Better Safeguards For The Food Supply

DEADLY ADULTERATION of milk with melamine has impelled scientists to develop new analytical methods to detect the compound in food. Yet melamine is only the latest in a long line of contaminants that have been added to food through the ages.

"Melamine is the compound of the hour," says R. Graham Cooks, an analytical chemist at Purdue University. But "there have been and will be many more 'alarm compounds' that people will need to detect in foods." So scientists are looking beyond melamine-detection methods to other techniques that can more broadly protect the food supply.

Some of those techniques are designed to overcome an analytical weakness exploited in the melamine incident. Dairy producers in China added the compound to their milk to inflate its apparent protein content. The higher the protein content appeared to be, the more these suppliers were paid for their milk. The stratagem worked because protein content is assessed indirectly by measuring a food's nitrogen content, and melamine is nitrogen-rich.

The two principal techniques for measuring nitrogen content are the Kjeldahl method and the combustion method.

An analyst begins the Kjeldahl procedure by boiling a sample with concentrated sulfuric acid, thereby converting any nitrogen in the sample into ammonium sulfate. Adding sodium hydroxide transforms the ammonium ions into ammonia. Mixing this gas with acid neutralizes some of the acid. The amount of the remaining acid can be determined by titration. With some calculation, that quantity reveals how much nitrogen—and, in theory, how much protein—was in the original sample.

In the combustion method, which is also known as the Dumas method, the sample is burned to release nitrogen gas. The amount of nitrogen is determined with a thermal conductivity detector. A simple calculation indicates the amount of nitrogen and, presumably, protein in the sample.

The 2008 milk crisis, as well as a 2007 scandal involving melamine contamination of pet food, prompted several organizations to look for ways to improve the measurement of protein content in food and to make such analyses less vulnerable to deceptive tampering. The U.S. Pharmacopeia (USP), a nonprofit public health organization that sets standards for the quality and purity of food and medicines, is playing a leading role in the effort.

Jeffrey Moore, scientific liaison for the organization's "Food Chemicals Codex" (FCC), a compendium of standards for food ingredients, is preparing an extensive review of existing and emerging protein-assessment methods. And USP will host a June 16–17 workshop at its Rockville, Md., headquarters to develop what it calls "a toolbox of analytical solutions to address food adulteration, with an emphasis on protein measurement."

"There's not a one size fits all kind of solution," Moore says. "There really needs to be a toolbox approach dependent on what the food matrix is and what the measurement will be used for." For instance, total protein content is relevant in evaluating the overall quality of milk. But for cheese production, manufacturers may want to know how much of a specific protein, such as casein, is present.

One potential tactic is to measure protein at a less fundamental scale than its elemental constituents, says Markus Lipp, director of food standards for USP. For example, enzymes can be used to break proteins into individual amino acids that can then be identified by high-performance liquid chromatography.

"Then there are biochemical methods, such as enzyme-linked immunosorbent assays, that look for bigger chunks of the protein molecule, such as a characteristic sequence of about seven to 10 amino acids," Lipp says. Spectroscopic methods can detect an electromagnetic signature of the whole protein. Chromatographic methods such as capillary electrophoresis or HPLC, sometimes used in combination with MS, can also identify the whole protein. Colorimetric methods, too, can be used for protein detection, Moore says.

So several protein assays are available, but that's only half the battle. For a commodity like milk, an assay needs to be "reliable, cheap, and fast, because you're talking about millions of tons of ingredients at that stage in the supply chain," Moore says. That rules out assays such as the enzymatic amino acid analysis, which requires a 24-hour digestion period. "But researchers are already looking into ways to accelerate the digestion step," Moore notes.

"In the end, every method has its advantages and disadvantages," Lipp says. "Some may be lengthier but more precise. Others might be quicker and cheaper but more easily misled by the presence of melamine, for example." He says that the USP workshop will provide stakeholders an opportunity to hash out "what is currently useful and what might emerge as being useful and how this can be integrated with the needs of various users of those methodologies within the food supply chain."

The Food & Drug Administration is working with USP both through the workshop and by helping to determine potential alternatives to traditional protein assays, says Gregory Diachenko, director of the analytical chemistry division at the agency's Center for Food Safety & Applied Nutrition. Once the experts come up with some promising procedures, FDA is ready to help with further development and validation of the new methods.

Ultimately, if the scientific community settles on some new methods for measuring the protein content of foods, USP might adopt the methods in FCC. Manufacturers comply with this source of internationally recognized quality and purity standards and associated testing procedures when producing food ingredients. Diachenko says FDA uses FCC specifications in some of its regulations, so the agency might need to alter its regulations to reflect the new FCC standards.

Although the ability to test food for quality and safety currently resides with analytical labs, Purdue's Cooks believes consumers will eventually want to test their food at home. And he thinks that the technology will be available "sooner rather than later."

To that end, Cooks and his colleagues have been shrinking analytical equipment to make it more affordable and portable. They have already developed a handheld mass spectrometer that weighs just 5 kg, so testing of samples can now be done in the field with no sample prep (Anal. Chem. 2008, 80, 7198). The Mini 11 spectrometer, which is compatible with a variety of ionization sources, has a detection limit in the parts-per-billion range for a variety of compounds. Cooks is working on an even smaller and lighter version.

The team has built 10 of these devices already and sold them at cost to researchers at other institutions, including universities, the Army Corps of Engineers, and pharmaceutical companies. Purdue has licensed some of the underlying technology to ICx Analytical Instruments.

So consumers may soon have a more precise instrument than the eye or nose to find out if the quality of their food is acceptable.