Pollen diet for caterpillars

January 2010. It is below freezing outside, but in a greenhouse at RWTH Aachen University winter has been kept at bay and clouds of butterflies are fluttering about in a carefully controlled atmosphere in a large tent-like net. Mechthild Schuppener has succeeded in establishing a breeding programme for small tortoiseshell butterflies. She needs small caterpillars for feeding experiments that she is conducting in the laboratory. She is trying to find out whether genetically modified Bt maize is harmful to butterflies.

Small tortoiseshell butterflies that are being bred in the greenhouse

Mechthild Schuppener harvesting maize pollen in the greenhouse for the feeding experiments with caterpillars.

Young, freshly hatched caterpillars eating a stinging nettle leaf

Young larvae eating a leaf. They leave the leaf veins.

Older caterpillars

Caterpillars in the final larval stage before pupation are kept in closed cages.

A caterpillar hangs from the lid of a plastic container ready to pupate.

The caterpillars hang upside down (here from kitchen paper) …


….before pupating.

Feeding experiment: Small caterpillars are offered defined amounts of pollen on pieces of stinging nettle leaf

Feeding experiment: A small caterpillar in the 3rd larval stage eating a piece punched from a stinging nettle leaf to which Bt pollen solution has been applied.

A day later the researcher checks to see how much of the leaf has been eaten by the caterpillars.

The trial larvae are raised in the climate chamber and their further development is observed.

In summer, pollen traps were set up at various distances from the trial field. The pot next to each trap contains stinging nettle plants, a typical host plant for caterpillars.

The pollen traps contain a gel onto which the pollen grains stick.

Pollen trap: a Petri dish filled with agarose gel for the pollen grains to stick to.

The pollen grains are counted under a binocular microscope.

Pollen grains on a stinging nettle leaf. They collect along the leaf veins.

On stinging nettle leaves, the pollen collects along the leaf veins, which the caterpillars do not eat.

Roused by the intrusion, large numbers of butterflies flutter around the tent, while a few remain motionless. The older ones look a bit tatty – their wing colours have lost a lot of their brilliance. Two butterflies have landed on a piece of kitchen paper soaked in honey solution and are using their proboscis to feed. Hundreds of new little caterpillars are feeding on the leaves of the stinging nettle plants. There are no big caterpillars because, as Mechthild Schuppener explains, “The older caterpillars in the final larval stage have to be kept in closed cages because they move too fast and we would not be able to monitor them otherwise”. She points to a butterfly that is evidently getting ready to lay a batch of eggs on the underside of a leaf. Mechthild Schuppener is visibly proud of her success in breeding the butterflies. She first had to learn how to handle the insects, find out about their special requirements and test how they can be kept under artificial conditions.

The caterpillars that she needs for her experiments are kept in a climate chamber in Petri dishes or large plastic containers. Preliminary experiments found that 25°C is an optimum temperature, with a day-night rhythm of 16:8 hours. To prevent the insects drying out, a piece of kitchen paper sprayed with water is placed between the dish and the lid. The older caterpillars can also hang from this when they are ready to pupate and turn into a butterfly.

The one-day die

Today a feeding experiment with small caterpillars is starting in the laboratory. It is designed to provide information about the extent to which the insects are harmed when they eat Bt maize pollen on their food plants. The experiment tests whether certain doses kill the larvae or impair their development. The Bt maize under investigation produces a Bt protein that kills the larvae of the European corn borer, a type of moth. It is therefore possible that other types of butterfly and moth could be at risk.

The insects used in the trial are in the third larval stage (L3), i.e. they have already shed their skin twice and are now around one centimetre in length. “When the caterpillars are smaller than that it is difficult to handle them,” says Mechthild Schuppener, placing a rolled-up caterpillar onto a tiny weighing scale. In order to be able to monitor their development later on, she has to weigh them at the beginning.

A cork borer is used to stamp 1 cm2 pieces from stinging nettle leaves. One piece is then placed in each Petri dish. Different concentrations of a Bt maize pollen solution are then applied to the leaf pieces. “The larvae are given 100-2000 pollen grains per square centimetre,” says Mechthild Schuppener, carefully using a pipette to drip pollen solution onto a leaf. “The maximum concentration is designed to simulate an extreme dusting of pollen.”

The pollen used in the feeding experiments comes from maize plants grown in the greenhouse. As a control, some of the larvae are given pollen from the conventional maize variety that was used to create the Bt maize being tested, and some are given pollen from another conventional variety. Between 15 and 30 insects are tested with each concentration of each maize variety.

The young larvae are exposed to the pollen diet for a day. In most cases, the whole leaf has been eaten by the end of the day. However, Mechthild Schuppener has to get the timing just right: “When they are shedding their skins, the larvae don’t eat. These periods of not eating can last for up to a day, so I have to take this into account when carrying out the experiments.” Afterwards, the insects are fed ‘normally’ again and their further development is observed closely in the climate chamber: When do they next shed their skin? How long is it until they pupate? How many larvae die? After about a week, when they are in the fifth and final larval stage before pupation, the insects are weighed again. At this point they weigh several times their starting weight.

Counting pollen grains

Laboratory experiments on their own are not enough to find out whether Bt maize harms non-target moths and butterflies. Field experiments are therefore being conducted to find out what amounts of pollen butterflies are exposed to under natural conditions. Although they don’t feed on maize plants directly, they can eat the pollen if it is blown onto their host plants, e.g. stinging nettles.

To measure pollen deposits, researchers are setting up pollen traps in three successive years on each side of the maize trial field, both in the field margin and at distances of five, ten, 15, 20 and 30 metres from the field edge, and up to 58 metres from the edge that is in the direction of the prevailing wind. The pollen traps consist of Petri dishes filled with agarose gel to which the pollen grains stick. A pot of stinging nettle plants is placed next to each pollen trap.

In a very tight timeframe of around five days during the maize-flowering season, the 75 traps are changed every day and leaves are cut from the stinging nettles. The pollen grains are then immediately counted under a binocular microscope. This has to be done quickly, especially in the case of the nettle leaves, because the leaves dry out fast and the pollen grains do not stick properly to the leaf surface.

It can already be seen from the figures obtained during the first two years of the trial that, as expected, the amount of pollen falls rapidly the further away from the field a trap is placed, and is several times lower on the leeward side of the field than on the side in the direction of the prevailing wind. By far the greatest amount of pollen is shed in the morning, while at night only very small amounts land in the pollen traps. One interesting finding is that far fewer pollen grains are caught on the stinging nettle leaves than in the pollen traps and that they tend to collect along the leaf veins, which the caterpillars do not eat.