The recalls involving powdered milk continue to pile up.
Since December, more than a dozen products containing powdered milk have been recalled due to the risk of Salmonella, including mini eclairs and cream puffs, mac & cheese products, chocolate-covered pretzels, potato chips, seasonings and white peppermint Hostess Twinkies.
“FDA investigators observed residues on internal parts of the processing equipment after it had been cleaned by the company and water dripping from the ceiling onto food manufacturing equipment. In addition, environmental swabs collected during the inspection confirmed the presence of Salmonella meleagridis on surfaces food came into contact with after being pasteurized.” – FDA news release
Anyone who has attended a food safety conference in the last few years has experienced some type of whole genome sequencing (WGS) presentation. WGS is the next big thing for food safety. The technology has been adopted by regulatory agencies, academics, and some food companies. A lot has been said, but there are still some questions regarding the implementation and ramifications of WGS in the food processing environment.
There are a few key acronyms to understand the aspects of genomics in food safety (See Table I below).
Pulse Field Gel Electrophoresis
Technique using restriction enzymes and DNA fragment separation via an electronic field for creation of a bacterial isolate DNA fingerprint; PFGE is being replaced by WGS at CDC and other public health laboratories
Whole Genome Sequencing
The general term used for sequencing—a misnomer—the entirety of the genome is not used, and depends on the analytical methodology implemented
Next Generation Sequencing
NGS is the next set of technology to do WGS and other genomic applications
Single Nucleotide Polymorphisms
A variation in a single nucleotide that occurs in specific position of an organism’s genome; Used in WGS as a methodology for determining genetic sameness between organisms
Multilocus sequence typing
A methodology for determining genetic sameness between organisms; Compares internal fragment DNA sequences from multiple housekeeping genes
16s RNA sequencing
A highly conserved region of the bacterial genome used for species and strain identification
Joseph Heinzelmann will be presenting: Listeria Testing Platforms: Old School Technology vs New Innovative Technology during the 2016 Food Safety Consortium | LEARN MOREIn 1996, the CDC established the PulseNet program for investigating potential foodborne illness outbreaks. PulseNet has relied on using bacterial DNA fingerprints generated via PFGE as comparisons for mapping potential sources and spread of the outbreaks. Due to a number of advantages over PFGE, WGS is quickly becoming the preferred method for organism identification and comparison. Moving to WGS has two critical improvements over PFGE: accuracy and relatedness interpretation. Like PFGE there are nuances when defining the difference between two very closely related organisms. However, instead of defining restriction enzymes and comparing the number of bands, the language changes to either single nucleotide polymorphisms (SNP) or the number of alleles. The other important aspect WGS improves is the ability to determine and interpret the relatedness of organisms more broadly. The frequent Listeria outbreaks and incidence from 1983-2015 provide an insight to what the future might hold with WGS implementation.1 The incidence report shows the increased ability to quickly and more accurately define relatedness between clinical cases creates a link of potential cases much faster.
WGS also provides key practical changes for outbreaks and recalls in the food industry. Sequencing provides a much faster response time and therefore means the outbreaks of foodborne illness decrease, as does the number of cases in each outbreak. As the resolution of the outbreaks increases, the number of outbreaks identified increases. The actual number of outbreaks has likely not increased, but the reported number of outbreaks will increase due increased resolution of the analytical method.
WGS continues to establish itself as the go-to technology for the food safety agencies. For example, the USDA food safety inspection service recently published the FY2017–2021 goals. The first bullet point under modernizing inspection systems, policies and the use of scientific approaches is the implementation of in-field screening and whole genome sequencing for outbreak expediency.
Agencies and Adoption
The success of FDA and CDC Listeria project provides a foundation for implementation of WGS for outbreak investigations. The three agencies adopting WGS for outbreak investigations and as replacement for PulseNet are the CDC, FDA and USDA. However, there are still questions on the part of the FDA for when WGS is utilized, including under what circumstances and instances the data will be used.
