The safety of dairy products relies on good farming, processing, transport, and storage practices, along with accurate screening for pathogens and drug residues. Continued advancements in dairy safety are now focused on novel techniques that use gene sequencing, metagenomics, and even image analysis and artificial intelligence (AI) to provide early warning signals of potential problems in an industry that produces one of the safest products in the country.
Explore this issueOctober/November 2017
Also by this Author
In 1938, milk-borne outbreaks were responsible for 25 percent of all disease outbreaks attributed to infected foods and contaminated water. By 2015, milk and fluid milk products were associated with less than 1 percent of reported outbreaks, according to the U.S. Public Health Service and FDA Grade “A” Pasteurized Milk Ordinance [PMO] 2015 Revision.
Drug residues in milk have been a concern, and sensitive and accurate analytical methods have been developed to detect and measure the presence of antibiotic residues in dairy products. Under the National Conference on Interstate Milk Shipments (NCIMS) Grade “A” program, state regulatory agencies report milk testing activities to the National Milk Drug Residue Database. In 2012, more than 3.7 million tests were reported to the database, and any milk containing illegal drug residues were not allowed to enter the human food supply.
PMO requires that a milk sample be tested from every bulk tank of raw milk collected at each farm, as well as a sample from every truckload of raw milk arriving at a dairy plant. Samples from every arriving truckload of raw milk are tested for the presence of at least four of six specific beta-lactam drugs: penicillin, ampicillin, amoxicillin, cloxacillin, cephapirin, and ceftiofur. If any test positive, raw milk samples from each farm that supplied the sample for that truckload must be tested.
In addition, the FDA Center for Veterinary Medicine conducted a Milk Drug Residue Sampling Survey, published in 2015, which analyzed raw milk samples from individual dairy farms that had been previously identified as having a drug residue violation in tissues from culled dairy cows at slaughter. These samples were compared to a control group of samples from farms that had not been identified with a previous residue violation. The milk samples were analyzed for antibiotics, non-steroidal anti-inflammatory drugs, and an antihistamine, a total of 31 different drug residues. A positive residue was defined as being at or about 50 percent of the established safe level/tolerance.
Out of the 1,912 total samples, there were 11 confirmed positive milk samples out of 953 (1.15 percent) targeted milk samples, representing 12 confirmed drug residues in the targeted sample group. One sample contained two confirmed drug residues. Among the 959 non-targeted samples, or the control group, there were four confirmed drug residues (0.42 percent). According to the FDA Center for Veterinary Medicine report about this sampling and testing, “the small number of positives in both the targeted and non-targeted groups is encouraging and the FDA continues to be confident in the safety of the U.S. milk supply.”
Additionally, the FDA report called for strengthening the NCIMS drug residue testing program to educate dairy producers on best practices to avoid these residues in both tissue and milk; to utilize the data to, if necessary, include testing for more diverse drug classes in milk; and to consult with state milk regulatory agencies to consider (on a case-by-case basis) collecting milk samples in conjunction with investigating illegal drug residues in tissue involving cull dairy cattle.
Analyzing the Microbiome
Investigators at the University of California, Davis, are taking dairy safety another step forward by identifying the raw-milk microbes, or the level of bacterial diversity that is found in shipments of raw milk that arrive at participating processing facilities in California. The researchers sampled and analyzed milk from 899 tanker trucks on arrival and then shortly after storage at two dairy processors in California’s San Joaquin Valley during the spring, summer, and fall.
Gene sequencing was used to analyze the samples, the same method already being used to study the gut microbiome and soil, according to researcher Maria L. Marco, PhD, an associate professor in the department of food science and technology at UC Davis. “The method has been revolutionary in medicine, agriculture, and many other fields where microbes can be either beneficial or detrimental,” she says.
Using DNA sequencing, Dr. Marco and her team found that the communities of milk microbiomes are highly diverse, with a core microbiota showing distinct seasonal trends. Milk collected in the spring had the most diverse bacterial communities, with the highest total cell numbers and highest proportions of Actinobacteria. A core community of microbes was found in all the raw milk samples, with 29 different bacterial groups and high proportions of Streptococcus and Staphylococcus, as well as Costridiales. The bacterial composition of milk stored in some silos at processing plants was distinct from that in the tanker trucks.
According to Dr. Marco, the research, which was published in a 2016 issue of American Society for Microbiology’s mBio, demonstrated “how the built environment in processing plants can have significant but still unpredictable impacts on the microbial quality of foods.” There are three major ways that this research can impact the dairy industry, she notes. First, it can help identify probable contamination points in processing, such as the pieces of equipment or precise steps where contaminants enter. Second, it can shed light on effective cleaning protocols, such as when and how to clean and how much time should elapse before a piece of equipment should be cleaned again. “Contaminant bacteria can build up over time, so our work is focused on helping processors refine their cleaning procedures,” she says. And third, using DNA sequencing will help increase the ability to predict spoilage. Understanding how to predict which milk would most likely result in a defect in cheese or other dairy product can improve the treatment and handling of milk and thus ensure consistently high-quality products, Dr. Marco emphasizes.
