Bee species visiting Medicago sativa differ in pollen deposition curves with consequences for gene flow

Authors: SANTA-MARTINEZ, EMMANUEL - University Of Wisconsin; CASTRO, CIBELE - Federal Rural University Of Pernambuco; Flick, Andrew; Sullivan, Michael; Riday, Heathcliffe; CLAYTON, MURRAY - University Of Wisconsin; Brunet, Johanne

Submitted to: American Journal of Botany
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/25/2021
Publication Date: 6/10/2021
Publication URL: https://handle.nal.usda.gov/10113/7709311
Citation: Santa-Martinez, E., Castro, C.C., Flick, A.J., Sullivan, M.L., Riday, H., Clayton, M.K., Brunet, J. 2021. Bee species visiting Medicago sativa differ in pollen deposition curves with consequences for gene flow. American Journal of Botany. 108(6):1016-1028. https://doi.org/10.1002/ajb2.1683.
DOI: https://doi.org/10.1002/ajb2.1683

Interpretive Summary: Pollinators pick up pollen from a flower and deposit that pollen on flowers visited in succession during a foraging bout. The foraging bout represents all the flowers visited in between the time a pollinator enters a patch and the time it leaves the patch. When a bee moves from a patch with genetically modified (GM) pollen to a conventional patch, it carries GM pollen and deposits that GM pollen on the stigmas of the flowers visited in succession in the conventional patch. The pattern of pollen deposition, or the decrease in the number of pollen grains deposited on flowers visited in succession, will help determine how many flowers must be visited, on average, by a pollinator before the GM pollen grains carried by that bee have been exhausted. Only the pollen on the pollinator's body can be used for pollination, the pollen in the pollen sacs is brought back to the hive to feed the young and is not available to pollinate plants. A pollinator with a steeper pollen deposition curve will deposit all the pollen grains on the first few flowers visited and will, therefore, carry pollen grains shorter distances and have lower gene flow risk. Differences in the shape of the pollen deposition curves between bee species may help predict gene flow risk by distinct bee species. We compared the pattern of pollen deposition over successive flowers and over distances between two grooming pollinators, the common eastern bumble bee, Bombus impatiens, and the alfalfa leafcutting bee, Megachile rotundata foraging on alfalfa flowers. For both successive flowers or distance, leafcutting bees exhibited steeper pollen deposition curves and hence shorter tails than bumble bees. These results suggest that pollen carried from a GM patch to a conventional patch by leafcutting bees will be carried to fewer flowers in the conventional patch relative to the GM pollen carried by bumble bees. Therefore, we expect the gene flow risk to be less severe with leafcutting bees relative to bumble bees. This is the first report of a difference in pollen deposition curves among distinct bee species. This information will be useful to alfalfa farmers and the alfalfa industry concerned about isolation distances to maintain seed purity or isolation distances to facilitate the coexistence of different agricultural seed markets. This information will also be useful to regulators concerned about risk of transgene escape and to scientists interested in how the foraging behavior of a pollinator may affect gene flow risk.

Technical Abstract: The pattern of pollen deposition or how pollinators deposit pollen from a particular donor onto successive flowers visited can affect pollen dispersal. In this study, we compared the pattern of pollen deposition over successive flowers and over distances for two grooming pollinators, one social bee, Bombus impatiens, and one solitary bee, Megachile rotundata foraging on Medicago sativa flowers. Plants carrying three copies of the ß-glucuronidase (GUS) allele were used as pollen donors for this tetraploid species. One individual bee was allowed to visit the GUS plants before being released on a linear array of untransformed M. sativa plants. Plant, raceme and flower visited in succession were marked and distance traveled between successive flowers measured. Stigmas were collected, stained and the blue pollen grains (GUS pollen) on each stigma counted. Different mixed effect Poisson regression models, which allowed for different rates of decay in pollen deposition, were fitted to the pollen data for each bee species. For both bee species, the regular Poisson model best fitted the pattern of pollen decay over successive flowers. Pollen decay over distance exhibited a slower rate of decay. For both successive flowers or distance, leafcutting bees exhibited steeper curves with shorter tails relative to bumble bees. The Poisson model is equivalent to an exponential model of pollen decay. This finding challenges whether the long dispersal tails detected in earlier studies are true biological phenomena or mere artifacts of the experimental or statistical methods used in these studies. This is the first report of a difference in pollen dispersal curves between two bee species that are both grooming pollinators. Differences in pollen dispersal curves could be used in agriculture to help predict differences in risk of gene escape between distinct pollinators.