5/1/2003

Consumers' Research Magazine

By Beatrice Trum Hunter

New Alternatives in Safety Testing (testing product safety without using animals)

Live animals, used to test product safety, have long played a fundamental role in the science of toxicology and pharmacology. Their use has kept many harmful substances out of the marketplace. However, new and emerging techniques offer opportunities to obtain reliable scientific data faster, less expensively, and more humanely. These techniques can reduce significantly our dependence on tests with live animals.

In November 1999, the White House Council on Environmental Quality achieved an agreement with animal protection groups that is likely to cause significant changes in a new program launched to test the toxicity of 2,800 high-production volume chemicals. The agreement, developed jointly by the Environmental Protection Agency (EPA), the Chemical Manufacturers Association, and Environmental Defense (formerly called the Environmental Defense Fund), is scheduled to begin this year and be completed by the year 2004.

The chemical industry has agreed to pay for the program. In addition, three federal agencies--the EPA, the National Institute of Environmental Health Sciences, and the National Toxicology Program--will provide $5 million for two years of research to develop non-animal tests. Animal protection groups urge a decrease in the number of animal tests. There is general consensus that presently, some animal testing is unavoidable and necessary for public health and safety. The goal is to reduce it to a minimum, and to develop and use effective alternative testing techniques whenever possible.

The Johns Hopkins School of Hygiene and Public Health and its Department of Environmental Health Science have been encouraging the development and use of in vitro (test-tube science using cells, tissues, and organs) for toxicology, to replace in vivo (live animal) toxicology.

In 1988, the Center for Alternatives to Animal Testing (CAAT) was established at Johns Hopkins. Funded by a coalition of more than 70 corporations, government agencies, and private individuals, CAAT had led in vitro testing techniques, validation of their safety and efficiency, and dissemination of information concerning non-animal models for toxicity testing.(*)

CAAT has awarded grants to develop and refine in vitro testing for neurotoxicity, skin irritation, eye irritation, and liver or kidney toxicities of various products. One grant, for example, concerns an in vitro model that may predict whether exposure to a test chemical would be painful, without having to subject test animals to undue stress. Another grant is for a project to develop models of human skin, using cultures of keratinocytes (cells in a layer of skin). Such models could replace or reduce the use of animals in studies of skin irritants or allergens.

The use of synthetic skin, instead of live rabbits, can also test chemicals for their corrosiveness. This new test, endorsed by the EPA, Consumer Product Safety Commission (CPSC), and the Occupational and Safety Health Administration (OSHA), is part of a federal effort to reduce animal experiments.

Two Traditional Tests. The [LD.sub.50] test is highly contentious. The abbreviation signifies the lethal dose for 50% of the animals exposed to the substance being tested. This test had been used traditionally to identify acute toxicity for cosmetics, detergents and other household/industrial products, pesticides, and food additives. The test requires the use of large numbers of laboratory animals. There is general scientific consensus that the test is crude, and that it gives no information about chronic effects.

Some federal bureaus and toxicology programs have dropped the requirement of using the [LD.sub.50] test. The FDA does not require [LD.sub.50] to be used, but does require some type of testing for acute animal toxicity to determine the safety or toxicity of regulated products.

The Draize Eye Irritancy and Skin Tests are as contentious as [LD.sub.50]. In the Draize test for eye irritancy, a test chemical is placed in one eye of live, restrained animals (usually rabbits), to learn the extent of damage that the chemical causes. The effects may be seen in blood vessel breakdown, and in corneal bulging and opacity.

A number of non-animal tests have been devised to replace live-animal use with the Draize test for eye irritancy. By using a battery of five or six in vitro tests, a range of irritant responses can be identified. In 1988, it was estimated that the cost of in vitro tests averaged about $50 per product, compared with $500,000 for in vivo tests.

One of the in vitro tests that can replace the Draize test is applicable for many cosmetics and toiletries, including eye-shadows, colognes, perfumes, shampoos, hair conditioners, bubble baths, skin creams, and skin cleansers. The test product is added to a synthetic mixture that simulates the living cornea in the human eye. The potential irritation is scored according to a matching of color gradations that signify molecular effects on the tissue.

The Agarose Diffusion Method (ADM) is a non-animal screening test, reported to reduce by 80% to 90% the numbers of animals needed for eye safety testing. It has been available since 1989. ADM is a tissue culture test system, using an overlay of agarose (a non-gelling fraction of the algae, agar). The test product is applied to filter paper discs, and placed on the agarose surface. The effect on the cells determines whether the test is positive or negative. ADM can be used to test cosmetic liquids, pastes, creams and powders, as well as products that can be diluted in water or hydroalcoholic systems (relating to water and alcohol). Results obtained from ADM were as favorable as those from the Draize test.

