Urine reveals our eating habits

We already know that a urine test can establish whether someone has an infection of the urinary tract or has taken illegal drugs. But there are lots more traces to be found in urine - if you know how to read them. Developing and refining techniques to get pointers to a person’s eating habits or to harmful substances in their urine is one of the pet projects being pursued by food chemist Prof. Hans-Ulrich Humpf and his working group at the University of Münster. "Innumerable metabolic products can be found in urine which tell us something about a person’s environment," says colleague Dr. Benedikt Cramer, "for example, whether they are exposed to noxious substances such as mycotoxins, i.e. the toxins found in fungi."

In order to draw any conclusions as regards environmental factors, the researchers first have to identify so-called biomarkers. These molecules can be shown to be present in urine or, for example, in blood, by means of mass spectrometry. They provide indications of the substances where they originate which are absorbed through the food eaten. The challenge consists in finding the exact bio-markers which can be assigned to a certain parent substance - and only to this one.

Hans-Ulrich Humpf and his team are in demand because of the expertise they have, and they are also involved in international projects. For example, they examine urine and blood samples taken from mothers and their newborn babies in Zimbabwe as part of a project being coordinated by Cornell University in the USA. To a noticeably frequent extent, children in Zimbabwe are subject to delays in their development. The international team of researchers aims to discover the causes of this. Some of the candidates are certain mycotoxins in maize such as fumonisins and aflatoxins. These are absorbed by the mothers and passed on to their babies via their milk, or they come from the maize porridge which is fed to babies at an early age. Over the past few years, the Münster team has tested several thousand samples to look for the relevant mycotoxin biomarkers.

As part of another project, the food chemists are carrying out analyses to see which mycotoxins are prevalent in rice and spices in Bangladesh. "The situation is particularly challenging in developing countries," says Humpf. "While food inspection in Europe provides assurance and allows relatively low amounts of food containing mycotoxins to end up on the market, people in such developing countries sometimes regularly have higher exposure to them," he adds. However, the question is whether even very low concentrations of mycotoxins can be harmful if people eat them in contaminated products over a longer period of time. "The problem is proving the existence of these substances when the traces are extremely low," says Cramer. The detection limit is falling in line with technological progress: the new mass spectrometer being used by the Humpf team, for example, is around five times more sensitive than the previous model and can demonstrate the existence of mycotoxins which were not previously identified.

In addition to the mycotoxins, Hans-Ulrich Humpfs team is also working on another type of traces in urine. Doctoral candidate Amelie Frank describes what they are doing: "Examining urine allows us to make statements about what a person’s diet is. Do they eat a lot of meat, or do they prefer cereals? Do they have a healthy diet, with lots of fresh fruit and vegetables? As a rule, traces of foodstuffs consumed can be detected up to ten to 24 hours after they have been eaten. This could be very interesting for doctors, for example, who need to have as accurate a picture as possible of the state of their patient’s health," she explains. Frank, a food chemist, suspects that some cheating goes on when people keep food diaries; and some patients, she adds, are simply not able to keep such a diary. Instead, regular urine checks could provide a more accurate picture.

There are currently a few hundred known biomarkers for demonstrating the consumption of certain foodstuffs. However, says Amelie Frank, many of the biomarkers included in the literature have not been tested for their suitability, or cannot even be used because they cannot be reliably assigned to one single foodstuff or at least to a small group of closely related fruits or some such. She is therefore on the lookout for biomarkers which can unequivocally demonstrate the consumption of apples, citrus fruits, tomatoes and paprika. "Routinely using them in practice is still a long way off," she says, "but we have a vision that they can perhaps be used one day."