From the time porcine epidemic diarrhea virus started hitting U.S. swine herds in 2013, veterinarians, researchers and producers have anxiously been trying to find out how to prevent the deadly virus from reestablishing itself. Work also continues to find out how PEDV got here and spreads in the first place.

Scott Dee with Pipestone Applied Research in Minnesota and researchers at the Animal Disease Research and Diagnostic Laboratory at South Dakota State University in Brookings in 2014 looked at the possibility that breeding herds could become infected following ingestion of PEDV-contaminated feed. Findings from that research, concluding that PEDV transmission could occur through contaminated feed, can be found in the Oct. 15, 2014, issue of National Hog Farmer in the article “Proof of Concept: Feed Can Carry PEDV.”

Taking that research a step further, Dee and the research team sought to see if PEDV reacts differently in the individual ingredients within a pig’s feed ration. This research looked at 18 ingredients commonly found in pigs’ diets from these categories — grain sources, porcine byproducts, fat sources and synthetic amino acids.

In this study it was found that soybean meal harbored PEDV for the longest duration post-inoculation at 180 days. Other ingredients harbored the virus for much shorter, but still significant, time. This research also looked at the effectiveness of feed treatments in neutralizing PEDV. More can be read about this research in the August 2015 issue of National Hog Farmer.

Along with the PEDV survivability in soybean meal, the virus was shown in the above-mentioned study to also survive in lysine hydrochloride, choline chloride, DDGS and several porcine byproducts for at least 30 days. Dee says one consistency across several of these ingredients was that they could be imported to the United States from China, and since the original strain of PEDV found in the United States was closely related to a variant found in China, questions arose as to whether the virus could have rode along with the feed ingredients.

So Dee thought the next logical research should be to see if PEDV could survive a simulated “oceanic trip”, providing a possible hint at the origin of the virus. Specifically, this study used a model developed to evaluate the transboundary risk of PEDV-contaminated swine feed ingredients, as well as the effect of two mitigation strategies during a simulated transport event from China to the United States.

In addition to Dee, those involved in the research were Casey Neill and Gordon Spronk, both also with PAR; Aaron Singrey, Travis Clement, Jane Christopher-Hennings and Eric Nelson, all with SDSU ADRDL; Roger Cochrane and Cassandra Jones, Kansas State University Department of Grain Science; and Gilbert Patterson with the Center for Animal Health and Food Safety at the University of Minnesota.

Going to the source

To start the process, select ingredients known to be imported from China such as organic and conventional soybeans and soybean meal, lysine hydrochloride, D-L methionine, tryptophan, Vitamins A, D and E, choline, carriers (rice hulls, corn cobs) and feed grade tetracycline were inoculated with a standard amount of PEDV. In addition, contaminated ingredients were treated with either a liquid antimicrobial (SalCURB, Kemin Industries) or a 2% custom medium chain fatty acid blend (in research by Cochrane and Jones at KSU) were tested for the survivability of PEDV.

To simulate a cargo transport from Beijing, China, to Des Moines, Iowa, the test ran for 37 days, and the samples were stored in an environmental chamber that was programmed to expose the samples to historical temperature and relative humidity levels to mimic the conditions of the land-and-sea transport.

Dee’s team decided to model the test shipping event to occur from Dec. 23, 2012, to Jan. 28, 2013, in order to coincide with the timing of the initial PEDV detection in the United States. To make this study and the impact on the ingredients as real as possible, the research team designed a temperature and percent relative humidity curve for both the land and oceanic segments of the study. Historical meteorological data was accessed for the land segments of the model using Weather Underground encompassing the periods of Dec. 23-30, 2012, and Jan. 17-27, 2013. Actual temperature and percent relative humidity data from within shipping containers traveling from Asia to the United States during this period of time were obtained from a published paper from the Xerox Corp. and were used to model conditions during the oceanic transport period. To simulate the effect of daily fluctuation, the team programmed the environmental chamber to cycle at four designated times each day (6 a.m., noon, 6 p.m. and midnight) and at three designated times each day for relative humidity (8 a.m., noon and 4 p.m.).

Finally, the ingredients were organized into one of four batch samples, each batch representing a specific segment of the trip.

The first batch simulated transport of contaminated ingredients from Beijing manufacturing plants (day 1 post-contamination). Batch 2 (segments 1 and 2) simulated manufacturing in Beijing and delivery to Shanghai, including time in the Anquing terminal awaiting shipment (days 1-8 PC). Batch 3 (segments 1, 2 and 3) represented time in Beijing and Shanghai (as described), the crossing of the Pacific and entry to the United States at the San Francisco terminal (days 1-27 PC). Batch 4 (segments 1-4) represented the all previous events, including transport to and storage in, Des Moines (days 1-37 PC).

On designated days, a batch of samples was removed from the environmental chamber and submitted for testing. Specifically, Batch 1 was submitted on day 1 PC, Batch 2 on day 8 PC, Batch 3 on day 27 PC and Batch 4 on day 37 PC. In other words, the same sample was not repeatedly opened and tested, but rather a new batch of samples was submitted on a designated day. Such sample management ensured that all sample containers remained sealed from the time they were inoculated until the time they were tested at the lab, minimizing the risk of cross-contamination, as well as enhancing repeatability of results as several segments were replicated across batches. To increase statistical power, each of the four batches contained two replicates of each ingredient in the control group and two replicates of each ingredient within each treatment group, for a total of 90 samples per batch.

