If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.



Anatomy Of A Pet Food Catastrophe

Investigators deployed an array of analytical tools while probing a wave of pet deaths in 2007

by Sophie L. Rovner
May 12, 2008 | A version of this story appeared in Volume 86, Issue 19

Perilous crystals
Credit: Courtesy of William Tolleson/FDA
Pets that ate the tainted food developed kidney stones. Polarized light microscopy (right) highlights the greenish brown kidney stones shown in kidney tissue (left).
Credit: Courtesy of William Tolleson/FDA
Pets that ate the tainted food developed kidney stones. Polarized light microscopy (right) highlights the greenish brown kidney stones shown in kidney tissue (left).

THE PET FOOD contamination crisis of 2007 began quietly, with just a few cats and dogs becoming seriously ill or dying unexpectedly. Regrettably, those first few victims were followed by many more. Ultimately, the incident led to massive recalls of pet food, multiple lawsuits against pet food suppliers that are only now being settled, and an intensive effort in labs throughout North America to track down the cause of the deaths. Even now, research continues into the properties and health effects of the adulterants responsible for the calamity.

As the case unfolded, pet food manufacturers recalled millions of containers of more than 150 brands of dog and cat food. "There definitely was a bit of a panic" among pet owners, "especially when the names of all the foods were still coming out, and people weren't sure if the food they were feeding their pets was going to turn up on the list," recalls Kristy Bennett, an associate veterinarian at Wheaton Animal Hospital in Kensington, Md. While her practice treated only a few of the affected pets, "quite a lot of veterinarians were seeing cases left and right," she says.

Animals that ate the adulterated pet food developed symptoms of kidney failure, including loss of appetite, vomiting, lethargy, frequent urination, increased thirst, and death. Stricken pets also had heightened levels of nitrogen compounds and phosphates in their blood and distinctive round, greenish brown kidney stones clogging their kidneys, says William H. Tolleson, a cell biologist at the Food & Drug Administration.

FDA, which regulates the manufacture and distribution of animal food and drugs, logged more than 18,000 phone calls after the recalls were announced. "This was the largest recall in FDA history," says Tolleson, who works in the agency's National Center for Toxicological Research (NCTR) in Jefferson, Ark. "There were more calls to FDA over this issue than anything we've ever handled before."

As many as 1,000 dogs and cats may have died in North America, according to Colorado State University veterinary pathologist Barbara E. Powers. She bases her estimate on a survey she helped conduct last year as president of the American Association of Veterinary Laboratory Diagnosticians.

Within two months of the first cases, chemists, toxicologists, and other investigators traced the deaths to two specific adulterants in pet food ingredients: melamine and cyanuric acid.

Melamine is used to make plastics, fertilizer, and other products but is not approved as an ingredient in human or animal food in the U.S. The compound "emits a very high level of nitrogen when it is analyzed," according to Paul K. Henderson, president and CEO of Menu Foods, one of the pet food manufacturers that unwittingly used a contaminated ingredient. "The reason this matters is that both the human and pet food industries routinely test for nitrogen as a way of confirming the desired protein levels in their raw materials.

"Why would the Chinese supplier do this? We believe it was to artificially increase the apparent level of the protein," Henderson said last June during his company's annual meeting.

Cyanuric acid is normally used to stabilize chlorine in swimming pools. The compound is a by-product of melamine synthesis and could have been an impurity in the melamine added to the pet food, Tolleson speculates.

Sadly, the 2007 incidents might not have been the first. A 2004 outbreak of kidney failure that reportedly struck thousands of dogs and cats in Asia originally was attributed to contamination of pet food ingredients by fungal toxins. But University of Georgia veterinary pathologist Cathy A. Brown and colleagues reported in September 2007 that they had found "compelling evidence" in tissue samples from autopsies of affected pets that both outbreaks were caused by the same toxic chemical combination (J. Vet. Diagn. Invest. 2007, 19, 525).

The latest drama began unfolding in February 2007, when Canada's Menu Foods began receiving complaints about illness and death in cats that might have eaten the company's pet food. On March 15, Menu Foods notified FDA that at least 10 animals had died, says R. Duane Satzger, director of the organic branch of the agency's Forensic Chemistry Center in Cincinnati.

The next day, Menu Foods announced a recall of some of the dog and cat food it had manufactured between Dec. 3, 2006, and March 6, 2007, because of concerns over kidney failure in pets. The start date for the recall was later expanded back to Nov. 8, 2006. The firm's customers, including Procter & Gamble and Nestlé Purina PetCare, also began initiating their own recalls.

