Most of us never give a second thought to the dust in our homes, except to sweep it up and throw it away. But for scientists, that dust is a goldmine of information about the chemicals we encounter every day. Unlike soil, water, or air, most of the chemicals in dust come directly from us or our stuff, and we spend so much time indoors that if scientists find something in dust, there’s a good chance it’s also in us. So when a team of Canadian scientists found some potentially cancer-causing chemicals in house dust, they knew they had to investigate further.
Hui Peng and the team didn’t set out to find these chemicals. Rather, the chemist at the University of Saskatchewan was hoping to solve a puzzle. One of the more concerning chemicals in dust are flame retardants that contain the element bromine, some of which can disrupt hormone activity or impair fetal development. But there’s far more bromine in house dust than the flame retardants can account for.
To solve that puzzle, Peng suspected he could use a new tool that he and others in the lab of Dr. John Giesy. The tool, called DIPIC-Frag, sifts through complex mixtures like dust and finds unknown chemicals. It’s an advanced sort of mass spectroscopy, a method of identifying chemicals by breaking them apart and then using an electromagnet to sort the electrically-charged fragments by mass.
Typically, finding the signal of a single chemical in the noise of a sample like dust is only possible by comparing the sample to a standard. But DIPIC-Frag uses a statistical analysis of the signals to find the chemical formulas of hundreds of molecules at once.
Bromine-containing molecules are particularly easy to identify with mass spectroscopy, because bromine atoms come in two different masses, or isotopes, of approximately equal frequency. The mass spectrometer can zero in on this pair of signals, eliminating the noise from other chemicals in the sample.
The method builds on years of advancement in what such analysis is capable of. “The science is good enough to do this,” said Dean Jones, a professor of medicine at Emory University who was not involved in the study. “They’re pushing the limits on what the technique is.”
When the team used DIPIC-Frag to look for bromine-containing molecules in dust samples collected from seven homes in Saskatoon, Saskatchewan, about 17% of the molecules they identified were the expected flame retardants. But 56% of the bromine-containing molecules had chemical formulas that marked them as related to azo dyes, commonly used to color clothing and other textiles.
It was a startling discovery. “Most of these compounds have never been reported before,” Peng said.
Most azo dyes are harmless by themselves, but when exposed to UV light or digested by bacteria on our skin, some can be degraded into smaller molecules that can enter our cells and damage DNA. What’s more, some compounds used in the manufacture of these dyes can still cling to clothes. Azo dyes that are known to degrade into harmful chemicals are banned in the European Union and seldom used elsewhere. The scientists were concerned that they had found so many previously undiscovered molecules similar to these toxic precursors and degradation products. The team decided to test how one of these molecules, a precursor called BNA, affected the DNA of the bacteria Salmonella .
Even at the low concentrations found in house dust, BNA caused mutations in the Salmonella DNA. The results might solve another mystery of dust: “most of the toxicity cannot be explained by compounds we’ve seen before,” Peng said.
But knowing whether BNA is in fact a cancer risk will take much more testing, said Jianxian Sun, a co-author on the study who performed the toxicity testing. Other compounds will be even harder to test, since many of them aren’t readily available for the scientists to buy and test.
In the meantime, there’s no reason to panic, Jones said. “We’re probably exposed to a million chemicals in our lifetime,” he said. Most of these might not be harmful enough to be worth regulating, but could be risk to individual people in certain cases—if they’re particularly sensitive, or if they’re exposed to higher amounts for a long time.
But the fact that Peng and his collaborators identified so many new chemicals is also good news for the fight against toxic chemicals he said, if it means that other potential threats can be discovered and examined more rapidly and cheaply. “We want to push this strategy in the future,” Peng said, using DIPIC-Frag to find chemicals that are already in the environment and decide if they should be regulated, rather than attempting to track all manner of industrial chemicals from their source.
Such regulation would be important for chemicals that show up in something as ubiquitous as house dust, said An Li, a professor of Environmental and Occupational Health Sciences at the University of Illinois at Chicago. “If you find high levels in house dust, it’s very likely humans are exposed through skin contact or food,” she said.
Li studies pollution in the Great Lakes, where DIPIC-Frag has also been useful. “Their machine is capable of something we cannot do here,” she said. Last year, Li and the Saskatchewan team found more than a thousand new bromine-containing chemicals in samples from Lake Michigan, some of which were also potentially toxic.
But most of the chemicals found in the Lake Michigan study were “natural products,” Peng said—not human-made, and therefore hard to decrease with human efforts. Dust, on the other hand, is almost entirely manmade chemicals.
Even so, Peng acknowledges that regulations of chemicals such as those in the house dust will be slow to change. “This is a very promising strategy, but it may take a long time” to assemble enough scientific evidence to know how to act.
The more immediate effect of methods like DIPIC-Frag may not be in regulation, but in medicine, said Jones, the Emory professor. In the future, he said, “we’ll still be regulating chemical by chemical,” rather than using techniques like DIPIC-Frag. But if techniques for analyzing thousands of chemicals at once become cheap, he said, “one could in fact have devices in a person’s home to evaluate what they’re exposed to.”
Jones thinks that cheap methods for evaluating what people are exposed to could do for the human “exposome” what genetic sequencing promises to do for the human genome—give us personalized medical advice based on massive amounts of data.
The next step is to check if chemicals like BNA show up in dust in places other than Saskatoon. George Cobb, a professor of environmental science at Baylor University, is collaborating with scientists at Murray State University in Kentucky to collect samples from sites around Texas and Kentucky. If those samples contain similar brominated dyes to the Canadian ones, it’s a good indication that this is a concern that crosses borders and regions.
“The method is very easy to adopt,” Cobb said. If he finds the same bromine-containing chemicals as the Canadian team, it will start the long process of testing the various chemicals for toxicity—in bacteria, in computer models, and finally in animals. Then another long review before any regulations are changed.
But to Cobb, that’s all in the future.
“What we really want to know first is what’s in there,” he said.