To limit variability to the level of the ingredient, the same quantity of ingredient, the same container type and the same environmental settings were used and samples were contaminated equally. To initiate this process, 30 grams of each ingredient were added to food storage containers to simulate a shipping container. Ingredients in the non-treated control group were treated with 0.1 milliliter of sterile saline. Ingredients in the liquid antimicrobial group were treated with 0.1 milliliter of product, based on an inclusion rate of 3 kilograms per ton of complete feed. Ingredients in the medium chain fatty acid blend group were treated with 0.6 gram of product based on a 2% inclusion rate. Individual treatments and saline placebo were added to the designated samples using separate syringes. The feed was then properly mixed, stirred manually for 10 clockwise rotations and 10 counter-clockwise rotations using individual wooden applicator sticks per ingredient. Following mixing, each individual container was manually shaken vigorously (50 times in a 10-second period). All samples were then inoculated with 2 milliliters of PEDV (passage 18, Ct = 17.15, total dose 491,520 FFU) and mixed as described. This quantity of PEDV was selected in an effort to provide a final mean Ct value in feed ingredient of approximately 25 (range equaling 19-30) following mixing, based on data from actual field cases of PEDV-contaminated feed, a challenge level used in previous published studies.

For the purpose of negative controls, 30-gram samples of PEDV-negative complete feed were inoculated with sterile saline. Duplicate negative controls were included in each of the four batches, across the control and treatment groups. For the purpose of positive controls, duplicate 5 milliliter samples of stock PEDV in minimum essential media were included in containers within each batch of ingredients in both control and treatment groups.

All prepared samples were then stored in the environmental chamber, which had a programmable temperature range of 4 to 21 degrees Celsius and a relative humidity range of 40 to 95%. To allow for ambient air exposure within the chamber, two 0.318 centimeter diameter holes were drilled into each plastic container.

Feed samples were then tested for PEDV via polymerase chain reaction. To follow-up the PCR test, the residual samples were tested for the presence of viable virus, or virus isolation. Four of the ingredients — conventional and organic soybean meal, lysine and Vitamin D — all tested positive for PEDV by virus isolation.

Samples treated with the liquid antimicrobial or the medium chain fatty acid blend were shown to have no detectable viable PEDV after the “journey.”

Further viability testing

Even though virus is detected in the samples, further testing was needed to confirm whether viable PEDV was present in any feed ingredient sample that had tested positive on PCR but negative on VI. This would be done through swine bioassay, or feeding the samples to piglets. Twenty-four 5-day old piglets in a biosecurity room at the SDSU Animal Resource Wing. The piglets came from a PEDV-naïve herd and were tested upon arrival by blood samples and rectal swabs from each pig. Piglets were divided into pens of four with individual feeding arrangements. Each of the four pigs in each unit would receive the same ingredient. Each pig in the unit received 1 milliliter of the designated inoculum orally via syringe and observed for a seven-day period. To minimize the number of animals needed for the study, pigs that were confirmed negative after day 7 PI would be inoculated with a different ingredient. A negative control unit was included in the design, with these pigs receiving sterile saline placebo. Piglets were monitored daily for clinical signs of PEDV, and ARW personnel collected rectal swabs from each pig.

In the event a pig presented clinical infection signs, swabs of diarrhea and/or vomiting were also collected, and were tested by PCR at the SDSU ADRDL. If PEDV was diagnosed in a specific unit, all animals were swabbed, humanely euthanized with intravenous sodium pentobarbital, the small intestinal tracts submitted for PCR testing, units were cleaned and sanitized, and re-stocked with new piglets as needed.

Samples selected for swine bioassay testing consisted of treated and non-treated ingredients from Batch 4. The non-treated ingredients tested included those which were PCR-positive and VI-negative, specifically Vitamins A and E, tryptophan, D-L methionine, soybeans (organic and conventional) and choline chloride. Viable PEDV was detected in piglets administered non-treated samples of choline chloride. Affected animals displayed evidence of mild diarrhea, shed PEDV in feces and samples of small intestine were PCR and immunohistochemistry-positive at necropsy. All other samples were bioassay negative. In regards to treated ingredients, LA-treated and MCFA-treated samples of soybean meal (conventional and organic), lysine, Vitamin D and choline chloride were tested. All piglets inoculated with the LA-treated or MCFA-treated ingredients were determined to be non-infectious, as piglets remained clinically normal throughout the testing period and all rectal swab and intestinal samples were negative by PCR.

Oceanic survivability

Under the conditions of this study, Dee says this research suggests that PEDV could have been transported via individual feed ingredients and survive an oceanic journey, and this latest research proved repeatability from the previous ingredient research in that choline lysine and soybean meal were positive in both studies. “Coming up positive again in conditions that are much less virus friendly than it is in setting outside in January 10 degrees in Minnesota,” he says. “It shows the virus is even living in tricky robust conditions. … 40-50 degrees and high relative humidity which were all pretty anti-viral when you look at it.”

Dee says this research builds on findings from a USDA study that looked at the totes for feed ingredients in shipping as a potential source of PEDV entry into the United States. “However, I don’t think it’s quite that simple as a bag that brought the virus to our country. For example, in our study, virus stored in containers in the absence of a protective feed ingredient (positive controls) died quickly, in contrast to virus stored in any of the five select ingredients discussed earlier. … This suggests that it’s not the container, it’s the contents within the container.”

An important question raised by the findings of this research, is that is PEDV can survive a trip across the ocean, can the same also be true of other transboundary diseases. Dee says this study also introduces a model which could enhance further transboundary research efforts, possibly employing surrogate viruses (bovine viral diarrhea virus for swine fever virus or Seneca Valley virus for foot-and-mouth disease virus) to test feed-related risks for other foreign animal diseases, further the investigation of the risk of organic ingredients, along with the continuing validation of mitigation strategies.

This research has recently been published in BioMed Central Veterinary Research journal.