Phthalate plasticizers have been added to polymers such as polyvinyl chloride for decades to impart flexibility and durability. Their benefit to society—seen in the plethora of synthetic materials they’ve made possible—is unquestionable.
But in the 1990s, after years of believing phthalates were safe, scientists began making connections between constant exposure to the chemicals and endocrine-disrupting health effects.
That discovery led to greater scrutiny of the compounds by scientists, advocacy groups, consumers, and regulatory agencies. The plastics industry has had to stay on its toes, not only to defend its products but also to figure out exactly which phthalates are problematic and to develop safer alternatives.
One thing scientists on all sides of the debate have learned is that the size and shape of phthalates are important. Many natural and synthetic chemicals encountered at home, at work, and at play have roughly the same size and shape as hormones such as estradiol and testosterone or share structural features with them. Some of those chemicals are capable of interacting in the body with hormone receptor proteins or with enzymes involved in the synthesis or activation of hormones.
That activity can fool the body into overreacting, underreacting, or responding at inappropriate times. To complicate matters, chemicals in the same class might lead to different toxic effects. In addition, the chemicals’ metabolites and oxidation products can be more problematic than the parent compounds—for example, phthalates are diesters that can be metabolized to monoesters, which can be more disruptive.
As toxicologists and endocrinologists try to understand the cause and effect of endocrine disruption in a sea of chemical influences, chemists and chemical engineers are looking at the chemical structures of phthalates to shine light on what leads to unwanted bioactivity. The goal is to design new molecules that provide the desired plasticizing properties of phthalates but avoid the toxicity.
To that end, Michael E. Baker of the department of medicine at the University of California, San Diego, has been using three-dimensional molecular modeling to look at how a diverse array of chemicals bind to hormone receptors (Endocr. Disruptors 2014, DOI: 10.4161/23273739.2014.967138).
A molecule typically has to be the right size and shape to park itself squarely in the binding pocket of a receptor protein, like a key fitting into a lock. But proteins are not static objects. They have flexibility that gives incoming molecules a bit of wiggle room to fit. Sometimes a square peg can fit into a round hole.
Estradiol and testosterone, which are steroid hormones derived from cholesterol, have a four-ring framework, labeled A to D. The hormones bind to estrogen and androgen receptors in cells, an action that triggers the receptors to bind to DNA to regulate gene activity.
Estradiol and testosterone are structurally identical in their B, C, and D rings, Baker explains. Where they differ is in the A ring: Estradiol has a phenolic A ring, whereas testosterone’s A ring is partially unsaturated and has keto and methyl substituents. Estradiol binds to its primary receptor about 1,000 times as strongly as does testosterone, Baker says, suggesting that the phenolic A ring is essential to the binding process.
Natural and synthetic chemicals that contain a phenolic ring are likely to interact with estrogen receptors, Baker says. The hydroxyl group on estradiol’s D ring is also important in binding and activating the estrogen receptor. So when compounds contain a phenolic ring and a properly spaced second hydroxyl group, they potentially can bind with higher affinity.
Baker’s structural studies of compounds that can bind hormone receptors include genistein and coumestrol, found in soybeans; zearalenone, produced by fungi living on cereal grains such as corn and wheat; and controversial synthetic chemicals such as the polycarbonate plastic building block bisphenol A (BPA), the surfactant building block 4-nonylphenol, and hydroxylated versions of the insulating fluids known as polychlorinated biphenyls.
Estradiol binds its primary receptor about 10,000 times as strongly as BPA, he notes, and the molecular modeling shows why. BPA is too short to completely fill the binding pocket—it only contacts amino acids at one end. Baker actually found that the BPA metabolite known as MBP, which has three more carbon atoms between BPA’s two phenol rings, fits better and could be a culprit in BPA toxicity.
Baker has yet to study phthalates, but he points out that some hormones, such as 27-hydroxycholesterol, bind and activate estrogen receptors despite having a long alkyl side chain—like some phthalates— that sticks out of the binding pocket. “The key point relevant to endocrine disruption for me is that estrogen receptors are capable of binding a diverse group of chemicals beyond estradiol,” Baker says. “This structural basis for binding is also relevant for other receptors.”
A case in point is a set of toxicology studies that in part link phthalate toxicity to the compounds’ ability to activate peroxisome proliferator-activated receptors (PPARs), which help control sugar and lipid metabolism. Researchers led by Nicolas Kambia of the University of Lille, in France, recently carried out a molecular modeling study to evaluate how likely some phthalates and nonphthalate plasticizer replacements are to bind and activate PPARs (J. Enzyme Inhib. Med. Chem. 2015, DOI: 10.3109/14756366.2015.1037748). They found that phthalates interact with PPARs via contacts with their aromatic ring and their alkyl side chains.
