If you overuse antibiotics on your bees, your colonies may fail due to inadequate gut microbes.
Nowadays it seems like there’s a new paper published every week showing how a robust gut microbiome is critical to honey bee health. Last month, we highlighted a paper showing that gut microbes shape how workers smell, which determines whether they’re recognized as colony-mates or intruders. Interestingly, the ability of workers’ gut microbes to rapidly change when placed in a new colony may explain why beekeepers can easily swap frames among colonies — the workers pick up microbes from their new colony fairly quickly and begin to smell like their “new sisters.”
But what happens when you change microbes in honey bee guts by killing them with antibiotics, such as those used to treat European foulbrood (EFB) or colonies in the same apiary as a positive American foulbrood (AFB) detection? It’s obviously beneficial to kill these serious pathogenic microbes and prevent them from overwhelming hives. But can you do too much of a good thing and harm bees by killing the good microbes while combatting the bad ones? This is the topic for our thirty-seventh Notes from the Lab, where we summarize “Long-term effects of antibiotic treatments on honeybee colony fitness: A modelling approach,” written by Laura Bulson and colleagues and published in the Journal of Applied Ecology [2020,00:1-10].
For their study, Bulson and colleagues used a nice existing dataset in which replicate treatments of ~30 age-controlled workers in hoarding cages were fed 450 µg/ml oxytetracycline (the most common antibiotic used to combat EFB and AFB) in sugar syrup for 5 days. Survival of the bees was monitored daily for 10 days following exposure and compared to controls, to determine how typical oxytetracycline treatment impacted microbiome disturbance and worker mortality over a period of two weeks.
Next, the authors assessed how mortality of individual bees in the short-term experiments could scale up to impact the colony over longer time periods by conducting simulations using the BEEHAVE model (Becher et al., 2014; www.beehave-model.net). BEEHAVE is a mathematical representation of a honey bee colony based on empirical data and accepted theory, and is considered by the European Food Safety Authority (EFSA) as a tool in regulatory risk assessments (EFSA, 2015). The model can be used to assess how stressors are likely to impact honey bee colonies using various data inputs, such as mortality of individual bees over time in response to antibiotics.
Each simulation of the BEEHAVE model was run with 1,000 mortality values estimated from the existing dataset, resulting in 1,000 replicate simulated colonies per scenario. To increase realism, 10 deformed wing virus-infected and 10 uninfected varroa mites (20 mites total) were added to colonies of 10,000 bees at the start of each simulation, as is standard in EFSA risk assessments. Simulations were run for a period of 10 years to give both short- and long-term estimates of potential impacts of antibiotic use.
Finally, to assess multiple plausible scenarios regarding the duration of microbiome disturbance (because this information is currently unknown), a gradient of antibiotic-induced mortalities was assessed. A 30-day effect duration was considered the minimum scenario and assumes antibiotics only increase a bee’s probability of dying during the treatment period (three applications of powdered oxytetracycline 4-5 days apart in the spring and autumn, which some beekeepers are known to use in their hives). This treatment period is similar to the recommendation from U.S. veterinarians when EFB or AFB is found, which is three applications spaced one week apart.
The maximum scenario (365-day effect duration) assumes gut microbiome depletion following antibiotic exposure is irreversible and results in a permanently increased probability of mortality. While this scenario is unlikely, it is possible. For example, honey bees acquire their core gut microbiome through social transfer, which might be prevented if antibiotics disrupt the micr ….