News Release

Bumblebees prefer virus-infected plants, and their buzz pollination increases seed yields

Peer-Reviewed Publication


Virus Payback

image: A bumblebee buzz-pollinates a tomato flower. <P> A foraging bumblebee (<i>Bombus terrestris</i>) extracts pollen from a tomato (<i>Solanum lycopersicon</i>) flower by gripping the cone-shaped structure at the center of the flower and vibrating its flight muscles at high frequency -- a process called buzz-pollination. view more 

Credit: Alex M. Murphy, Sanjie Jiang and John P. Carr

Plants produce volatiles, air-borne organic chemical compounds, to attract pollinators and seed dispersers, and to repulse plant-eating animals and microbes. Humans have used them for thousands of years as perfumes and spices. A study published on August 11th in PLOS Pathogens reports that virus infection can change a plant's volatile profile to increase reproductive success of the infected--virus-susceptible--host, and might so counteract the selective pressure that favors evolution of plant resistance.

A research team led by John Carr, Beverley Glover, and Nik Cunniffe, all from the University of Cambridge, UK, examined the effects of infection by cucumber mosaic virus (CMV), on tomato and Arabidopsis plants and bumblebees. Both Arabidopsis and tomato are self-pollinating--they do not depend absolutely on pollinating insects or birds for fruit and seed production. But in tomato, an additional process called buzz pollination takes place. Buzz pollination involves vibration of the flowers by insects such as bumblebees to increases pollen release. This in turn enhances fertilization rates and increases seed and fruit production.

Infection by CMV, a major pathogen of tomato which can also infect Arabidopsis, the researchers found, induced changes in the composition of volatiles emitted by plants of both species. The researchers grew plants in individual containers, and collected air with emissions from CMV-infected and mock-infected control plants. Organic compounds were collected on filters and analyzed by mass spectrometry.

Not only could the researchers see the changes, but the bumblebees could smell them. For tomatoes, the bumblebees showed a preference for the infected plants, even when these were covered up to eliminate visual cues, or when the plants had no flowers, suggesting that the leaves of infected plants were at least partially responsible for emission of the attractive volatile mix.

Arabidopsis is not a natural target for bumblebees, and the researchers observed no preference in visitation time and frequency between infected and uninfected plants. That said, the bees could smell the difference, because they could be trained to selectively avoid either infected or un-infected Arabidopsis plants. For the training, the researchers provided sugar water on infected plants and a (bitter) quinine solution on uninfected ones. The bees can taste but not smell the difference between the two liquids but, for example, once they had learned to associate the 'infected smell' with sweet taste and the uninfected one with bitter taste, they preferentially visited the infected plants.

Being able to train bees in this way, the researchers could then use mutant virus or mutant Arabidopsis strains to test whether specific genes contributed to the changes in volatiles. Either Arabidopsis mutants with defects in gene silencing or CMV mutants that lack a virus protein that blocks silencing in the plants abolished the ability of the bees to learn to discriminate between these infected and uninfected plants. This suggests that gene silencing plays a role in inducing the changes in volatile production or emission.

CMV-infected tomato plants are smaller but have similar number of flowers compared with uninfected ones. Without buzz pollination, infected plants also have dramatically decreased seed production, with less than 10% of the yield in uninfected plants. Using an electric toothbrush to artificially buzz pollinate the flowers of CMV-infected plants, however, the researchers were able to substantially rescue seed production, with seed numbers reaching approximately half the level seen for uninfected non-buzzed flowers.

As with the artificial buzz pollination, when bumblebees buzzed tomato flowers, seed yield was increased. Bee-pollinated flowers on CMV-infected plants yielded tomatoes that contained more seeds than the tomatoes that developed from bee-pollinated flowers on uninfected plants. These results imply that the greater buzzing activity on flowers of CMV-infected plants resulted in increased seed production.

Given the results in domesticated tomato plants under controlled conditions, the researchers wondered about implications for wild buzz-pollinated plants growing under natural conditions. They developed a mathematical model to address this question and report that pollinator preference for infected susceptible plants (with the resulting increases in offspring) could indeed outweigh underlying strong selection pressures favoring pathogen resistance. The findings were consistent over a wide range of conditions in the wild, suggesting that the interaction could allow genes for disease susceptibility to persist in plant populations.

"Under natural conditions", the researchers propose, "helping host reproduction by encouraging bee visitation might represent a 'payback' by the virus to susceptible hosts". In general, they hope their results "may be useful in developing strategies to increase pollinator services for crops under conditions of cultivation, as well as for a better understanding of the interplay of plant pathogens, wild plants and pollinators under natural conditions".


In your coverage please use this URL to provide access to the freely available article in PLOS Pathogens:

Please contact if you would like more information.

Funding: Major funding for this project was provided to JPC by the Leverhulme Trust (Grant numbers RPG-2012-667 and F/09741/F: Additional funding to JPC and studentships to support JHW and SCG came from the Biotechnological and Biological Sciences Research Council (Grant number BB/J011762/1: Other additional funding was obtained from the Isaac Newton Trust ( grant number 12.07/I to AMM). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing Interests: The authors have declared that no competing interests exist.

Citation: Groen SC, Jiang S, Murphy AM, Cunniffe NJ, Westwood JH, Davey MP, et al. (2016) Virus Infection of Plants Alters Pollinator Preference: A Payback for Susceptible Hosts? PLoS Pathog 12(8): e1005790. doi:10.1371/journal.ppat.1005790

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.