprotein H

https://doi.org/10.1016/j.mimet.2016.07.014Get rights and content

Highlights

  • Apolipoprotein H is used as a novel capture agent for foodborne pathogens.

  • ApoH binds both viral and bacterial pathogens with high affinity.

  • A unique co-capture system is proposed for one-step virus/bacteria detection.

Abstract

The need for improved pathogen separation and concentration methods to reduce time-to-detection for foodborne pathogens is well recognized. Apolipoprotein H (ApoH) is an acute phase human plasma protein that has been previously shown to interact with viruses, lipopolysaccharides (LPS) and bacterial proteins. The purpose of this study was to determine if ApoH was capable of binding and efficiently capturing two representative human norovirus strains (GI.1 and GII.4), a cultivable surrogate, and four bacterial pathogens (Escherichia coli O157:H7, Listeria monocytogenes, Salmonella enterica serovar Enteritidis, and Staphylococcus aureus). Experiments were carried out using an ApoH-conjugated magnetic bead-based capture followed by pathogen detection using nucleic acid amplification. For all three viruses studied, > 10% capture efficiency (< 1 Log10 loss in RT-qPCR amplifiable units) was observed. The same capture efficiencies were observed for the bacterial pathogens tested, with the exception of E. coli O157:H7 (approximately 1% capture efficiency, or 2 Log10 loss in CFU equivalents). The efficiency of the capture steps did not vary as a consequence of input target concentration or in the presence of an abundance of background microflora. A complementary plate-based capture assay showed that ApoH bound to a variety of human norovirus virus-like particles. ApoH has the potential to be a broadly reactive ligand for separating and concentrating representative foodborne pathogens, both bacteria and viruses.

Introduction

Foodborne pathogens are responsible for approximately 9.4 million illnesses each year in the United States, with an estimated total illness cost of $14.0 billion annually (Batz et al., 2012, Scharff, 2012). About 58% (5.5 million) of these illnesses are caused by human norovirus (Scallan et al., 2011). While the bacterial foodborne pathogens are responsible for fewer illnesses (3.6 million each year), they constitute the majority of serious outcomes (Scallan et al., 2011). Nontyphoidal Salmonella spp., the most common bacterial causative agent, is the leading cause of food-related hospitalizations (35%) and deaths (28%) (Centers for Disease Control & Prevention, 2013), while Listeria monocyotogenes is the third leading cause of foodborne disease deaths (19%) (Scallan et al., 2011, Gandhi and Chikindas, 2007). Two other important foodborne pathogens are shiga toxin-producing Escherichia coli (STEC, for which E. coli O157:H7 is the most common serovar.), and Staphylococcus aureus, which causes self-limiting food intoxication associated with food handling and subsequent temperature abuse (Rangel et al., 2005, Kadariya et al., 2014).

Regulators and industry alike use microbiological testing as one tool to manage these foodborne disease agents. The inability to cultivate human norovirus in vitro means that separation and concentration of low levels of virus particles from food matrices is necessary prior to detection, which is done using molecular amplification methods. Conversely, it is possible to enrich low concentrations of most bacterial pathogens, however enrichment steps are time-consuming, often taking up to two days. Waiting for test results has many implications, both financial (e.g., storage and warehousing, reduced product shelf-life) and public health in nature. In fact, users of testing methods almost universally recognize that the enrichment and/or concentration process is the single most limiting factor in the development of near real-time methods for foodborne pathogen detection (Dwivedi and Jaykus, 2011).

There is a need to develop methods that either shorten enrichment times, or allow bypass of enrichment steps. Separation and concentration of pathogens from the sample matrix prior to detection is necessary in order to achieve these goals. This so-called “pre-analytical sample processing” is of increasing interest to the detection sector and approaches have been reviewed elsewhere (Stevens and Jaykus, 2008, Dwivedi and Jaykus, 2011). The most commonly used methods for pathogen separation and concentration are immuno-magnetic separation (antibody-conjugated paramagnetic beads) or affinity magnetic separation (ligand-conjugated paramagnetic beads) (Fluit et al., 1993a, Fluit et al., 1993b, Liebana et al., 2009, Stevens and Jaykus, 2008). However, depending on the ligand, these methods have their disadvantages, and none is ideal (Dwidedi and Jaykus, 2011). For example, it is very difficult to find broadly reactive ligands for capture of all human norovirus genotypes (and for some bacteria as well); binding affinities can be poor; and cost and/or shelf-life can be problematic. No ligand has binding specificity to a comprehensive and diverse group of foodborne pathogens.