This type of testing is not intended to replace the widely used diagnostic assays that effectively identify pathogens such as Listeria or Salmonella and Campylobacter in milk. Instead, it is an additional approach that can help identify a potential safety risk. “This has shown that we have to be mindful that frequent sampling is needed and that, by using methods we never had before, we can really monitor the equipment to keep the contaminants down,” Dr. Marco says.
Metagenomics and metatranscriptomics have moved food safety, including dairy safety, into the arena of nontargeted screening to give early warning signals of deviations that could indicate a safety issue. The use of genomics also provides a more precise method of detecting, characterizing, and identifying pathogens in foods such as milk, according to Martin Wiedmann, DVM, PhD, Gellert Family Professor in Food Safety at Cornell University. Sequencing DNA and RNA means that the microbiomes can be profiled all along the milk supply chain.
Cornell is collaborating with IBM Research as part of the Consortium for Sequencing the Food Supply Chain (the university is one of several members). The goal of the consortium is to categorize and understand microorganisms and the factors that influence their activity in a normal, safe environment, and to develop the science and the tools that can be used for analysis. Researchers at Cornell are using the university’s own approved and licensed dairy farm and processing facility as a “model system for how we can implement on a routine basis these types of tools,” Dr. Wiedmann says.
Another focus of the program is defining the baseline for “normal” raw milk, and then being able to define “abnormal” milk, he points out. “We have started developing the knowledge to detect some of these abnormalities earlier and trace them back to identifying the cause and, therefore, more effectively and more rapidly address or further characterize the abnormalities.”
One example of how these techniques can be applied has been the identification of certain bacteria that can make refrigerated, pasteurized milk in partially filled containers turn gray or be streaked with gray, as discussed in a July 2017 article in the Journal of Dairy Science. “We found that there are organisms that required oxygen to make the color compound, and we were able to identify which genes are responsible. So if you have these genes and they are expressed and they have enough oxygen, then you are going to get this defect. It was a microbial contaminant that causes the problem, not a disgruntled employee tampering with the product, as had been suspected,” Dr. Wiedmann says.
Other applications include individualized troubleshooting to identify the likely cause of a defect, such as a taste defect, so that intervention can begin to eliminate that cause. Using the tools for genomic sequencing, it is now possible to quickly take a bacterial isolate and accurately identify the microbes involved. “We want to get to the point where individual dairy processors can do this type of testing. As futuristic as it may sound, I think it is feasible that it will probably happen in three to five years in more sophisticated plants,” Dr. Wiedmann notes. Metagenomics testing is likely to supplement traditional dairy testing methods, not replace them, and will allow for more risk-based testing, he says.
U.S. dairy products are “probably some of the safest products around, and the countries where we export pay premium for U.S. dairy products because of the excellent safety record,” Dr. Wiedmann elaborates. “Anything we can do to show that we use cutting-edge tools will hopefully improve that ability for the U.S. to export some of these products.”
A team of researchers at Osaka University and Rakuno Gakuen University, both in Japan, have developed a technique that uses a camera and AI to monitor lameness among dairy cows. Lameness, if untreated, can result in declining quantity and quality of dairy production. The researchers waterproofed and dustproofed a camera-based sensor capable of measuring distance to an object and set it in a cowshed. Based on the large number of cow gait images taken by the sensor, the researchers could characterize cow gaits and detect cows with lameness through machine learning.
Professor Yagi Yasushi at Osaka University says this research “will mark the start of techniques for monitoring cows using AI-powered image analysis. By showing farmers cow conditions in detail through automatic analysis of cow conditions, we can realize a new era of dairy farming in which farms can focus entirely on health management of their cows and delivering high-quality dairy products.”
Raw, Unpasteurized Milk
One dairy product that continues to be associated with disease outbreaks is raw, unpasteurized milk and cheese. The FDA does not regulate the intrastate sale or distribution of raw milk, leaving that up to each state. Thirty-one states allow consumers to purchase raw milk directly, although in many states it can only be purchased at the farm, at farmers’ markets, or through a cow-share program. Twelve states allow its purchase at retail stores. Raw milk cannot be sold across state lines or internationally. In Canada, it is illegal to sell or buy raw milk.
Research published in 2017 of the CDC’s Emerging Infectious Diseases reported that unpasteurized dairy products are responsible for almost all of the 761 illnesses and 22 hospitalizations in the U.S. that occur each year because of dairy-related outbreaks attributed to Shiga toxin-producing Escherichia coli, Salmonella spp., Listeria monocytogenes, and Campylobacter spp. People who consume raw milk are 838.8 times more likely to experience an illness and 45.1 times more likely to be hospitalized than people who consume pasteurized dairy products. The cause of most of these outbreaks is pathogen contamination at the dairy farm, according to the report.
Dr. Marco describes raw milk as a “microbial zoo.” The soil, gut, and aerosol bacteria found in raw milk means that it is a product “that should not be considered probiotic. It has the wrong kind of bacteria, the kind that can make you sick, particularly children and people who are immunocompromised or are recovering from an illness.”
Dr. Wiedmann grew up drinking raw milk as a child, but does so no longer. “I would not let my kids drink raw milk or my friends or pregnant friends, elderly people, or people with weakened immune conditions,” he says. “There are too many risks, and the benefits are anecdotal at best. The risks are very clear, very well described, and ironclad with regards to the science.”