Pollen from multiple sunflower cultivars and species reduces a common bumblebee gut pathogen

Abstract

Pathogens are one of the factors driving pollinator declines. Diet can play an important role in mediating pollinator health and resistance to pathogens. Sunflower pollen (Helianthus annuus) dramatically reduced a gut pathogen (Crithidia bombi) of Bombus impatiens previously, but the breadth of this effect was unknown. We tested whether pollen from nine H. annuus cultivars, four wild H. annuus populations, H. petiolarus, H. argophyllus and two Solidago spp., reduced Crithidia in B. impatiens compared to mixed wildflower pollen and buckwheat pollen (Fagopyrum esculentum) as controls. We also compared hand- and honeybee-collected pollen (which contains nectar) to assess whether diet effects on pathogens were due to pollen or nectar. All Helianthus and Solidago pollen reduced Crithidia by 20-40-fold compared to buckwheat pollen, and all but three taxa reduced Crithidia compared to wildflower pollen. We found no consistent differences between hand- and bee-collected pollen, suggesting that pollen alone can reduce Crithidia infection. Our results indicate an important role of pollen diet for bee health and potentially broad options within the Asteraceae for pollinator plantings to manage bee disease.

Keywords: Bombus impatiens; Crithidia bombi; bumblebees; goldenrod; pollinator decline; sunflower.

Figure 1.

Mean raw Crithidia count per 0.02 µl (±s.e.) for the 19 pollen taxa. Pollen treatments are: buckwheat (pink), wildflower mix (yellow), our positive control of H. annuus ‘China’ (orange), honeybee-collected taxa (grey), hand-collected taxa (blue) and Solidago spp. (green), which were honeybee-collected. Different letters associated with bars indicate statistically significant differences between pollen treatments after a post hoc Tukey's test. Full explanations for all taxa names are provided in electronic supplementary material, table S1; ‘HC’ refers to hand-collected for the three taxa where we had both honeybee and hand collection. Standard errors were calculated by back-transforming least-square means plus or minus least-square mean standard errors.

Figure 2.

Comparison of hand- versus honeybee-collected pollen for (a) H. annuus, ‘Black Oil Seed’, (b) H. annuus, Germany, (c) H. annuus, wild California and (d) comparison pooled across all Asteraceae taxa used in the experiment (17 treatments). Asterisks (*) denote statistically significant differences between collection method. Standard errors were calculated by back-transforming least-square means plus or minus least-square mean standard errors.