In recent public forums, the FDA has acknowledged that there are situations when a recall would be a potential solution based on WGS results in the absence of any clinical cases.2 One critical question that still exists in spite of the public presentations and published articles is a clear definitions of when WGS surveillance data will be used for recall purposes, and what type of supporting documentation a facility would need to provide to prove that it had adequate controls in place.
A key element is the definition between agencies for sameness or genetic distance. The FDA and FSIS are using a SNP approach. A sequence is generated from a bacterial isolate, then compared with a known clinical case, or a suspected strain, and the number of different SNPs determines if the strains are identical. The CDC is using the Multilocus sequence typing (MLST) approach.
Simple sequence comparisons are unfortunately not alone sufficient for sameness determination, as various metabolic, taxa specific and environmental parameters must also be considered. Stressful environments and growth rates have significant impact on how quickly SNPs can occur. The three primary pathogens being examined by WGS have very different genetic makeups. Listeria monocytogenes has a relatively conserved genomic taxa, typically associated with cooler environments, and is gram positive. Listeria monocytogenes has a doubling time of 45–60 minutes under enrichment conditions.3 These are contrasted with E. coli O157:H7, a gram negative bacteria, associated with higher growth rates and higher horizontal gene transfer mechanisms. For example, in an examination of E. coli O104, and in research conducted by the University in Madurai, it showed 38 horizontal gene elements.4
These two contrasting examples demonstrate the complexity of the genetic distance question. It demonstrates a need for specific definitions for sameness within a microbiological taxa, and with potential qualifiers based on the environment and potential genetic event triggers. The definitions around SNPs and alleles that define how closely related a Listeriamonocytogenes in a cold facility should be vastly different from an E. coli from a warm environment, under more suitable growth conditions. Another element of interest, but largely unexplored is convergent evolution. In a given environment, with similar conditions, what is the probability of two different organisms converging on a nearly identical genome, and how long would it take?
MLST vs. SNP
As previously stated, the three agencies have chosen different approaches for the analytical methodology: MLST for CDC and SNP of the FDA and USDA. For clarity, both analytical approaches have demonstrated superiority over the incumbent PFGE mythology. MLST does rely on an existing database for allele comparison. A SNP based approach is supported by a database, but is often used in defining genetic distance specifically between two isolates. Both approaches can help build phylogenetic trees.
There are tradeoffs with both approaches. There is a higher requirement for processing and bioinformatics capabilities when using a SNP based approach. However, the resolution between organisms and large groups of organisms is meaningful using SNP comparison. The key take away is MLST uses a gene-to-gene comparison, and the SNP approach is gene agnostic. As mentioned in Table 1, both approaches do not use every A, T, C, and G in the analytical comparisons. Whole genome sequencing in this context is a misnomer, because not every gene is used in either analysis.
Utilizing WGS for companies as a preventive measure is still being developed. GenomeTrakr has been established as the data repository for sequenced isolates from the FDA, USDA, CDC and public health labs. The data is housed at the National Center for Biotechnology Information (NCBI). The database contains more than 71,000 isolates and has been used in surveillance and outbreak investigations. There is a current gap between on premise bioinformatics and using GenomeTrakr.
The FDA has stated there are examples where isolates found in a processing facility would help support a recall in the absence of epidemiological evidence, and companies are waiting on clarification before adopting GenomeTrakr as a routine analysis tool. However, services like NeoSeek, a genomic test service by Neogen Corp. are an alternative to public gene databases like GenomeTrakr. In addition to trouble shooting events with WGS, NeoSeek provides services such as spoilage microorganism ID and source tracking, pathogen point source tracking. Using next generation sequencing, a private database, and applications such as 16s metagenomic analysis, phylogenetic tree generation, and identification programs with NeoSeek, companies can answer critical food safety and food quality questions.
Today food companies will have access to a new whole genome sequencing (WGS) test that could help them prevent dangerous pathogens from getting into their products. Released by Clear Labs, the test takes a detailed approach to identifying pathogen strains in samples, providing information about their geography and from which food groups they originate.
