Press release for the start of the diXa project

Major EU grant to combine toxicogenomics data

Once upon a time, you could run a lab experiment to identify the harmfulness of a substance that yielded a single result, then start doing your calculations on the back of an envelope. Those days are over. In toxicogenomics – Kleinjans’s primary research field – experiments now look at how all 30,000 or so human genes react to a certain substance. The result: a huge pile of data to analyse and interpret. The latest technology that Kleinjans now has in-house yields 2.5 million results per measurement. These data have to be gathered, saved and checked, then subjected to statistical techniques to interpret them at a biological level. In other words: to find out whether the substance is doing something to a cell. The upshot: laboratory workers who are first and foremost toxicologists, like Kleinjans, now also have to be well versed in informatics and computer science. For this reason, around 20% of the Toxicogenomics department now consists of ‘toxico-informaticians’.

So how does all this work in practice? Kleinjans: “First you throw a substance onto a few cells in a culture dish. Then you measure all gene reactions, all protein reactions, after 12 hours, after 24 hours, and so on. This yields an amazing amount of data on how the cell reacts to that substance. Even we don’t yet understand some of this information, because till date we haven’t yet unravelled all the different functions of the genes. That’s no problem, just as long as you save and store your data well, so that a scientist from Kuala Lumpur can make a fantastic discovery in six years or so. After all, with the knowledge from the day after tomorrow you can look back on today’s data.”
And this is where the new project comes in, which Kleinjans coordinates: Data Infrastructure for Chemical Safety, or DiXa for short.* Seven European consortia, each using toxicogenomics to work towards alternatives to animal testing, will be making all their raw data available. “Until now, sharing data hasn’t been part of our culture. So it’s a good thing that we all know one another within the participating consortia, because mutual trust is essential in this project.” Over the coming three years a web-based database will be developed to preserve data in a sustainable manner, and subjecting this data to meta-analyses will provide deeper insight into the working mechanisms of toxic substances in human cells. “Because ultimately these little cells need to predict what will happen in the intact human, and they need to do this better than the average rat experiment.”

Coin toss
This won’t be so hard, according to Kleinjans, because the average laboratory animal toxicity tests correctly predicts the consequences for humans in only 60% of cases – barely better than a coin toss. “Very few new medications have been introduced in the last ten years. This is because in 30% to 40% of cases, the last stage of development shows a prospective medication to be harmful for humans. Despite the fact that it seemed to be fine in the laboratory animals. A lot of money goes down the drain this way.” And these days, it’s more and more difficult for harmful effects to slip through the cracks of the testing circuit: “If we had to approve paracetamol with the knowledge we have now, it wouldn’t happen. Paracetamol kills liver cells, which can potentially also result in hepatitis. Use of paracetamol appears to be the major cause of acute liver damage, perhaps because people exceed the recommended dose of 4 grams per 24 hours, but possibly also because some of us are more sensitive to it. In the States, FDA is considering lowering the recommended dose. Now that’s progressive insight.”

One industrial sector desperate for alternatives to animal testing is the cosmetics industry. European law stipulates that new cosmetics can no longer be tested on animals. “But new cosmetics do need to be evaluated before they can be allowed to enter the market. And another piece of legislation that puts demands on safety still requires animal testing. Of course, no-one would die if new lipsticks and creams could no longer be developed. But economically, this is a problem for Europe.” To date, only two testing models have been developed that have actually replaced animal testing: the eye-irritation test, which means live rabbits no longer need to be used (cow eyes from offal are used instead); and the skin-irritation test, which shows whether skin is damaged by a particular substance. “With cosmetics you might think that the eyes and skin are the most important targets to research. But the latest cosmetics are also intended to penetrate through the skin and polish away wrinkles, for example. So these preparations have to be biologically active: you can basically call them a drug.”

So why have only two alternatives to animal testing actually been put into practice? The reason, according to Kleinjans, lies with the European regulators. “The staff of the agencies that have to assess and approve this are civil servants rather than scientists, who have no view of modern developments. The older generation of regulators understands nothing of toxicogenomics. Of course, it’s ridiculous that it takes 10 or 20 years before a reliable alternative to animal testing is taken up in legislation. The government – both national and European – shouldn’t just accept that, in my view. They pump a lot of money into this sort of research and then the best they can do is hope that something will come of it. I think that’s a strange approach to resource management.” This, however, does not stop Kleinjans and his colleagues from forging on with the development of alternatives to animal testing. “This is a major scientific challenge. The field has become more technological, more exciting, more complex. We think we’re on the brink of a revolution of biology.” In December last year, Kleinjans’s colleagues sent their first alternative test to the European Centre for the Validation of Alternative Methods in Italy. “And it’s been there ever since. Aside from concluding that it’s relevant, they’ve done nothing with it. But in the coming years we’ll be shooting one test after the other to Italy. That was the first, but it certainly won’t be the last.”
* In addition to the Maastricht UMC+, where Kleinjans works, the DiXa project includes the European Molecular Biology Laboratory (Germany), the Max Planck Society for the Advancement of Science (German), Genedata (Switzerland), Imperial College London (UK), the EU Joint Research Centre (Italy) and the University Hospital of Cologne (Germany).