A cell tissue test using a barrier, called transepithelial electrical resistance (TER), is another alternative to the Draize test. Normally, cells set up a resistance barrier between adjacent cells to prevent charged particles from passing between them. TER is lowered if the cells are disrupted by a toxicant. In this test, an apparatus continuously bathes the cells in liquid. Each chemical applied to the system decreases TER in a particular way, and over time, produces a "signature" that can predict roughly the eye irritancy of the product.

Another test uses epithelial cells from rabbits, grown in culture dishes, that are exposed to test chemical solutions such as household bleach, rubbing alcohol, or other potential irritants. These substances injure the cells by disrupting their secretions of an enzyme, plasminogen activator (a normal cellular product). The degree to which a test chemical inhibits the release of the enzyme correlates quite well with the chemical's toxicity.

Corneal epithelial cells from the rabbit also can be grown in multilayered sheets of culture. They can be "wounded" by killing and removing the cells from a small swatch of the tissue. Normally, such cells would heal within 48 hours. However, if treated with a toxicant, the wound's closure rate is slowed. Measuring the time required for healing correlates with the degree of toxicity.

Mast cells--cells that release substances in response to tissue inflammation or injury--can also be used in eye irritancy tests.

The Chorioallantoic Membrane Vascular Assay (CAMVA) can also be used instead of the Draize test. The chorioallantois is a nourishing fetal membrane found inside of the eggshells of birds and reptiles. Young, fertilized chicken eggs are used in CAMVA. Tiny windows are cut out of the eggshells to expose the white membranes. The test substance is applied. The membrane reacts quite like the human eye.

Bacteria can be used in a test that substitutes for the Draize test. Certain bacteria have the ability to emit light. When the bacteria and the test substance are mixed in a solution, the change in expected light emission signals a potential irritancy of the substance.

The rabbit's corneal cells are not entirely equivalent to human corneal cells. A ready source of human corneal cells is available for tissue testing from the tissue normally discarded during human eye surgery. These cells may give more accurate results than corneal cells from rabbits.

In the Draize test for skin irritancy, the skin of restrained guinea pigs or rabbits is shaved and painted with the test chemical daily for several weeks. The same test substance is injected into the animals. Blisterings and swellings of the skin are noted.

As an alternative to this procedure, the ears of mice are painted with the test chemical, and the lymph node tissues are examined for signs of toxic or allergic reactions. This test uses fewer animals and is reported to be less painful and stressful to the test animals. Also, the mouse test can be completed in one week; the guinea pig or rabbit painting and injection test requires three to four weeks.

Newer Safety Testing. Four regulatory agencies--the FDA, EPA, CPSC, and OSHA--will accept results from contact dermatitis testing methods that use fewer animals than in traditional testing.

Tissue testing is one alternative. For toxicity testing, human skin can be grown in a laboratory dish from the epidermis (the outer protective non-vascular layer of skin). The equivalency of living skin can be reconstituted from cells and matrix molecules.

Several chemical tests are available to gauge a range of skin reactions. One, known as a Critical Micelle Concentration Test, monitors a substance's potential to break down into a single molecule. A single molecule is more likely to penetrate the skin and cause irritation than are clusters of molecules.

The value of tissue testing has been established. In 1996, the first European human tissue bank was founded. Donors can make arrangements to offer their tissues after death, similar to organ donations. The tissues are kept in test tubes, to be used for in vitro testing.

Over time, it has become increasingly clear that toxicity tests do not necessarily uncover additional types of damage that can be inflicted in the human body, such as carcinogenicity, teratogenicity, and mutagenicity. This awareness has led to mandated tests which might reveal the potential of substances to induce such effects.

One technique that uses tissue cells has been described as recapitulating the cancer process in a petri dish. Embryo cells from Syrian hamsters are stimulated to divide uncontrollably. This technique is useful for mandatory carcinogenic testing. In use with more than 300 chemicals, this technique was able to predict the outcome of live-animal rodent tests 80% to 90% of the time.

Several different tests are being used for genotoxins (substances that poison genes). The Ames Test, based on bacteria, can test for mutagenesis (substances that produce mutations in the genes). The Williams Test shows DNA repair in hepatocytes (celis in the liver). Some tests can identify genotoxins and carcinogens in polycyclic aromatic hydrocarbons (PAHs) in tobacco smoke; nitrosamines in cured meat, and in salted and pickled foods; and heterocyclic amines formed in some cooking processes.

In vitro testing for drugs can screen compounds for their effects on fetal development. Stem cells are general-purpose embryonic cells that can develop into any type of cell in the body. Stem cells have a very reliable pattern of development into tissue. Researchers can measure precisely any disruptions to the number of cells, their quality, and the timing of development. This technique provides a means of looking for subtle chemical effects that might result in birth defects in particular organs.

