Send

Effects of Bt maize on honeybees

(2008 – 2011) University of Würzburg, Dep. of Animal Ecol. & Trop. Biology (Zoology III)

Topic

The project investigated the potential effects of the genetically modified Bt maize cultivar MON89034xMON88017 on honeybees. This maize produces three Bt proteins, making it resistant to both the European corn borer and the Western corn rootworm.

The project examined the following questions:

  • Does ingestion of Bt maize pollen have undesirable effects on the development of bee colonies or on individual honeybee larvae?
  • Does ingestion of Bt proteins reduce the bees’ resistance to other stress factors, e.g. pathogens?
  • How much maize pollen is actually carried into beehives as a function of the structure of the surrounding agricultural landscape?

Experiment description

In order to differentiate between maize varietal effects and Bt effects, two other conventional maize varieties were planted as a control in addition to the genetically modified variety and the isogenic parent variety.

During the maize-flowering period two honeybee colonies are placed in each of the flight tents.

Honeybee hives were placed in flight tents with a base area of 4x12 metres and a height of 3 metres during the maize-flowering period.

Traps were set up in front of the artificial hives to catch dead bees, so that they could be counted.

The hives were opened to document the development of the colony.

The bee with the green marking is the queen

The bee with the green marking is the queen.

Bees in individual cages during the feeding trials.

Bee larva

Effects of Bt proteins on honeybees in a semi-field trial

Two beehives with approx. 1000 workers were placed in a flight cage on each of 32 plots on the trial field (eight plots with each of the four different maize variants) at the start of the maize-flowering period. As a control, nine free-flying bee colonies were placed on a field with Phacelia tenuifolia (scorpionweed, a plant that attracts bees).

A trap was used to count the number of dead bees each day. Changes to the size of the colony were documented throughout the trial period using a special method (Liebefelder estimation method). Various colony development parameters were recorded several times during the flowering period: the number of open and capped brood cells, the number of bees and a special brood index. In the second year of the trial, the amount of pollen brought in each day was also recorded by observing the forager bees, and the various flowering stages were documented to determine how much pollen was available.

Newly emerged worker bees raised in the flight tents were removed and taken to the laboratory to assess their weight on emergence and individual lifespans.

Examination of bee intestines

In collaboration with Prof. Christoph Tebbe’s team (vTI Braunschweig), the project investigated the digestibility of maize pollen, the breakdown of Bt proteins in the bee intestine and the influence of Bt maize on the composition of the gut microflora under semi-field conditions. To achieve this, newly emerged bees were marked and placed in the trial colonies. After nine days, they were removed, deep-frozen and later used for gut dissections in the laboratory. Measurements were taken of the amount and digestion rate of the maize pollen grains eaten and DNA sequences were used to investigate the bacterial communities in the different gut sections.

Feeding experiments with adult bees

Feeding with maize pollen: Bees were fed pollen from the various maize varieties in the field trial both individually and in groups of 30 bees. The feeding experiments were continued until all the bees were dead.

Feeding with pure Bt protein: Bees were fed with different concentrations and combinations of pure Bt proteins to test whether the three Bt proteins produced by the maize variety under investigation have undesirable effects on honeybees when combined.

Feeding experiments with bee larvae

In an in vitro larval test, young bee larvae were raised under standardised laboratory conditions on artificial food, to which pollen from the various maize trial variants and pure Bt protein had been added. In this experiment it is possible to measure exactly how much Bt protein the larvae eat. Once the artificially raised bees have hatched, they can be further examined for sublethal effects. Since the Bt proteins target insect larvae, the in vitro larval test starts at a potentially sensitive phase of honeybee development, which means the results can be particularly revealing.

Combined effects of Bt proteins and pathogens on honeybees

Colonies on the trial plots with Bt maize and the isogenic variety were infected with bee diseases: in 2008 with Nosema, a gut parasite, and in 2009 with Varroa mites. The development of the bee colonies was assessed and documented in the same way as described above. In addition, experiments were conducted in the laboratory to investigate the interaction of Bt pollen and Nosema.

