| dc.description.abstract | Despite decades of research, many key aspects related to the physiological
processes and mechanisms insect herbivores use to build themselves remain poorly
understood, and we especially know very little about how interactions among nutrients
and allelochemicals drive insect herbivore growth processes. Understanding the
physiological effects of these interactions on generalist herbivores is a critical step to a
better understanding and evaluation of the different hypothesis that have been emitted
regarding the benefits of polyphagy. I used both lab and field experiments to disentangle
the respective effect of protein, carbohydrates and allelochemicals on a generalist
herbivore, the grasshopper Melanoplus differentialis.
The effect of protein and carbohydrates alone were examined using artificial diets
in choice and no-choice experiments. Results were plotted using a fitness landscape
approach to evaluate how protein-carbohydrate ratio and/or concentration affected
performance and consumption. Growth was best near the self-selected ratio obtained
from the choice experiment, most likely due to the fact that the amount of food digested
was also higher on that ratio. By contrast, development time was not best near the
preferred ratio most likely due to the trade-off existing between size and development
time. These results illustrate how nutrient availability can shape an ecological trade-off:
growing big or growing fast.
When an allelochemical (gramine, an alkaloid commonly found in grass) was
introduced to the artificial diets, it had an interactive effect with protein and
carbohydrates on performance and consumption and performance were generally
improved on diets that contained higher amount of protein.
Host plants of two wild populations were determined by gut content analysis and
fed in choice and no-choice experiment. First the plant material was dried and ground
and its protein, carbohydrate, terpenoid and phenolic content analyzed. Both populations
regulated for the same protein-carbohydrate intake. However performance was different,
due to variation in plant allelochemical content. This demonstrated for the first time that
nutrient regulation, not toxin dilution, is directing food selection behavior in a generalist
herbivore.
Finally the role of macronutrients was analyzed in a context of cost of
detoxification by measuring microsomal p450 production in the presence/absence of
gramine. In the presence of choice, nutrient regulation was altered when gramine was
present in the protein-biased diet. In the absence of choice, insects performed better on
carbohydrate biased diet. I found that gramine elevated the level of microsomal protein
in the fat body. | en |