In an exclusive interview with Food Safety Tech, Mahni Ghorashi, co-founder of Clear Labs, explains how he expects the company’s new test, which has a five- to seven-day turnaround time, will offer companies with a more accurate yet less expensive alternative to protecting consumers by actively monitoring their supply chain for emerging pathogens.
Food Safety Tech: What differentiates this WGS test from current available solutions?
Mahni Ghorashi: No one has been able to provide the food industry with modern whole genome sequencing techniques for food safety. What we’re releasing is a quantum leap in terms of what’s been available on the market today. Whole genome sequencing has been largely siloed to regulatory bodies like FDA and CDC to trace outbreaks and inform investigations—the technologies and techniques that they’re using are actually fairly old; they’re some of the original WGS techniques that emerged on the next-gen sequencing platform. We’ve taken the most advanced techniques on the NGS platform for human disease exploration and personalized medicine and adapted them for food industry.
What gives our WGS test a competitive advantage over legacy-based methods is two fold:
1. Clear Labs has a 2-million+ entry-curated database of genomic information and sequences for the accurate ID of food ingredients (pathogenic organisms and microbiomes). Its accuracy and the confidence level that comes behind our matching is a huge step above anything that’s available in the public domain today.
2. Being able to place pathogenic strain information in the context of overall food ingredients and samples. The whole genome sequencing test we developed has been specifically catered for the food industry, and for food samples in particular, [versus] FDA’s GenomeTrakr, CDC’s PulseNet, or other food safety labs that are offering full genomic sequencing of pathogen strains—they’re using some of the earliest methods to do this. On the NGS platform, we’re able to put those strains in the context of food ingredients and suppliers: Specifically, [matching] bacterial strains with food ingredients [and] suppliers.
FST: Does this test target specific foods?
Ghorashi: Our platform particularly shines in complex foods. The value of next-gen sequencing and DNA barcoding over PCR-based technologies, which is the gold standard in food safety, is its stability to break down complex food ingredients into all of their known parts, and to look in a universal and unbiased way into food samples. It’s untargeted, so you don’t have know what it is that you’re looking for—and that’s the real power.
FST: What impact do you anticipate for this test, especially in the context of FSMA?
Ghorashi: Our customers are using [the test] for monitoring ingredient supplies and the effectiveness of preventive and sanitary controls [and] to match specific pathogen strains to specific food ingredients. They are using it for proactive testing for FSMA compliance—there’s a lot of movement in this direction and hefty budgets are being allocated to put new preventive controls in place in response to FSMA; whole-genome sequencing will play a big role, and we anticipate large-scale partnerships with agencies and private industry on that front. And the most obvious use case is that it’s being used for techniques to mitigate or reduce the risk of product recall and outbreak.
We’ve been able to significantly reduce the price point on whole-genome sequencing, and all of our tests across the board, because we’re intimately familiar with how the inner workings of these platforms and how to best optimize them for scale and cost efficiency. We think the test will be more accurate and leaps and bounds ahead of what’s available, as well as cost competitive. We’re excited about the work we’re doing and its impact on food safety. I don’t think the food industry—retailers and manufacturers—have ever had access to these kind of tools and they’re being made available just in time for FSMA, as the industry moves towards a more proactive approach to food safety and [takes] preventive measures in their supply chains. Hopefully we’ll soon be living in a world where outbreaks, illness and the financial toll are a thing of a past.
Clear Labs also just released a microbiome test that helps companies associate microbiomes with specific food ingredients.
Mahni Ghorashi: The microbiome test we’ve developed is able to sequence samples from the human gut and from food, and look at how the microorganisms are interacting. Our customers for this test have been large brands that have advanced R&D departments and academic research centers that are looking for how diet research and the microbiome interact together and how new product development can help us move toward personalized diets when it comes to prebiotic and probiotic diets.”