Cell culture techniques can be used to test chemicals that cause birth defects. Cells shed by the human fetus can be obtained from the amniotic fluid of a pregnant woman who is monitored because she is at risk for producing offspring with genetic diseases. The shed cells are exposed to potential teratogens (substances that induce birth defects). Then, the cells are checked to learn to what extent they have produced a set of biomolecules called "stress proteins" that otherwise would not have been produced. These proteins may play a protective role. The amniotic fluid cells respond dramatically to toxic agents in vitro tests. The technique could be used to monitor environmental stress to the fetus, and also be used in standardized laboratory tests for teratogens.

To test toxins, "DNA chips will be the source of the next reduction in animals used," predicts Horst Spielmann, Director of Berlin's National Center for Documentation and Evaluation of Alternative Methods of Animal Experiments. DNA microarrays are used commonly to track patterns of gene expression (an array is an arrangement). Microarrays could be used as a genetic technique to study cellular responses to test compounds. A single DNA chip carries an array of hundreds, or even thousands, of short strands of DNA. Each one acts as a probe for a specific gene. Such arrays might reveal which gene is turned on by the cell in response to a toxic compound. Because arrays probe the human cell's activity directly, they might serve better than animal tests in predicting toxicity to humans.

All heart-cell cultures beat rhythmically in culture flasks. If they are exposed briefly to a toxic volatile chemical (such as the solvent carbon tetrachloride or to the inhaled anesthetic halothane), the contractions of the beating-heart culture become irregular. This feature can be used to screen toxic volatile compounds for cardiac toxicity. There are numerous harmful volatile compounds in many household and office products, including paint removers, glues, and aerosol sprays that can induce an irregular heartbeat.

Cultured kidney and liver cells can be used to probe the mechanisms of toxicity in these organs, and to flag suspected toxins such as halogenated hydrocarbons (pesticides) early in the toxicity assessment process. Isolated kidney tubules are very sensitive to certain heavy metals, and to injury from certain organic compounds. In a living organism, tubule injury can result in acute kidney failure.

In vitro tests can be used to measure an inflammatory response. Many compounds can irritate sensitive tissues, making them red, swollen, and painful. Certain drugs, such as the antibiotic tetracycline, can make skin more susceptible to irritating and damaging effects from exposure to ultraviolet light (a phototoxin). Testing can be done with cultures of human cells called peripheral blood mononuclear cells. Like the skin cells, they are more likely to die from exposure to ultraviolet light if pretreated with a photo-sensitizing chemical.

In vitro liver cell cultures can be used to test the toxicity of common analgesic drugs, such as acetaminophen. Also, such cultures can make it possible to learn more about molecular mechanisms that underlay inflammation, membrane damage, and tissue toxicity.

A new in vitro test, with monoclonal antibodies, may replace a widely used technique known as the mouse ascite method. This traditional test involves a million mice yearly, and involves injection of tumors into mouse abdomens and extraction of antibodies with needles. Although the National Institutes of Health have not banned the mouse ascites method, the agency approves and encourages the use of the in vitro method.

The energy-producing processes within a cell can be used as a sensitive test to detect toxic substances in food, water and soil. The assay uses bits of membranes from the cell's mitochondria--the main site of energy production within cells. They contain the enzymes that generate energy. The mitochondria are sensitive to many types of chemicals. In vitro bioassays for toxic compounds can be conducted with mitochondria from fresh beef hearts obtained from slaughterhouses.

Isolated rat liver cells are useful in flagging a toxic metal such as cadmium. When present in the body, cadmium causes cells to produce more metallothionein (a protein that binds cadmium) in order to protect the cells. Total protein synthesis and metallothionein induction could serve as markers for cadmium toxicity.

Cultured mouse embryos can be used to test for toxins such as lead and mercury at the stage when the embryo's new genome begins to express itself. This expression makes it possible to quantify the direct genetic damage caused by the toxin, without having to use live animals.

The lowly earthworm provides a clue that soils have been poisoned. Although many earthworms thrive in soils in good tilth, they are scarce in soils that have been treated heavily with chemical applications. Now, the earthworm may replace laboratory animals as a tool for environmental toxicology.

Tests may indicate how substances such as PCBs or heavy metals impair an earthworm's immune system. Investigators periodically extract and examine the earthworm's immune cells to check their ability to engulf bacteria. Certain immune processes are common to many species, including humans. The earthworm may be useful as an early warning system to indicate whether further, more sophisticated testing is necessary. Also, the earthworm has a unique capability not present in test animals; it provides, through its soil contact, a means of evaluating clean-up work at hazardous waste sites.

