The roles of transporters in the human metabolic network

点击翻译按钮获取中文摘要

When you eat a foodstuff or a pill (i.e. a pharmaceutical drug) it is important that the relevant molecules go to the places where they will be of most benefit. How nutrients and drugs are absorbed and distributed (and eventually excreted) is thus a topic of high importance. When it goes wrong in the case of drugs they may not work properly or may even be toxic; this latter is known as an Adverse Drug Reaction, and they account for more than 5% of hospital admissions. Cells are bounded by cell membranes, whose job it is to stop them letting in any old rubbish. Instead, these membranes contain proteins called transporters that serve to ferry small molecules into and out of cells as part of the day-to-day reactions (metabolism) that keep us alive. Such transporters account for fully one third of the gene products involved in these biochemical networks. It turns out in general terms that, since all they recognise is a molecule, without knowing its 'purpose' (nutrient, drug, vitamin, etc), just these same transporters are involved in transporting nutrients, vitamins and pharmaceutical drugs into cells; the problem is that we do not tend to know which transporters transport which substances, and this correlation-based assessment is what we wish to find out here.To do this we shall study how transporter levels and the extent of small molecule uptake vary together between different cells and tissues; given enough of these paired measurements we can work out which changes in which transporters best account for the changes in small molecule uptake, including molecules not used in the learning phase, and can then test directly that they do indeed transport the molecules we claim. Because the transporter levels naturally vary between tissues they must naturally be controlled, and several substances (such as vitamin D) are known to affect these levels dramatically. This means that we can expect to be able to modulate the expression of transporters and different tissues by adding a second molecule (a so-called 'binary weapon'), and thereby 'force' particular substances to go only to those tissues. We shall test this by adding a library of small artificial (man-made) molecules (so-called 'fragments') and seeing which of them can do this (at least to some degree). Computer methods help us to find larger and more potent molecules that possess these same fragment signatures and that would therefore be predicted to have the desired targeting effects. Importantly, we shall curate all of the data in a web-accessible database.