Science - September 29, 2011

Poisoning bees with pesticide

It is not obvious that there is anything untoward going on. The bees at Droevendaal experimental farm are busily buzzing around. But some of them have been fed for the last three months with the pesticide imidacloprid.

In fact, a small number of bees are not only getting regular doses of imidacloprid but are also being deliberately weakened. A ‘pollen trap' in the hive ensures that a fair proportion of the pollen that is brought in is snatched away from the bees. Not that researcher Sjef van der Steen is so keen to make the bees' lives difficult. On the contrary, he is very fond of bees.
This summer the forty populations at Droevendaal have been the subject of an experiment with the controversial pesticide imidacloprid (see text box). Van der Steen is looking at whether and how the bee populations suffer if exposed to this pesticide and/or to pollination differences. The emphasis is on bee populations, says Van der Steen. ‘We are the first to do this sort of research at population level. You can study an individual bee, but the results cannot be transferred one-to-one to a whole population. A population might react quite differently from an individual bee. You should regard a bee population as a super-organism.
Incredibly poisonous stuff
The use of imidacloprid on agriculture and horticulture is coming under heavy fire. ‘Imidacloprid is often mentioned as a pesticide that could play a role in bee mortality', confirms Van der Steen. ‘Its use has been permitted in this country for about ten years. Recently a strong lobby has grown up for taking the substance off the market, because of its possible role in the bee deaths. But bee mortality is caused by a number of factors. As far as we know at present, the main one is the Varroa mite. But environmental factors such as the availability of pollen or the presence of pesticides play a role too.'
Van der Steen readily acknowledges that imidacloprid is ‘incredibly poisonous' stuff. It is used as a seed coating in certain crops, and via this route it ends up in nectar and pollen. Crop spraying also causes it to be washed into surface water. But is it a factor in the bee deaths as well? Van der Steen: ‘How dangerous a pesticide is for a bee population depends not only on how poisonous it is but also on the level of exposure. Although imidacloprid is poisonous for bees, the exposure level is low. But it certainly cannot be ruled out that problems could arise if the exposure time was different for some reason. It is possible, but we don't know. That is why we look at both best-case and worst-case scenarios.'
Worst case
Vitality is the key concept in Van der Steen's research. How vital does a bee population remain under long-term exposure to imidacloprid? And is that exposure worse for vital populations than it is for populations that have been put on short rations and are therefore less vital? Van der Steen is interested here in changes in bee numbers during the season, their reproduction rates, their stocks of pollen food known as bee bread, and the levels of vitellogenin in the blood. The latter is the bee's storage protein, explains Van der Steen. ‘The young bee converts protein from pollen into reserve protein. A significant amount of this protein is vitellogenin, also known as yolk protein. The queen, the larvae and the young bees (1 to 4 days old) are fed on it.'
Some of the bees are having a particularly hard time: those which are on both imidacloprid and short rations. The pesticide is given by feeding the bees twice a week on sugar water with a sub-lethal dose of imidacloprid in it. Van der Steen: ‘That is the worst-case scenario. In it we simulate the situation in which bees only feed off plants infected with imidacloprid and also cannot collect much pollen.' At the other end of the spectrum are the lucky bees which get to keep all their pollen and are not dosed with any poison at all. The bees were subjected to a thorough study three times during the course of this summer. The last examination has just been completed. Van der Steen cannot say anything about the results yet as the analysis is still in full swing. The test will certainly not do anything to silence the discussion about the role imidacloprid. If a tentative indication of that role emerges, it would be a great result. Van der Steen: ‘Follow-up research will have to show exactly what the relation is between exposure to imidacloprid and the vitality of bee populations.'