Potential exposure risk of honeybees in relation to Bt maize

Twelve landscape areas were selected that differed from one another in terms of the proportion of land area under maize cultivation and the availability of alternative pollen sources during the maize-flowering period. Trial colonies with pollen traps were set up there at the start of the maize-flowering period. With the pollen traps it is possible to measure the pollen spectrum used by the bees each week. The aim was to draw conclusions about how much maize pollen is actually carried into beehives as a function of the structure of the surrounding agricultural landscape, and assess the potential exposure risk to bees in relation to Bt maize.

In order to analyse the foraging distance of honeybees for maize pollen, four observation hives were used and moved around all 12 locations.

Results

Effects of Bt proteins on honey bees in a semi-field trial

There was no difference in mortality between the bee colonies in the different maize variants. Nurse bees that have to eat large quantities of pollen to feed the brood exhibited no significant weight differences after nine days in the flight tent.

The newly hatched worker bees transferred to the laboratory did not exhibit any negative impacts of Bt maize pollen in terms of their weight on emergence or life expectancy.

Examination of bee intestines

No differences in the digestibility of the various maize varieties was found in nurse bees forced to collect only maize pollen during nine days in the flight tent.

No Bt-specific influence was found on the frequency and diversity of the gut bacteria.

Combined effects of Bt proteins and pathogens on honeybees

The scientific results have not yet been published.

Feeding experiments with adult bees

Feeding with maize pollen: No differences were found in the life expectancy of honeybee worker bees between the different pollen treatments.

Feeding with pure Bt protein: The maximum quantity of Bt protein that bees are exposed to when eating Bt maize pollen was calculated (amount of pollen eaten multiplied by the quantity of protein produced in the pollen). The bees were given the various Bt proteins in a sugar solution as well as a protein mix in four concentrations. This test also covered a worst-case scenario, in other words it tested concentrations that far exceeded the amount bees are exposed to in theory. The results show no dose–effect relationship, even for high doses, between the tested Bt proteins and bee mortality.

The survival of larvae fed on maize pollen
‘Bt’ feeding with Bt maize pollen (MON810 and MON89034xMON88017, n = 40),

‘C’ feeding with pollen from conventional maize (three different varieties, n = 59)

‘H’ feeding with pollen from an ornamental plant heliconia rostrata as a positive control; this plant is known to be toxic to bees (n = 10)

Feeding trials with bee larvae
The in vitro larval test is recommended as a standard method for the approval of pesticides and transgenic crop plants.

Feeding with maize pollen: Pollen from the various maize varieties on the trial field and pollen from the Bt maize variety MON810 were added directly to the brood food (L3 stage). The pollen in the food was shown to be eaten by the larvae. 100 per cent of the larvae that received Bt maize pollen survived. The survival rate of bee larvae fed on pollen from the conventional maize varieties was slightly lower. There was therefore no significant difference between the different maize variants. The weight of the larvae before pupation did not differ either.

Feeding with Bt proteins: The individual Bt proteins and a protein mix were added to the brood food in five different concentrations (L2 stage). Once again, a worst-case scenario was tested. No negative Bt effects were found on the survival rate or pre-pupal weight of the bee larvae.


Potential exposure risk of honeybees in relation to Bt maize

Over a period of 22 days, pollen collected by the bees was removed with the help of pollen traps and the maize pollen ratio was calculated for the surrounding landscape. The colony’s development was documented before and after the trial period. In order to quantify the amount of pollen eaten by nurse bees in the different landscapes, bee intestines were dissected. The results show that maize pollen plays a role in the honeybee pollen diet and therefore in the provision of protein. However the use of maize pollen is not dependent on the density of maize fields in the surrounding area.

662 bee dances were observed and used to calculate the coordinates of the pollen sources visited by the forager bees at all locations. The foraging distances for maize pollen were significantly shorter than those for other types of pollen. In other words, in the observation hives less effort was spent on foraging for maize pollen than for other types of pollen.