Using a next-generation sequencing approach to DNA metabarcoding for identification of adulteration and potential sources of mercury in commercial cat and dog foods

https://doi.org/10.1016/j.scitotenv.2021.146102Get rights and content

Highlights

  • Quantified mercury in pet food, used DNA to test whether actual ingredients match package label.

  • Mercury was present at concentrations above suggested MTLs in 3% of tested foods.

  • Methylmercury concentrations were high, but at safe levels, in fish-based foods.

  • Based on genetic results, adulteration is common in commercial pet foods.

  • No single ingredient is the main source of mercury in commercial pet foods.

Abstract

Studies have demonstrated that some commercial pet (i.e., cat and dog) food products contain high concentrations of mercury (Hg), and some products have Hg concentrations that are higher than expected based on the ingredients included in the package ingredient list. Additionally, concentrations of methylmercury, a particularly toxic form of Hg commonly associated with fish-based ingredients, are largely unstudied despite the widespread use of such ingredients in pet food products. This study aimed to quantify total Hg and methylmercury in a variety of commercial pet food products (n = 127), and use genetic tools to determine if specific ingredients contributed to high Hg concentrations in the final product. Results indicate that total Hg concentrations were above suggested maximum tolerable limits in three of the tested pet food products, and that methylmercury concentrations were at safe levels in all tested products. Next-generation amplicon sequencing using ten barcode primers was conducted to target distinct taxa and to determine if one primer set outperformed the others in amplifying the often heavily degraded DNA found in pet food products. The 16sUniF_16sUniR primer set generated a relatively higher number of reads across the broadest set of taxa, although several of the primer sets were useful in identifying common animal- and plant-based ingredients in commercial pet food products. Combined with the Hg results, it was demonstrated that pet food product ingredients are consistent among and between product lots. However, these results also revealed that adulteration is prevalent in pet food products.

Introduction

Over the past century, mercury (Hg) concentrations in the environment have risen in response to anthropogenic activities. Hg is a ubiquitous element in the environment, and bioaccumulates and biomagnifies in aquatic and terrestrial food webs, resulting in high Hg concentrations in some organisms (e.g., tuna) that are commonly used as ingredients in many commercial pet (i.e., cat and dog) foods. Studies have shown Hg exists in pet foods, including both commercial (Atkins et al., 2011; Adb-Elhakim et al., 2016; Luippold and Gustin, 2016; Kim et al., 2018; Paulelli et al., 2018; Sires et al., 2019) and home-prepared (Pedrinelli et al., 2019) foods. These results are disturbing because Hg is a strong neurotoxin that can exist in many forms, with organic forms (e.g., methylmercury) being the most toxic, and is especially toxic to cats (Takeuchi et al., 1962; Chang and Reuhl, 1983; Opinion of the Scientific Panel on Contaminants in the Food Chain, 2008). Ensuring pet foods have low Hg concentrations can reduce potential health impacts in cats and dogs related to Hg toxicity and poisoning (e.g., ataxia, loss of balance, seizures, death). Unfortunately, this is made more challenging by the fact that there is currently no established regulation for a safe concentration of Hg in pet foods. There are several proposed maximum tolerable limits (MTLs) for cats and dogs, ranging from 67 to 500 parts per billion (ppb) depending on the source, with values of ≤100 ppb reported most frequently. This concern was recently discussed in more detail (Dunham-Cheatham et al., 2019).

Recently, significant associations have been detected between Hg concentrations and specific ingredients used in pet foods. Pedrinelli et al. (2019) demonstrated that high Hg concentrations were significantly correlated with olive oil and pinto beans in home-prepared pet foods. In contrast, Luippold and Gustin (2016) revealed that commercial pet foods dominated by fish-based ingredients, according to the package ingredient list, had the highest Hg concentrations. Fish are well-known to contain high concentrations of Hg, especially large-bodied, long-lived species at higher trophic levels. Typically, the Hg present in these organisms exists as methylmercury, an especially toxic form of Hg. Thus, it is important to determine not only the total Hg concentrations, but also methylmercury concentrations in pet foods. Furthermore, some of the chicken-based dog foods tested by Luippold and Gustin (2016) and Sires et al. (2019) also had high Hg concentrations, although chicken is not commonly considered to be a source of Hg. These results indicate that lesser abundant ingredients may be contributing Hg to the final pet food product, or that Hg-rich ingredients were included in the product but not included in the package ingredient list – a legal issue known as adulteration. For this study, the term “adulteration” is used solely to indicate the occurrence of: 1) the inclusion of an ingredient in the product that is not included in the package ingredient list, and/or 2) the exclusion of an ingredient in the product that is included in the package ingredient list.

