Nutrient enrichment is associated with altered nectar and pollen chemical composition in Succisa pratensis Moench and increased larval mortality of its pollinator Bombus terrestris L

Abstract

Pollinators are declining worldwide and possible underlying causes include disease, invasive pest species and large scale land use changes resulting in habitat loss and degradation. One particular cause of habitat degradation is the increased inflow of nutrients due to anthropogenic combustion processes and large scale application of agricultural fertilizers. This nutrient pollution has been shown to affect pollinators through the loss of nectar and pollen-providing plant species. However, it may also affect pollinators through altering the nectar and pollen chemical composition of plant species, hence influencing pollinator food quality. Here, we experimentally investigated the effect of nutrient enrichment on amino acid and sugar composition of nectar and pollen in the grassland plant Sucissa pratensis, and the subsequent colony size and larval mortality of the pollinating bumblebee Bombus terrestris. We found less of the essential amino acids glycine and arginine in the pollen of fertilized plants, and more arginine, ornithine and threonine in the pollen of control plants. Nectar glucose and pollen fructose levels were lower in fertilized plants as compared to control plants. Furthermore, bumblebee colonies visiting fertilized plants showed more dead larvae than colonies visiting control plants. Our results suggest that the fitness of bumblebees can be negatively affected by changes in their food quality following nutrient pollution. If similar patterns hold for other plant and pollinator species, this may have far reaching implications for the maintenance of pollination ecosystem services, as nutrient pollution continues to rise worldwide.

Fig 1. NMDS plots of amino acid composition of nectar and pollen produced by fertilized (black) and control (blank) plants.

Environmental fit (envfit function, 1000 permutations; vegan package, R) showed a significant difference between control and fertilized treatments of the amino acid composition of the nectar (R² = 0.067, P = 0.021) and of the pollen (R² = 0.32, P < 0.001).

Fig 2. Proportions of amino acids in the nectar and pollen produced by fertilized (black) and control (blank) plants.

Significance of the differences between the fertilization and control treatment were determined through Wilcoxon signed-rank tests using Bonferroni corrections to correct for multiple testing (*** P < 0.001; ** P < 0.01; * P < 0.05).

Fig 3. Proportion of sugars in the nectar and pollen of fertilized (black) and control (blank) plants.

Significance of the differences between the fertilization and control treatment were determined through Wilcoxon signed-rank tests using Bonferroni corrections to correct for multiple testing (** P < 0.01; * P < 0.05; (*) P < 0.09).

Fig 4. Larval mortality recorded each week of the experiment in colonies visiting control plants (blank) and fertilized plants (black).

Generalized linear modeling with repeated measures showed a significant effect of fertilization (F = 6.18, P = 0.025) and week (F = 15.92, P < 0.001). There was no significant interaction between week and fertilization treatment (F = 1.76, P = 0.13).

Fig 5. Proportion of living workers recorded each week of the experiment in colonies visiting control plants (blank) and fertilized plants (black).

The proportion was calculated as the ratio of living workers to the number of living workers in the first reference week (= 100%). Generalized linear modeling with repeated measures showed no significant effect of fertilization (F = 1.82, P = 0.20), a significant effect of week (F = 238.83, P < 0.001) and a significant interaction between week and fertilizan treatment (F = 2.50, P = 0.025).