Apolipoprotein H (ApoH), also known as beta2-glycoprotein I, is an acute phase protein found circulating in human plasma (Bouma et al., 1999). ApoH has already been shown to interact directly with a number of different viruses such as hepatitis B, rotavirus and human immunodeficiency virus (Adlhoch et al., 2011, Stefas et al., 1997, Stefas et al., 2001). It also binds with high affinity to LPS on Gram-negative bacteria (Agar et al., 2011) and specific proteins on two Gram-positive pathogens, S. aureus (Zhang et al., 1999) and Streptococcus pyogenes (Nilsson et al., 2008). The broad range specificity of ApoH may provide a viable alternative to current organism-specific techniques (e.g., antibody capture) for the concentration and separation of viral and bacterial pathogens from food or environmental samples.

The purpose of this study was to determine whether ApoH-conjugated magnetic beads could be used to capture norovirus (representative human genotypes and the Tulane virus, a cultivable surrogate) and select Gram-negative (E. coli O157:H7 and S. enterica serovar Enteritidis) and Gram-positive (L. monocytogenes and S. aureus) bacteria from solution. The efficiency of this concentration and separation technique was also determined.

Section snippets

Virus strains

The genogroup I, genotype 1 (GI.1) Norwalk virus (NV) (courtesy of Dr. Christine Moe, Emory University, Atlanta, GA) and a GII.4 New Orleans outbreak strain (courtesy of Dr. Shermalyn Greene, North Carolina State Laboratory of Public Health, Raleigh, NC) were obtained as stool samples and diluted 10–20% in phosphate buffered saline (PBS) (MP Biomedicals, Santa Ana, CA). Tulane virus was propagated in the LLC-MK2 cell line as previously described (Farkas et al., 2008) and recovered after three

ApoH bead-based capture of whole virus particles

A comparison of the RT-qPCRU corresponding to whole virus particles before and after capture using ApoH-coated beads is presented in Fig. 1. In virtually all instances, there was no statistically significant difference between input virus concentration and that recovered using the ApoH beads (p > 0.05), indicating a high degree of capture efficiency. On a percentage basis, > 23% of the input virus was captured by the beads; hence, virus loss during capture was always < 1 Log10 RT-qPCRU. The limit

Discussion

The need to shorten and simplify methods to detect bacterial and viral pathogens in food and environmental samples is well established. The most commonly used approach to achieve this is concentration and purification of the target from the sample matrix just prior to detection. Pathogen concentration methods range from general (e.g. high speed centrifugation, filtration) to specific (immunomagnetic separation), with a emphasis on separating the organisms from sample matrices, removing

Acknowledgements

We thank E. Benedict for her help troubleshooting buffer conditions. This work was supported in part by the Agriculture and Food Research Initiative Competitive Grant no. 2011-68003-30395 from the U.S. Department of Agriculture, National Institute of Food Agriculture through the NoroCORE project. Additional support was provided by the U.S. Air Force Institute of Technology Civilian Institute Program Scholarship. The views expressed in this article are those of the authors and do not reflect the

References (37)

  • C.A. Boulanger et al.

    Precision and accuracy of recovery of Legionella pneumophila from seeded tap water by filtration and centrifugation

    Appl. Environ. Microbiol.

    (1995)
  • B. Bouma et al.

    Adhesion mechanism of human beta(2)-glycoprotein I to phospholipids based on its crystal structure

    EMBO J.

    (1999)
  • Centers for Disease Control & Prevention

    Surveillance for foodborne disease outbreaks–United States, 2009–2010

    MMWR Morb. Mortal. Wkly Rep.

    (2013)
  • H.P. Dwivedi et al.

