Metabolic “breathalyzer” could diagnose disease from our breath
By Darren Quick
February 6, 2012
Scientists - and dogs - have known for some time that our breath can reveal much more about us than our estimated blood alcohol content. Researchers at the University of Wisconsin-Madison believe that "breathalyzer"-like technology they currently have under development could be used to diagnose a wide range of diseases, such as diabetes, cancer and infections. Such technology, which relies on the fact that many diseases alter the body's metabolism in distinctive ways, would provide a non-invasive method of detecting disease even before typical symptoms appear.
Under different conditions, the human body relies on different sources to produce energy. "When we're healthy we use the food that we eat," says co-author Warren Porter, a UW-Madison professor of zoology. "When we get sick, the immune system takes over the body and starts tearing apart proteins to make antibodies and use them as an energy source."
When the body shifts from sugars to proteins for its energy source, different biochemical pathways in the body are engaged, which results in distinct changes in the carbon isotopes that show up in exhaled carbon dioxide. It is these changes that the researchers are relying on to signal the earliest stages of disease.
For their study, the researchers looked at mice with metabolic symptoms similar to those seen in women with polycystic ovary syndrome (PCOS), which causes a wide range of symptoms including infertility, ovarian cysts, and metabolic dysfunction. PCOS, which affects around one in ten women, can currently only be diagnosed after puberty by excluding all other likely diseases - obviously a time-consuming process.
"The goal is to find a better way of diagnosing these women early on, before puberty, when the disease can be controlled by medication or exercise and diet, and to prevent these women from getting metabolic syndromes like diabetes, obesity, and associated problems like heart disease," says study senior author Fariba Assadi-Porter, a UW-Madison biochemist and scientist at the Nuclear Magnetic Resonance Facility at Madison.
By measuring the isotopic signatures of carbon-containing metabolic byproducts in the blood or breath, the researchers were able to detect distinct metabolic changes in the mice. This was done by injecting glucose containing a single atom of the heavier isotope carbon-13 into the mice in order to trace which metabolic pathways were most active in sick and healthy mice. Within minutes, the researchers were able to measure changes in the ratio of carbon-12 to carbon-13 in the carbon dioxide exhaled by the mice. The researchers say the technique is sensitive enough to detect statistically significant differences between even very small populations of healthy and sick mice.
"The pattern of these ratios in blood or breath is different for different diseases - for example cancer, diabetes, or obesity - which makes this applicable to a wide range of diseases," explains Assadi-Porter.
The researchers say this approach has numerous advantages. Not only does it simplify diagnosis by relying on a single measure to survey the workings of the entire body, but the rapid feedback it provides could help ascertain the effectiveness of treatments.
"With this methodology, we have advanced methods for tracing metabolic pathways that are perturbed in disease," says Assadi-Porter. "It's a cheaper, faster, and more sensitive method of diagnosis."
Although the technique currently uses a machine about the size of a shoebox, the researchers envision a small, hand-held "breathalyzer" that can easily be taken remote areas. They have co-founded a company, Isomark, LLC, to develop the technology and hope it could also provide a better understanding of the underlying biology of disease and whether the biochemical changes they can detect are causative or side effects.
The research team's study is published in the journal Metabolism.
Source: University of Wisconsin-Madison.
Just enter your friends and your email address into the form below
For multiple addresses, separate each with a comma