The impact of the microbiome and the correlations between bacteria of the human gut and the bacteria in the food we eat. The prevailing thesis at the moment is that the microbiome has a significant impact on our health when it comes to disease risk and diet, inflammation and mood disorders. We’re seeing very forward thinking brands like Nestle, ConAgra and Mars putting a lot of attention on the impact of the microbiome when it comes the development of new products, [such as] prebiotics and probiotics, or even specific food products as it pertains to the microbiome. We believe that this intersection— nutrigenomics and the personalized diet—is going to be a massive market, and we’re at the early stages of that.
Americans consume 350 billion pounds of food each year, with one out of six people falling victim to foodborne illness, and 3000 dying. The significant amount of Listeria outbreaks hitting the industry (most recently, the staggering number occurring in produce) has left many food safety and quality assurance professionals searching for better methods of prevention and detection. Using big data, specifically metagenomics, to improve food safety and detect potentially deadly outbreaks is indeed where the future is headed.
DID YOU KNOW? The estimated U.S. cost of one case of Listeriosis is $1.4 million. Listeria is a prime concern due to the high percentage of fatalities that occur as a result of contracting Listeriosis. And what’s worse is the fact that many of the cases are preventable.
During Food Safety Tech’s Listeria Detection & Control Workshop this week, John Besser, Ph.D., deputy chief of CDC’s Enteric Diseases Laboratory Branch, outlined how the agency is leveraging metagenomics to find unrecognized problems in the food supply. Perhaps the most important element of disease surveillance is that it enables the detection of new issues, especially those whose presence was previously unknown.
Pathogen-specific surveillance allows the detection of more outbreaks, which will in turn make the food supply safer, because it will enable industry to understand the root causes of outbreaks and help them address problems much sooner. The CDC is focused on genome-based outbreak detection because of its ability to achieve faster detection—and with greater precision in identifying the source. The method has also helped the agency solve outbreaks with fewer cases occurring, and it concurrently helps rule out sources.
PulseNet, a nationwide database (comprised of 87 labs in the United States) that links cases most likely to share a cause for illness, has prompted food safety improvements across a variety of products, including sprouts, peanut products, leafy greens, flour, melons, eggs and poultry. Combine this capability with the Listeria initiative, which was launched in the mid-2000s, and the CDC has been able to find more (and smaller) outbreaks than ever before. In fact, there’s been a dramatic increase in the number of outbreak cases that have been solved (with the food source being identified). During the pre-whole genome sequencing (WGS) stage (September 2012–August 2013), only one outbreak was solved; in year one of the WGS project (September 2013–August 2014), four cases were solved; in year 2 of the WGS project (September 2014–August 2015), nine outbreaks were solved. In these respective time periods, the median number of cases per cluster dropped from six to four to three. In addition, the number of cases linked to a food source jumped from 6 to 16 to 93 during this respective time period.
Besser also discussed the role of metagenomics, or the study of total genetic material recovered directly from environmental samples. A couple of years ago, this was science fiction and wasn’t possible, he said. But as we look to the future, metagenomics will become a lot cheaper as computers become more powerful—and at break-neck speed. He referenced IBM Research, who earlier this year announced a project being conducted in conjunction with Mars, Inc. and Biorad for sequencing the food supply chain (calling it the “largest-ever metagenomics study”).
Metagenomics enables the profiling of communities of microbiomes anywhere in the food supply chain. And the method is fast—it can potentially shave weeks off the process of identifying clusters of interest. In addition, it can increase the value of interviews conducted with patients who have fallen ill (Think about it: Do you remember what you ate two weeks ago? What about a month ago?).
Currently there are several limiting factors surrounding metagenomics: Cost; sequencing read length and error rate; specific software (and pipelines); computing processing power and bandwidth; and the signal-to-noise factor. However, with the rapid rate in which technology has been improving in this space, the high likelihood of these issues being addressed and resolved in the not-so-distant future will present exciting opportunities in outbreak prevention and detection.
