Flowering plants have a range of defensive mechanisms to protect against herbivory, including secondary compounds. There is widespread evidence that secondary compounds not only occur in foliar tissue but also floral tissue, including floral rewards. Thus, pollinators such as bees may be exposed to secondary compounds in their consumption of nectar and pollen. Studies have shown that consumption of these compounds can both benefit bees (e.g., by enhancing memory and reducing parasite infection) or harm bees (e.g., by reducing reproduction and in some cases increasing mortality). Given widespread concerns about bee poisonings and population declines, documenting the effects of secondary compounds on bees has important implications for bee health. Locoweeds, Astragalus and Oxytropis species that contain the alkaloid swainsonine and poison livestock, have been implicated in honey bee losses, which is curious considering these species often do not produce nectar. Here we measured concentrations of swainsonine in pollen (the principal reward of many locoweeds) and tested the degree to which swainsonine affected the survival and/or reproduction of its principal pollinators, including honey bees and bumble bees. We predicted that swainsonine would have dose-dependent effects on the survival and reproduction of honey and bumble bees.
We found swainsonine in the pollen of locoweed (Astragalus lentiginosus) at concentrations of at least 0.17%. Chronic consumption of swainsonine at this concentration can cause a variety of neurological disorders in livestock, including death. In no-choice feeding trials, we found dose-dependent effects of swainsonine on honey and bumble bees. We found no evidence for immediate bee mortality following consumption of swainsonine. Honey bees were deterred from feeding on the highest concentrations of swainsonine, but swainsonine had no significant effect on honey bee mortality over 10 days. In comparison, bumble bees showed no deterrence at any swainsonine concentration, but suffered mortality at intermediate swainsonine concentrations. Despite these mortality effects, for surviving bumble bees we found no sublethal effects of swainsonine on estimates of (micro)colony reproduction. These results suggest that swainsonine has dose-dependent and species-specific effects on diet consumption and bee survival over the time-periods of our experiments. However, documented bee poisonings and death immediately following visitation to locoweeds is likely not attributable to swainsonine.