With a steady increase in meat and poultry consumption globally, confirming the microbial safety of these products is more critical than ever to reduce biological hazards. While various pathogens can be present within meat, Salmonella represents a common presence, and cause of foodborne disease in humans with around 1 million cases and 300 deaths annually in the United States alone.
As food contamination is a frequent source of Salmonella outbreaks, not only from animal sources but also through plant products, reducing the risk across the food supply chain is of key focus.
In this on-demand SelectScience webinar, Erin Crowley, Chief Scientific Officer at Q Laboratories, Inc., and Martin Wiedmann, Gellert Family Professor in Food Safety at Cornell College of Agriculture and Life Sciences, discuss the major public health risks posed in poultry and meat production, where Salmonella serovars Typhimurium and Enteritidis are associated with outbreaks from poultry, eggs, and meat. Hear about the global impact of Salmonella on the meat and poultry supply chain and validation methods for detection of the organism in food.
MW: It has the same pathogenicity as motile Salmonella Typhimurium. Although it seems like there are different events that led to Typhimurium becoming the non-motile strain, it actually happened multiple times, once in Europe and once in the US. It appears to be a phage, a genetic insertion into the H2 gene. Subsequently, some of the Typhimurium strains have acquired multi-drug resistance. For all practical intents and purposes, they are both similar in terms of shell-based variant characteristics. There are no differences in treatment, or likeliness to cause disease in humans.
EC: A good summary of my presentation with the Thermo Scientific RapidFinder Salmonella Multiplex assay is one of the serotype-specific assays. It will tell you if Salmonella species is present and if it is, whether it is Salmonella Typhimurium or Enteritidis. Over the past 10 years, with the exception of the FDA shell egg rule, we have seen that Salmonella Enteritidis detection assays are being used. There is a list of those that are approved in the FDA Bacterial Analytical Manual (BAM). In the National Poultry Improvement Plan (NPIP), there are some serotype-specific assays listed there, most of which are qualitative and use molecular or immunological technologies.
MW: It's been known that some companies have developed serotype-specific assays for serotypes other than Typhimurium or Enteritidis. For example, we had some big issues with the serotype Heidelberg. There is an option to develop specific assays for serotypes other than Enteritidis and Typhimurium.
MW: Typically, it will take two to three days. The real challenge is to get the organism into Phase 2 because that requires some passages to get them to express those. In the US, the classical serotyping process is limited by the fact that you need a lot of different antisera. So commercial labs may offer serotyping, there are some out there, but I don't think there are that many anymore. It is getting rapidly replaced by whole-genome sequencing. Whole-genome sequencing data allows for a very precise and rapid prediction of serovars of Salmonella. We've actually published a paper on that and depending on the platform, it can be extremely fast, less than 24 hours.
EC: I agree with Martin as far as the difficulty with the Kauffmann-White traditional method of serotyping. At Q Laboratories, which is a contract testing laboratory, we often see a request from our clients that want to do a comparative analysis of this serotype-specific strain-typing or whole-genome sequencing option with the Kauffmann-White protocol. That is something that we offer, however, the Phase 2 flagella step is often very difficult. Getting into the individual factors that are needed to confirm the serotype is also challenging because of the practical application. The availability of the resource and limited vendors that offer those, often involving custom-making by order and sometimes that could be a matter of a few weeks to a few months to estimate when they are available. In theory, if we had everything that we needed and be able to predict what factors we would need, it could be anywhere from four to 10 days to get that serotype, but often it could be 2 to 3 weeks if we're working closely with our vendors.
MW: Another important thing for people to realize, in the US the public health agencies or the public health laboratories have stopped doing classical serotyping according to Kauffmann-White, relying only on whole-genome sequence data. In addition, when we see discrepancies with the whole-genome sequence-based prediction in Kauffmann-White serotyping, more than 90% of the Kauffmann-White serotypes cases were incorrect because they're not always easy to read. Sometimes the reagents might not have been quality controlled properly, etc. To be honest, Kauffmann-White will virtually disappear very quickly.
MW: Infectious dosage is a tough concept because it implies that there is one dose that will make a person sick and another one that doesn't. There is no such thing as an infectious dose. There is what is called a dose-response curve, which means the more Salmonella you ingest, the more likely you are to get sick. Even when we ingest really high numbers, there's no guarantee that you will get sick. So, I can't definitively answer that question unfortunately. There is a wide range of differences in terms of virulence, the dose-response curve will differ considerably between isolates.
MW: The most common transmission from poultry to humans will depend usually on the environment. Number one is undercooked poultry products - undercooked chicken, raw chicken, people who consume raw chicken in general. Number two would probably be cross-contamination between raw chicken and raw poultry and other products either in a restaurant or at home. If you store your chicken above your salad and the chicken drops down or drips onto your salad which you then consume, or if you use the same cutting board and utensils for chicken and other products, that is cross-contamination. The third one is what I would call direct zoonotic transmission, and that would be people touching chickens and working with them, etc. An example of other transmission routes that are less indirect is if you use chicken manure or chicken feces to fertilize, for example, produce, that can transmit it from the chicken to the produce, and then the produce can make people sick. There are actually a number of different routes by which this transmission can and does occur, which is one of the reasons why control at the primary source, at the chicken, would be so effective because it will cut all of these different routes of transmission from poultry to humans off.