    Detection of pathogens in foods: the current state-of-the-art and future directions

    Crit. Rev. Microbiol.

    (2011)
  • B.I. Escudero-Abarca et al.

    Selection, characterization and application of nucleic acid aptamers for the capture and detection of human norovirus strains

    PLoS One

    (2014)
  • T. Farkas et al.

    Characterization of a rhesus monkey calicivirus representing a new genus of Caliciviridae

    J. Virol.

    (2008)
  • A.C. Fluit et al.

    Detection of Listeria monocytogenes in cheese with the magnetic immuno-polymerase chain reaction assay

    Appl. Environ. Microbiol.

    (1993)
  • A.C. Fluit et al.

    Rapid detection of salmonellae in poultry with the magnetic immuno-polymerase chain reaction assay

    Appl. Environ. Microbiol.

    (1993)
  • Cited by (9)

    • Glycoproteomics analysis reveals differential site-specific N-glycosylation of donkey milk fat globule membrane protein during lactation

      2023, Food Chemistry
      Citation Excerpt :

      APOH, also known as beta2-glycoprotein I (beta2-GPI), is a single-chain glycoprotein, nearly 19 % of the molecular weight of which consists of glycans, and is involved in metabolic pathways such as lipid metabolism, coagulation, and the production of antiphospholipid antibodies in the body (Lévesque, Jubinville, Hamon, & Jean, 2021). Almand, Goulter, and Jaykus (2016) have shown that APOH can be used as a broad reactive ligand for the concentration and isolation of foodborne pathogens. Javeed et al. (2021) identified 14 α2,3/2,6-linked sialylated N-glycans on APOH using the N-glycome strategy; however, information related to the site specificity of these glycans was missing.

    • A new sight to explore site-specific N-glycosylation in donkey colostrum milk fat globule membrane proteins with glycoproteomics analysis

      2022, Food Research International
      Citation Excerpt :

      Fucosylated and sialylated N-glycans have glycan epitopes similar to those on the surface of target cells, which act as soluble decoy receptors that are recognised and bound by pathogens (Brijesha, Nishimura, & Aparna, 2017; Yu et al., 2018). Within this biological context, APOH has a broad-spectrum affinity for microorganisms, including hepatitis B and C viruses, and many gram-positive and gram-negative bacteria, which can be used as a broadly reactive ligand for the concentration and isolation of foodborne pathogens (Almand, Goulter, & Jaykus, 2016; Levesque et al., 2021). GO enrichment analysis of all identified DCMFGM N-glycoproteins revealed that the majority of N-glycoproteins were located at the membrane, cell surface, lysosome, and extracellular space (Fig. 4), which is consistent with the conclusion that N-glycoproteins are mainly secreted as membrane proteins (Chen et al., 2021).

    • Pathogen enrichment from human whole blood for the diagnosis of bloodstream infection: Prospects and limitations

      2019, Diagnostic Microbiology and Infectious Disease
      Citation Excerpt :

      Examples include lipopolysaccharide for gram-negative bacteria, lipoteichoic acid for gram-positive bacteria (Messner et al., 2013), or chitin for fungi (Ruiz-Herrera and Ortiz-Castellanos, 2010). Ligands such as antibodies (Kim et al., 2017; Vutukuru et al., 2016; Yung et al., 2009), aptamers (Shen et al., 2016), lipoproteins (Almand et al., 2016; Vutukuru et al., 2016), components of the extracellular matrix (Henderson et al., 2011, Singh et al., 2012b), bacteriophages (Singh et al., 2012a), zinc-coordinated bis(dipicolylamine) (Lee et al., 2014), polymyxin B (Raman et al., 2017), or cell wall active antibiotics (Wang et al., 2014) have been immobilized on biomaterial surfaces to capture pathogens and PAMPS from blood samples (Didar et al., 2015; Herrmann et al., 2015; Kang et al., 2014). While ligands with high specificity and affinity for their targets enable efficient binding of specific pathogens, a high degree of specificity narrows the spectrum of pathogens that can be depleted, thereby increasing the probability of false negative results and limiting the suitability of selective pathogen capture for diagnostic purposes.

    View all citing articles on Scopus
    View full text