Despite occurring as an extremely toxic form of Hg and likely being present in fish-based pet food products, very little research has focused on quantifying methylmercury in pet foods and identifying ingredients responsible for its presence in the product. Fortunately, the development of next-generation sequencing (NGS) technology over the last decade has made food traceability through DNA barcoding techniques possible for even complex and highly processed mixtures such as pet food products. These techniques may allow for the identification of adulteration in pet food products, a growing concern among consumers and the basis of several lawsuits in recent years (Dunham-Cheatham et al., 2019; Palumbo et al., 2020). NGS may also provide a means to validate that pet food products are truly “grain-free”, “corn-free”, or “chicken-free”, for example, claims that are becoming more common in many commercial pet food products.

To build upon recent work investigating the reliability and accuracy of DNA-based approaches in testing for adulteration in products generated by the global supply chain (Xing et al., 2019; Palumbo et al., 2020), we had two major objectives. First, we set out to determine total Hg and methylmercury concentrations in commercially available pet food products, and second, to identify animal and plant-based ingredients present in these products using NGS-based DNA metabarcoding. To accomplish these objectives, the following questions were addressed:

  • 1.

    What are the concentrations of total Hg and methylmercury in commercial cat and dog foods?

  • 2.

    Can the source(s) of animal and plant proteins be identified in pet foods, and do the identified proteins match the ingredient list on the packaging label?

  • 3.

    How consistent is the composition of a commercial cat and dog food? Do Hg concentrations and ingredient proportions vary among lot numbers and among packages from the same lot number?

Section snippets

Sample description and processing

All pet food samples (n = 127) for this project were donated by faculty at the University of Nevada, Reno, acquaintances of the authors, the Reno-Sparks, Nevada community, a local pet food store, and followers of the project's social media account (facebook.com/petfoodmercury). Donations of samples were requested via email, social media, flyers, and personal communications. Only donations that included the original packaging were accepted. Replicates of select foods were purchased to test for

Hg and methylmercury concentrations

Total Hg concentrations varied in all pet food types (Table S3). The 59 dog foods tested ranged from 0.3 to 57.4 μg Hg kg−1, with an average of 8.4 μg Hg kg−1, and the 14 dog treats tested averaged 4.82 μg Hg kg−1. The 17 dry cat foods tested ranged from 1.1 to 142.9 μg Hg kg−1 (average 24.8 μg Hg kg−1), and 35 wet cat foods ranged from 0.4 to 162.6 μg Hg kg−1 (average 16.2 μg Hg kg−1). Some pet foods were more heterogeneous than others, based on the standard deviation of replicate Hg

Hg in pet foods

Overall, these results indicated that the majority of commercial pet foods may have relatively safe Hg concentrations; however, several of the tested foods contained concerning concentrations of total Hg. None of the treats tested contained high Hg concentrations. The majority of the samples with high Hg concentrations were cat foods, especially cat wet foods, and foods that contained fish-based ingredients. Three foods (CD5, CD14, CW20) had total Hg concentrations above suggested MTLs

Conclusion

It is important that all commercially available products be safe and the composition be consistent through time to prevent accidental and unnecessary risk to the consumer. Pet food products are no exception. Although the majority of pet food products had Hg concentrations below suggested MTLs, this study highlights that not all products meet this limit. Without an established and enforced limit for contaminants (e.g., Hg and methylmercury) in pet food products, the pet food industry may

Funding source

Funding for this work came from an internal grant from the College of Agriculture, Biotechnology & Natural Resources at the University of Nevada, Reno, an award from the Office for Undergraduate Research at the University of Nevada, Reno, and private donors.

CRediT authorship contribution statement

Sarrah M. Dunham-Cheatham: Conceptualization, Formal analysis, Investigation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration. Kelly B. Klingler: Conceptualization, Methodology, Formal analysis, Investigation, Writing – original draft, Writing – review & editing, Visualization. Margarita Vargas Estrada: Investigation, Writing – review & editing. Mae Sexauer Gustin: Conceptualization, Writing – review & editing, Funding acquisition.

Declaration of competing interest

The authors declare no conflicts of interest.

Acknowledgements

The authors would like to thank the private donors that supplied samples and financial support to the project. Additional thanks is given to: Chris Fields and Mark Band at the University of Illinois for their guidance and support with DNA sequencing and data processing; Mary Peacock and Michael B. Teglas for their helpful discussions and use of their laboratory facilities for isolating DNA; Jacob Kastner for his graphic design assistance in generating the final versions of the figures; Samantha

References (28)

  • B.J. Callahan et al.

    Exact sequence variants should replace operational taxonomic units in marker-gene data analysis

    The ISME Journal.

    (2017)
  • B.J. Callahan et al.

    High-throughput amplicon sequencing of the full-length 16S rRNA gene with single-nucleotide resolution

    Nucleic Acids Res.

    (2019)
  • S.M. Dunham-Cheatham et al.

    What is in commercial cat and dog food? The case for mercury and ingredient testing. The case for mercury and ingredient testing

    Science of the Total Environment

    (2019)
  • H.T. Kim et al.

    Evaluation of arsenic, cadmium, lead and mercury contamination in over-the-counter available dry dog foods with different animal ingredients (red meat, poultry, and fish)

    Frontiers in Veterinary Science

    (2018)
  • Cited by (5)

    View full text