MW: Many companies that sell you PCR assays will not and cannot tell you what it is, but I can say that the vast majority of them will target invA, invasive gene A, which is one of the universal virulence genes in Salmonella. If you're interested in more details, we actually published a paper recently that shows that it is variation in the assay gene sequences such that some of the most common subtypes or subspecies of serotypes may not be detected as easily, particularly if these PCR parameters are not designed as well as they could be.
EC: Yes, I agree. I also think some of the serotype-specific assays targets the invA gene. It's definitely looking at group D specific parameters and probes as well.
MW: There's a lot of effort to try to classify Salmonella into different virulence types. Conceptually, this is much more difficult than for other pathogens such as enterohemorrhagic E. coli where we have a few key virulence genes. Salmonella has in excess of 10 different genes because of various serotypes having different combinations of Salmonella pathogenicity islands. We are not really close yet to a specific virotype based on this so that we can say: "This one is 10 times more likely to cause disease than another type." It’s definitely a work in progress. You can define which virulence gene the Salmonella has, but it is not very easy to translate that into either relative virulence, or likeliness to causing disease in a human, or even likeliness to causing disease in different types of animals. Part of the challenge is that a lot of the research on Salmonella is really limited to maybe two or three serotypes. There is a lot of research on Typhimurium or Enteritidis, for instance, you have some in Newport and Dublin, but for the other serotypes we don't have enough research, which is why that is a major area of investigation.
MW: To the best of my knowledge there is no consideration yet to add this to MLG, but again I look forward to what Erin has to say. The exception obviously is Salmonella Enteritidis testing for eggs, which is a very specific requirement in the US that eggs ought to be tested for Salmonella Enteritidis and I'm not sure whether there is a specific Enteritidis method in the MLG. Beyond that, the FSI sometimes does encourage companies in the US to do serotype-specific detection. For example, if they have a big issue with Salmonella Heidelberg in a poultry operation.
EC: I agree with Dr. Wiedmann. The USDA does not have official guidance on serotype-specific assays or methodology like FDA BAM. However, in partnership with NPIP, there are certainly serotype-specific assays that are requested within the poultry industry to help with better monitoring and risk assessment. One way that I know that this is going to be changing probably in the next couple of years is that when we're doing method validation, one really critical part of the validation is doing inclusivity and exclusivity evaluation to challenge a variety of strains and serotypes with the assay to see how sensitive its detection is. We have been looking at non-motile and motile Salmonella for a while, and the requirement with the validation guidelines is that you have to look at, at least, 100 different serotypes for the inclusivity method and analyze those strains with the new method enrichment scheme. What we're seeing now at the validation and certification body at the request of the regulatory agency is that we have more of the specific serotype and numerical profiles with the inclusivity panel in order to more accurately understand the public health implication. This is really something that is going to become a lot more of a common expectation than previously before. So that lets me know that we may see some more serotype-specific expectations if not actual methods here in the next couple of years.
MW: Just two quick clarifications. Number one, as Erin pointed out, egg testing would not be in the MLG because it is actually done by the FDA. The second one, back to the question of infectious dose, the WHO uses a number of 1 in 400 of the chance that a single Salmonella could make a person sick. If you ingest one Salmonella in a meal, your chance of getting Salmonella is 1 in 400. Other sources are 1 in 2000, but that at least gives you an idea of how likely Salmonella is to cause infection.
MW: If you want to test a sample for multiple types at the same time, it would really be a PCR assay to detect these multiple serotypes, etc. Anything beyond that will be highly experimental such as metagenomic sequencing out of an enrichment culture. If you're talking about doing subtyping and multiple isolates, whole-genome sequencing (WGS) would perform well. Detection of multiple types from the same sample is still very experimental. Some specific labs might offer you methods that can screen for five or six different methods, but whether you have access to these sort of specialized methods really depends on where in the world you are.
EC: We could divide this up between what we call a screening assay and a confirmatory assay. The screening assay would be something like the pan-Salmonella assay that just detects the general Salmonella species. Some of the serotype-specific species, that are multiplex assays that are available such as the one we were talking about today, can be an option, but getting into the confirmatory element of that where you're really strain typing and then subsequently, serotyping is more of a genomic, meta-genomic option. It is really a combination of the two. If you're looking strictly for automated options, it's the combination of multiple assays that would help you arrive at that result assuming everything goes well.
Learn more about the pathogenicity and transmission of Salmonella: Watch this on-demand webinar here>>
SelectScience runs 3-4 webinars a month across various scientific topics, discover more of our upcoming webinars>>