Menu Foods discovered that production of the problematic food coincided with introduction of an ingredient from a new supplier. In late 2006, Menu Foods had added ChemNutra of Las Vegas as a source for wheat gluten, a concentrated vegetable protein used as a thickener or binding agent in pet food. Other pet food manufacturers also purchased small quantities of wheat gluten from ChemNutra and ultimately initiated recalls as well. ChemNutra itself had acquired the wheat gluten from a Chinese supplier.

AT THE TIME of the initial recall last year, the specific defect in the wheat gluten was a mystery, despite a battery of tests. Assays for minerals, heavy metals, ethylene glycol, pesticides, toxins formed by molds, and intentionally added toxins came up empty.

Then on March 23, researchers at the New York State Food Laboratory said they had identified the rat poison aminopterin as a toxin in the pet food (C&EN, April 2, 2007, page 11). FDA was unable to corroborate the finding. But Daniel H. Rice, director of the New York lab, stands by the aminopterin identification. "If lab findings don't support each other, it's usually due to different methodology," he says. "In this case, FDA instrumentation was different than ours." Rice notes his lab detected the aminopterin only at parts-per-billion levels, so it probably wasn't a factor in sickening pets anyway.

On March 30, FDA and Cornell University—whose animal health experts had been enlisted in the investigation by Menu Foods—announced that they had independently identified melamine as a contaminant.

Cornell researchers had detected the melamine in urine and kidney samples from affected pets, and both Cornell and FDA's Forensic Chemistry Center (FCC) had detected the contaminant in the pet food.

FCC made an initial tentative identification of the melamine using a mass spectrometric (MS) technique that relies on the open-air ionization method known as direct analysis in real time (DART) (C&EN, Oct. 8, 2007, page 13). The group confirmed the finding with additional DART and gas chromatography (GC)-MS runs.

The finding was a bit mysterious because the small amount of information available in the scientific literature indicated that melamine wasn't particularly toxic.

FCC staff next examined samples of the suspect wheat protein concentrate under a light microscope and separated out what appeared to be foreign particles. Fourier transform infrared and Raman spectra of the particles, as well as DART and GC-MS, confirmed the presence of melamine. FCC's analysis also unearthed further triazine contaminants, particularly cyanuric acid. Satzger says that total adulterant levels in the protein concentrate ran as high as 25% by weight.

The DART technique is much quicker than GC-MS because it requires no sample preparation, Satzger notes, but it depends on expensive equipment that's not widely available. Throughout April 2007, FCC scientists labored to develop a GC-MS method to rapidly screen pet food samples—one that could be replicated in other labs across the country with commonly available instrumentation. One challenge involved finding a way to test for multiple triazines, which have widely varying solubilities, in a single run. FCC researchers working in collaboration with FDA's Pacific Regional Laboratory Northwest, in Bothell, Wash., found that a solvent system of acetonitrile, water, and diethylamine did the trick.

During this same period, other FCC scientists were examining crystals from kidney tissue provided by FDA's Rockville, Md.-based Center for Veterinary Medicine, Satzger says. Raman mapping studies indicated the crystals could be melamine cyanurate. Lab tests also showed that mixing melamine and cyanuric acid in water forms melamine cyanurate crystals.

Meanwhile, Tolleson was gathering data about the toxicity of the pet food contaminants using cultured cells from the immune system known as macrophages. He found that cyanuric acid by itself was only slightly toxic; by contrast, melamine and the combination of both compounds were toxic to the macrophages.

In April 2007, Procter & Gamble toxicologist George P. Daston showed that the triazine compounds are deadly to rats when consumed in combination, Tolleson says. Rats that ate the mixture developed the same kidney disorder seen in pet cats and dogs. Neither Daston nor other P&G representatives would comment for this story because the company is involved in ongoing litigation associated with the contaminated pet food.

Other researchers at the University of Guelph, in Ontario, "mixed melamine, cyanuric acid, and cat urine and reproduced the kidney stones that are quite characteristic for this melamine-associated renal failure syndrome," Tolleson says. "Using Fourier transform infrared spectrometry, they were able to get a very clear signature that melamine cyanurate was present in those crystals in affected cats."

In November 2007, University of California, Davis, veterinary toxicologist Birgit Puschner and colleagues reported on the results of feeding melamine and cyanuric acid individually and in combination to a few cats (J. Vet. Diagn. Invest. 2007, 19, 616). The researchers found that a single 32 mg/kg dose of the combination can cause acute renal failure in cats. Puschner suggested that the compounds cause renal damage at least in part through precipitation of melamine cyanurate crystals out of acidic urine in the kidneys.