“When a flower doesn’t bloom you fix the environment in which it grows, not the flower.” A quote, by Alexander Den Heijer, trainer, speaker, purposologist, that rings true in food safety. When there is a contamination issue in food processing, one must fix the environment in which food is being processed. Safe food is a product of a clean environment.
We have better environmental sampling programs in our food manufacturing plants and processing facilities, and we have sanitation standard operating procedures, so why are we seeing a prevalence of Listeria, and in rising numbers? I recently sat down with Jeff Mitchell, vice president of food safety at Chemstar, about the recent increase in Listeria outbreaks and how you can rid your facility of the dangerous pathogen.
We’re seeing Listeria—in product recalls and outbreaks—over the last couple of years, and in multiple numbers. Why do you think this is happening?
Jeff Mitchell: The distribution of Listeria in the environment has not changed, and the processes that we use for processing food really haven’t changed. What’s changed is the way that we collect data. We have PulseNet now, which gathers information. If someone goes to a medical treatment facility with a foodborne illness, they’re going to investigate that and they’re going to get the whole genome sequencing on the pathogen.
There’s a difference between understanding what transient Listeria is and resident Listeria. I think there are a lot of sanitation efforts being put forth to eliminate the resident populations—those are the populations we’re most concerned about, and they’re the ones that are being related back to a lot of these recalls.
If I have resident Listeria in my facility, why can’t I find it?
Food Safety Tech is organizing a Listeria Detection & Control Workshop, May 31 – June 1, 2016 in St. Paul, MN. LEARN MOREMitchell: Resident populations of Listeria are found in a biofilm—most bacteria aggregate within a biofilm. A biofilm is a survival mode for the bacteria; it protects it from sanitizer penetration. That layer actually masks it from sampling. You could swab a surface or an area and not pick it up, because the biofilm is masking it.
Jeff goes on to discuss the type of sanitation program that companies should have in place to get rid of resident Listeria. You can learn about the steps you need to take in my video interview.
The CDC has declared the Chipotle E. coli outbreaks over. As for its origin(s), we may never know. Yesterday the CDC provided its latest and final update regarding the two outbreaks, stating that investigators used whole genome sequencing to dig a bit deeper, and isolates tested from those sickened in the second outbreak (sickened five people in three states) were not genetically related to isolates from the people who fell ill in the initial outbreak (55 sickened in 11 states, with 21 hospitalizations).
“We are pleased to have this behind us and can place our full energies to implementing our enhanced food safety plan that will establish Chipotle as an industry leader in food safety,” said Steve Ells, founder, chairman and co-CEO of Chipotle in a company statement. “We are extremely focused on executing this program, which designs layers of redundancy and enhanced safety measures to reduce the food safety risk to a level as near to zero as is possible. By adding these programs to an already strong and proven food culture, we strongly believe that we can establish Chipotle as a leader in food safety just as we have become a leader in our quest for the very best ingredients we can find.”
While the outbreaks “appear” to be over, the fact that the source will remain a mystery is a bit unsettling. All the CDC can tell us is that the “likely” source was a common meal item or ingredient served at Chipotle Mexican Grill. Regulatory officials simply cannot trace a food or ingredient to the outbreak. “When a restaurant serves foods with several ingredients that are mixed or cooked together and then used in multiple menu items, it can be more difficult for epidemiologic studies to identity the specific ingredient that is contaminated,” according to the CDC’s final update on the outbreak.
The most recent reported illness started on December 1, 2015. No deaths were reported as a result of either of the outbreaks.
Today Chipotle released its Q4 2015 earnings, reporting a 6.8% decrease in revenue ($997.5 million) compared to Q4 2014. However, 2015 revenue increased 9.6% over 2014.
The problems are not over for the restaurant chain either. On January 28, Chipotle was served another subpoena that broadened the scope of the existing DOJ investigation. The company stated the following in a release, “The new subpoena requires us to produce documents and information related to company-wide food safety matters dating back to January 1, 2013, and supersedes the subpoena served in December 2015 that was limited to a single Chipotle restaurant in Simi Valley, California. We intend to fully cooperate in the investigation.”