Endocrine disrupters found in bottled water

Naomi Lubick, Environ. Sci. Technol., April 8, 2009

New research analyzing mineral water held in bottles made from polyethylene terephthalate (PET) raises questions about whether contaminants might leach from PET into the water where they mimic estrogen’s effects.

[ ALSO NOTE this earlier study presented at the Society for Risk Analysis 2001 Annual Meeting. ]

In the study reported March 10, 2009 online in Environmental Science and Pollution Research, ecotoxicologists Martin Wagner and Jörg Oehlmann of Johann Wolfgang Goethe University (Germany) report evidence of the bottles’ estrogenicity from multiple tests, but they have yet to pinpoint the exact source.

Billions of bottles and food containers made of PET are sold every year. The plastic is considered safer than others that contain endocrine-disrupting compounds, such as polyvinyl chloridewhich is made with phthalates and bisphenol A (BPA) and polycarbonate, which has been shown to release BPA into liquids at high temperatures.

For the new study, Wagner and Oehlmann used both a yeast-based assay and a reproduction test with the New Zealand mud snail Potamopyrgus antipodarum to tease out whether traces of chemicals in PET or other compounds mimic estrogen’s activity. The researchers tested 20 brands of mineral water sold in either glass or plastic bottles or both.

The yeast-based assay of different samples of mineral water showed that more than half the brands of water had “significantly elevated estrogenic activity,” the researchers note. On average, the effects seen were similar to those elicited by a dose of about 18 nanograms per liter of 17β-estradiol (a natural estrogen). For all but one brand, mineral water stored in plastic bottles had higher estrogenicity than the same water stored in glass bottles. And multiuse PET bottlesmeant to be refilled several timesshowed lower estrogenicity than the bottles meant for one-time use.

The researchers also raised snails in both plastic and glass bottles for up to 56 days. The animals in plastic bottles were shown to have higher reproductive rates than average, and they created more than twice as many embryos as the snails housed in glass bottles.

The formulation and possible chemical content of PET vary. Past studies from William Shotyk of the University of Heidelberg (Germany) and his colleagues (e.g., DOI10.1021) have documented that antimony leaches from plastic water bottles in different amounts according to temperature, length of time on the shelf, and other storage conditions. Antimony or other heavy metals could be acting as an endocrine-disrupting compound in plastics, but such effects are not proven.

The new study has generated some disagreement. PET has long been approved for food container use by the U.S. Food and Drug Administration. In response to the new research, industry trade groups (such as the PET Resin Association in North America) reaffirm that past studies have shown no endocrine-disruption effects from PET in standard lab animal tests. The German Federal Institute for Risk Assessment, known as BfR, released a statement on March 18 calling the results inconclusive. The BfR statement noted that nonylphenols in bottle lids, for example, might be contaminating the waters tested. The agency called for further research.

But Frederick vom Saal of the University of Missouri Columbia, who advised the researchers on the new experiments, believes that the new findings add PET to the list of plastics that expose humans and the environment to harmful contaminants. “If these snails are showing a response that is clearly an estrogen response” to endocrine disrupters in the plastic or the water, says vom Saal, “then there’s enough estrogenicity in these containers that they can stimulate abnormal reproductive behavior in an animal.” The snails are a “canary in the coal mine,” he says, and “as far as we know, not as sensitive to these hormones as a human cell.” The study also clearly shows high estrogenic activity coming from some but not all ethylene-based plastic bottles, adds vom Saal.

“We are now trying to figure out which compounds are hiding behind this estrogenic activity,” which will be a “hard task,” Wagner says. “We can’t expect that there’s one chemical present at high concentrations that’s causing this activity. It may be more realistic to expect the presence of 5 or 10 estrogenic compounds” acting in concert. Also, estrogenicity of compounds such as phthalates or BPA occurs at such low concentrations that detecting the chemicals’ presence at active doses is difficult.

“It’s a work in progress,” Wagner adds. “You can’t conclude [from the current results] that it’s dangerous to drink mineral water, but it’s clear that we are exposed to [many] more endocrine disrupters than we currently think.”