"That pretty much brought the story full circle," Tolleson says. "Melamine cyanurate was present in the contaminated raw material, it was detected in the crystals that were responsible for kidney failure, and feeding animals with melamine cyanurate reproduced the kidney toxicity."

As Puschner suggested, the behavior of melamine and cyanuric acid depends on pH. Tolleson determined that different pH values in different zones in the body influence the solubility of melamine and cyanuric acid, "just as you would predict from the acid-base dissociation constants for those compounds." The compounds are most soluble in the acidic conditions of the stomach and least soluble near pH 5, which is found in parts of the kidney, he says.

Under neutral conditions, the compounds form a stable and insoluble hydrogen-bonded network; the network is destabilized at pH extremes. Accordingly, melamine cyanurate crystals could be expected to form in blood, which has a neutral pH. But they don't. "The hypothesis is that serum proteins might sequester one or both of the triazines and allow them to remain in circulation until they can be deposited in the kidney, where they precipitate," Tolleson says. He is now searching for triazine-binding proteins in the blood and urine and is studying proteins found together with melamine cyanurate in crystals.

Shortly after the wheat gluten story broke, another problematic pet food ingredient came to light. Wilbur-Ellis, a San Francisco-based distributor of agricultural products, discovered that rice protein concentrate from its Chinese supplier was contaminated with melamine and related compounds.

FDA found other problems when it analyzed the rice protein, Satzger says. Microscopy suggested the product contained wheat starch, casting doubt on its basic composition. FCC's subsequent chromatographic analysis of both the rice and wheat gluten products showed they were actually ground wheat mixed with the triazine adulterants. Satzger notes this mixture would have been cheaper to produce than the genuine pet food ingredients.

DOGS AND CATS were not the only animals that consumed pet food containing the tainted ingredients. Before realizing the pet food was contaminated, manufacturers had incorporated it into feed given to hogs, chickens, and fish. Recalls were announced for the feed, and thousands of animals that hadn't yet been consumed were temporarily quarantined. About a month after the feed problem first became known, FDA announced that a U.S. unit of Canada's Tembec had been using melamine as an unapproved binding agent in feed ingredients for livestock, fish, and shrimp for years. The feed ingredients were then recalled.


FDA's Center for Veterinary Medicine subsequently evaluated the toxicity of melamine cyanurate in hogs, chickens, and fish, Tolleson says. And the agency's Center for Food Safety & Applied Nutrition assessed the risk for humans who consumed fish or pork derived from animals that had eaten the contaminated feed. "It turned out that it wouldn't be possible to consume enough meat from those animals to produce a toxic response in humans," Tolleson says.

FDA's NCTR plans to perform pharmacokinetic and toxicological studies of melamine cyanurate in rats and miniature pigs, whose renal physiology "is closer to that of a human than most typical lab species," Tolleson says.

While the technical studies continue, the legal and financial mess spawned by the pet food adulteration is also gradually unfolding. This February, a federal grand jury in the U.S. indicted ChemNutra and its owners; Xuzhou Anying Biologic Technology Development Co. and its Chinese owner; and the export broker Suzhou Textiles, Silk, Light Industrial Products, Arts & Crafts (SSC) and its Chinese president for their roles in the contamination scheme. SSC had obtained the contaminated wheat from Xuzhou Anying on behalf of ChemNutra. According to the indictment, the shipments were falsely labeled to avoid inspection in China. ChemNutra's owners reportedly deny the allegations involving their company.

Menu Foods' Henderson says it's unlikely that customers of the Chinese companies will be able to obtain restitution because they have been shut down (C&EN, July 30, 2007, page 31). However, his firm has sued ChemNutra. Menu Foods itself faces considerable liability. The recall cost the company $54 million. Furthermore, on April 1, 2008, Menu Foods announced it had reached a tentative mediated settlement with pet owners who had filed claims against the company. A court hearing is scheduled for May 14.

Work continues to make the pet food and livestock feed supply safer in the U.S. On May 13, FDA will host a meeting to discuss the development of ingredient, processing, and labeling standards for pet food. And the agency will hold a meeting on May 14 to present changes made to its Animal Feed Safety System project since last year. Changes include efforts to rank feed hazards according to the risks they pose to animal and public health and a risk-based method for prioritizing inspections of feed-production venues.

As a step toward preventing replays of this pet food adulteration saga, the U.S. State Department has approved an FDA plan to place staff in China (C&EN, March 24, page 28). The Chinese government must also approve the plan, which is designed to improve the safety of products shipped to the U.S. FDA also hopes to establish a collaborative relationship with authorities in India.

Cover Story


This article has been sent to the following recipient:

Chemistry matters. Join us to get the news you need.