Applying Newer Technology. Molecular biology is used in newer safety testing. Toxicologists are moving from a high to a low-dose method of testing, by using ultrasensitive probes that can detec ttoxic effects at a cellular level. Instead of studying animals stressed to the maximum with near-lethal doses of test toxin, toxicologists may look for biological processes much closer to normal ones in order to understand the unifying biomechanisms by which damage in inflicted. Then, they can formulate preventive measures for those who are exposed.

With these changes, engineered models derived from human cells, tissues, or organs are useful. For example, instead of the traditional high-dose animal models, cultured liver tissue can be used to test for cancer risks. The new models may allow toxicologists to demonstrate definite cause and effect, rather than merely to suggest possible associations.

The molecular structure of a substance can predict many toxic properties. Using a database of more than 12,000 chemicals, investigators have been able to identify substances so similar to the known toxicants that there is no need to test them individually.

Ultrasensitive chemiluminescent and bioluminescent tests, using DNA detection and immune assays, can be used to test for pathogens where food is prepared. Results can indicate good or poor practices of sanitation and personal hygiene. The potential of these tests has not yet been utilized fully.

Computerized models can determine if a substance is irritating or poisonous, based on its chemical and physical properties. By using this information, as well as a database compiled from other tests, it is possible to narrow down the number of ingredients in a product that should be tested. Mathematical models can give information about the quantitative structure/activity of a substance. The relationships can identify potentially toxic chemicals. Compatible software programs have been developed to predict endpoints, based on the structure of a chemical. This information lessens dependence on the traditional tests such as [LD.sub.50] and the Draize tests.

Computers also play an important role in new safety testing techniques using cloned human genes. The National Institute of Environmental Health Sciences (NIEHS) recently opened a center to test chemicals for their ability to interfere with functions of human genes. Such interference can lead to cancers or genetic mutations. The testing program uses copies of about 2,000 genes out of some 80,000 total. The cloned genes are arranged in a grid pattern on a glass slide, and then are exposed to test chemicals. Computers can rapidly determine whether the substances turn the genes off or on. Initially, NIEHS plans to use the cloned gene test with chemicals already known to cause cancers or genetic mutations. If the tests demonstrate reliability, they may replace some laboratory animal tests.

The Eye Irritancy Test

In the spring of 1933, a woman designated anonymously in the archives of the FDA as "Mrs. Brown" wanted to darken her eyelashes. She applied an eyelash dye product that promised to "radiate personality." Instead, she suffered constant pain for three months. Ultimately, her corneas sloughed off, and she became blind.

Mrs. Brown's experience earned her a place in the "Chamber of Horrors" exhibit that the FDA presented to members of Congress to demonstrate the need for passage of the Federal Food, Drug and Cosmetic Act of 1938. Enactment of this legislation probably prevented countless numbers of tragedies similar to Mrs. Brown's. The law authorized the FDA to prohibit the sale of harmful cosmetics. As a result, many manufacturers tested the safety of products intended for use in an around the human eye. The Draize test came into common usage.

Animal Test Limitations

Most developments in understanding human diseases such as hepatitis, heart disease, and cancer, have come from clinical investigations. Animal research has served mainly to dramatize what has already been discovered. In at least one instance, undue emphasis on animal research actually delayed discoveries. Prior to 1963, dozens of clinical studies linked smoking to lung cancer. But, conflicting results from animal studies convinced many investigators that no such relationship existed. The consequent delays in public health warnings may have cost unnecessary suffering and loss of lives.

Cosmetic Testing

The FDA does not require cosmetic manufacturers to conduct safety tests for their products. However, the agency strongly urges them to conduct toxicologic or other tests to substantiate the safety of products. If the tests are not performed, the products may be considered misbranded, and subject to regulatory action unless the product's label bears a cautionary statement such as: "Warning: the safety of this product has not yet been determined."

Responsible manufacturers usually conduct tests in order to ensure the safety of their products. Testing protects them from liability suits and bad publicity that can result from unsafe products.

A marketing ban becomes effective July 1, 2000 on all cosmetic products containing animal-tested ingredients within the European Union (EU) countries. A new proposal would extend the regulation, by turning it into a full-scale ban on all animal testing of cosmetics within three years after acceptance. The proposal would ensure compatibility with World Trade Organization (WTO) agreements, and avoid marketing disagreements between WTO member countries.

(*) In vitro models fail to mimic the complexity of the in vivo model, which uses a whole living organism. Single in vitro tests are not reliable; a battery of different tests yields better results. In vitro screening tests are useful for screening out severely irritating compounds in products in the early stage of testing, so that they can be eliminated long before the products undergo further testings.

Validation of an in vitro test requires that it be used with a wide variety of compounds, and be tested in different laboratories. In vitro findings must be compared to in vivo testing data. The results must demonstrate that the in vitro test is predictable, reliable, and reproducible. Also, it must be somewhat comparable to the results obtained with live animals.

Mrs. Hunter is CR's food editor