Di(2-ethylhexyl) phthalate, known as DEHP, is currently the most widely used phthalate. Kambia and his coworkers found that DEHP is capable of weakly binding PPARs. The Lille researchers also found that, among the pool of phthalate replacements in the offing, diisononyl cyclohexane-1,2-dicarboxylate (DINCH), tris(2-ethylhexyl) trimellitate (TOTM), and acetyl tributyl citrate (ATBC) appear incapable of binding PPARs. On that basis, they expect those chemicals to have low toxicity, and laboratory studies are bearing that out.
However, PPAR binding is just one way phthalates could cause toxic effects. “Our docking study highlights the need for a more comprehensive toxicological evaluation of the biological effects of these alternatives as well as their main metabolites,” Kambia and coworkers write.
As phthalate toxicity studies proliferated a decade ago, researchers quickly zeroed in on phthalates with three to six carbons in their alkyl side chains as being the most problematic, says Mark S. Holt, director of plasticizer market development and advocacy at Eastman Chemical, a major plasticizer manufacturer.
Among the most cited studies supporting this finding is work by Environmental Protection Agency endocrinologist L. Earl Gray Jr. and his colleagues. The researchers dosed pregnant rats with different phthalates and then studied the effect on sexual development and testosterone production in the male offspring (Toxicol. Sci. 2000, DOI: 10.1093/toxsci/58.2.350).
They found that short-chain dimethyl and diethyl phthalates had no effect relative to DEHP, which was used as a benchmark. Dipentyl phthalate, with five-carbon chains, was the most potent, disrupting sexual development about three times as much as DEHP, which has eight-carbon chains. Going beyond eight carbons, diisononyl phthalate (DINP) is about an order of magnitude less active than DEHP.
The results suggest that the problematic compounds have some properties in common: their alkyl chain length, their molecular weight, and the ability of digestive enzymes to cleave an alkyl chain to form a monoester, Holt says. Some toxicity studies show that feeding rats a phthalate monoester produces similar effects as feeding them a diester, and some researchers have found that the monoesters activate PPARs.
The nonphthalate alternatives now being deployed are promising for avoiding toxicity, Holt says, although it turns out they aren’t actually new compounds. During the past 80 years, perhaps 2,000 plasticizers have been commercialized, Holt explains. Few of them have been used on a large scale, but they are a good pool to select from.
“You might not find a more highly studied class of industrial compounds,” Holt says. But researchers are doing additional testing on the replacements just to be sure.
“The level of toxicity testing being done on these compounds is now higher than in the past,” he says. “No one is going to switch to a nonphthalate plasticizer without a full set of data on PPAR proliferation, testosterone suppression, reproductive toxicity, carcinogenicity, or mutagenicity. The industry is making sure it has all the i’s dotted and the t’s crossed.”
For example, Eastman is celebrating its 40th anniversary of producing di(2-ethylhexyl) terephthalate (DEHT), which is now one of the leading phthalate replacements. In Gray’s EPA studies, DEHT did not show any disruption of sexual development in male mice relative to DEHP. Being a terephthalate, with the ester groups on opposite sides of the phenyl ring instead of adjacent to each other, is an indication that DEHT might not readily bind hormone receptors and/or metabolize to a monoester, Holt says.
When designing alternative plasticizers, says Daniel F. Schmidt of the department of plastics engineering at the University of Massachusetts, Lowell, chemists have been looking at molecules with similar molecular weight to existing plasticizers that are not too volatile or too soluble. He suggests it’s also a good idea to avoid aromatic rings.
Schmidt points to BASF’s approach to developing DINCH as a smart way to balance performance and toxicity concerns. “When people began demanding nonphthalate plasticizers, they took DINP, which is made in large amounts, and hydrogenated it. The result is DINCH, which acts a lot like its phthalate analogs but is nonaromatic. That seems to substantially reduce endocrine disruption.”
Even so, Schmidt emphasizes that despite the progress being made there’s a lot scientists still don’t know about chemical structure and toxicity. “It wouldn’t surprise me at all if a conversation very much like this one we are having now concerning phthalates would occur decades from now involving some other class of compounds we currently think of as not so bad. A measure of precaution is warranted, tempered with the realization that we will never be able to provide an absolute guarantee that any substance is totally safe.”