Surprising results from a pesticide and temperature stress field trial
Everybody wants to lock eyes on a big, healthy queen when inspecting a hive — the kind that make you exclaim in admiration — but that isn’t always what we find. Some reports estimate that over 50% of queens are replaced within their first six months in a hive, and having a “poor queen” is a commonly cited reason for colony losses in Canadian and U.S. surveys.1,2 Given that colonies experiencing queen loss or supersedure are three times more likely to die in the next fifty days,3 queen quality can have direct consequences for colony survival.
But what makes a poor-quality queen? Last spring and summer, when travel was restricted and the laboratory was shut down because of the pandemic, I was spending almost all of my time in the field. I wanted to know what pesticide exposure and temperature stress did to queen performance in colonies. I had done a lot of research in the laboratory, but I did not know how these stressors impacted queens in the field.
Luckily, the package bees I ordered for this experiment — which were coming from Tasmania, Australia — squeaked into the country on one of the few shipments that arrived before the planes were grounded in March 2020. I actually wasn’t around to install the packages myself: A colleague did it for me because I was on a plane to Chicago to give a keynote talk for the Bee Prepared conference in Joliet, Illinois. When I got off the plane, the conference had been cancelled, and the world was a different place. Back to Canada I went, happy to have socially-isolated field work to keep me busy.
Now, the results from those field trials are out. In two papers published back-to-back in Scientific Reports and PLoS One, my colleagues (from North Carolina State University and the University of British Columbia) and I describe the surprising resilience of queens. In the first article, we tested how contact exposure to different pesticides commonly found in wax affected queen performance and quality metrics (mass and sperm viability).4 In the second paper, I looked at the same effects of heat stress and cold stress.5 While we did find some negative effects of temperature stress, in both papers the results were overwhelmingly boring; that is, these stressors did not have nearly as big of an impact as we were expecting.
We were interested in testing pesticide exposure because Dr. Kirsten Traynor — the former editor of this magazine — published a landmark paper in 2016 which showed that increased levels of in-hive pesticides were associated with queen loss and supersedure.6 Dr. Joe Milone, a then-doctoral student at North Carolina State University and collaborator on this project, used the chemical residue data that Traynor published to recapitulate a cocktail of compounds at the same relative concentrations as found in wax. The queen is in constant contact with wax, yet contact pesticide exposure is a topic seldom investigated for queens. Since absorption rates from wax to the bee are not known, it is hard to know what doses are relevant.
As you might expect, miticides are some of the most abundant compounds in wax. Fluvalinate, coumaphos, and an amitraz degradation product accounted for the biggest proportion of the cocktail components, despite discouragement of using the former two compounds due to widespread mite resistance problems. We should not assume that these compounds are benign just because they are approved for use in beehives — queens exposed to coumaphos during development, for example, are more likely to die, and those that survive are smaller than control queens.7 And fluvalinate concentrations in wax were so high, it was actually the most hazardous residue. Other fungicides, herbicides, and insecticides (chlorothalonil, chloropyrifos, fenpropathrin, pendimethalin, atrazine, and azoxystrobin) were included in the mixture, too, using Traynor’s data as a guide.
Back at the NCSU laboratory, Dr. Brad Metz and Erin McDermott exposed queens to different concentrations of each of these chemical components as well as a complete cocktail, and after two days, they measured the quality metrics that are standard for their queen clinic protocol (including queen mass, sperm viability, and sperm counts). The pesticide concentrations varied from 1 to 32 times the median concentration found in wax, and although the dose was very small (a two microliter drop on the thorax) we had previously observed changes in protein expression resulting from even lower doses.
Applying pesticides to the thorax may not seem like the most relevant part of the cuticle to target; after all, it would mainly be the queen’s feet (tarsi) and abdomen that contact the wax. But thorax exposure is a standard method in honey bee toxicology research, and is widely used to conduct topical toxicity evaluations in workers. When applied with a solvent, the pesticides can be absorbed through the cuticle and transferred to the rest of the body via the bee’s open circulatory system. As beekeepers, we commonly apply paint or glue to a queen’s thorax with no ill effects; apparently, these are not hazardous substances, but I have yet to find a study specifically investigating that question.
Metz and McDermott found no differences between control queens and queens exposed to the residues alone or in combination. But that was a short-term test, and I was itching to see what would happen if the exposed queens were introduced to colonies in the field. So, I set up my field trial and got to work.
First, I measured baseline queen masses and egg laying patterns, calculated as the percent of cells laid relative to the number of cells actually available, to account for pre-existing differences between individual queens. Then, to keep me blind to the treatments, a colleague of mine exposed the queens (10 in each group) to either the pesticide cocktail, the solvent alone, or no manipulation. I returned the queens to their hives, waited a couple of weeks, and measured laying patterns and queen mass again. I also looked at sperm viability (to test for sperm death a longer time after exposure), queen proteins that I previously proposed as diagnostic markers for pesticide stress (see “Queen Forensics,” July 2019 issue of ABJ), as well as the protein composition of the queen’s eggs and emergence mass of workers (to test if there were vertical effects of stress on the queen’s progeny). Across the board, I found no differences between the groups — not even in the proposed pesticide stress biomarkers.
So, if direct pesticide exposure is not causing a decline in queen quality, why does pesticide residue data correlate with queen failure and supersedure? Some of Milone’s other work, which was also published this year, shows that exposed workers have altered royal jelly secretions8 and are less able to produce viable queens.9 Therefore, our best guess is that these pesticides mainly impact queen health indirectly by impairing the workers’ ability to care for their queen, rather than through direct harm to the queen herself.
But I wasn’t done experimenting. By this time, it was the middle of June and I still had a whole summer of pandemic lockdown to fill, so I started another trial to investigate the impacts of temperature stress. We and others have investigated the perils of temperature stress extensively in the laboratory, and it is well-established that hot and cold temperatures reduce the queen’s sperm viability (see “How to keep queens in the Goldilocks zone, even during shipping,” April 2021 issue of ABJ), but I hadn’t yet tested how that translates to performance in colonies.
Using a new batch of queens (supplied locally, to avoid potential double-stress events that could occur during long-distance transport), and again remaining blind to the groups, I investigated the impact of ….