Bees

Newly emerged bees were obtained from a healthy-looking colony of Apis mellifera carnica located at the University of Lausanne. In short, dark-eyed pupae were carefully removed from capped brood cells with sterile tweezers and transferred to sterilized plastic boxes as described previously [36 (link)]. Boxes with pupae were kept with a source of sterile sugar water (50% sucrose solution, w/v) at 35°C with 80%–90% humidity for 2 d until the bees emerged, followed by a reduction in temperature to 32°C. For each box, one or two newly emerged bees were dissected, and their homogenized hindguts (in 1 ml 1x PBS) cultured on growth media as described below. To minimize the chance of including contaminated bees in colonization experiments, we excluded cages for which bacterial growth was observed for the tested bees. For the colonization of newly emerged bees, bacterial strains were inoculated from glycerol stocks and restreaked twice. Details on bacterial strains and culture conditions can be found in S1 Table. Bacterial cells were harvested and resuspended in 1x PBS/sugar water (1:1, v/v) at an OD₆₀₀ of 1. For colonization, bacterial suspensions were added to a source of sterilized pollen and provided to the newly emerged bees (for details, see S5 Text). MD bees were kept under the same conditions, with the same food sources, but without being exposed to bacteria. The mid- and hindgut (S9 Fig) of gnotobiotic bees were dissected at day 10 post colonization and stored at −80°C until further use.
To obtain age-controlled hive bees, several brood frames without adult bees were transferred from the hive to a ventilated Styrofoam box that was kept in an incubator at 32–34°C overnight. The next morning, the newly emerged bees were collected, marked on the thorax with a pen, and reintroduced into the hive. These bees were recollected 10 d later, and their mid- and hindguts were dissected and stored at −80°C until further use.
The colonization experiment was repeated at two different time points of the year (spring and fall, referred to as experiment 1 and experiment 2 in this study). Whenever possible, we included bees from both experiments in our analysis, such as for CL and MD bees. However, this was not possible for all mono-colonizations because of bacterial contaminations (as detected by qPCR) or in a few cases because of the presence of above-threshold viral loads. The precise numbers of bees included per condition are listed in S5 Table.

Worker larvae were obtained from the honey bee (Apis mellifera ligustica) colonies at the Honey Bee Research Facility, School of Life Sciences, Arizona State University, Mesa, Arizona. Queens were confined to a fully drawn comb in an excluder cage (46 × 24 × 6 cm) as described by Peng et al. (1992 ). On the fifth day, the bees were shaken off the comb and the comb was brought into the grafting room to obtain 1.5–2 day old larvae.
Seven different larval diets were prepared by changing the sugar and water concentrations (Table 1). Sugars and the yeast extract were dissolved in distilled water and freshly thawed commercial royal jelly purchased from a local bee supply company was added to the mixture and mixed thoroughly on a shaker. The diets were divided into 2 ml centrifuge tubes and kept at -18° C in a freezer until they were used. The diets were thawed and brought to 34° C in a water bath just before feeding.
A total of 350 larvae were grafted; there were 7 treatment groups, 5 replicates, and 10 larvae in each replicate. The first day 5 aliquots of 200 mg food were placed in a polyethylene Petri dish (100 × 15 mm) and 10 larvae were grafted on each aliquot (Figure 1A). The Petri dishes were placed into a polyethylene tub (20 cm × 40 cm) containing 16% sulfuric acid, transferred into a humidity chamber, and kept there at 34° C and 90% RH. On the second day, 40 mg and the 3^rd day 80 mg of larval food/larvae were placed in new Petri dishes and the larvae were gently placed on top of the fresh food (Figure 1B). On the 4^th day, 120 mg and on the 5th day 180 mg of food/larvae was placed in Petri dishes and the larvae were transferred onto the food. On the 6th day larvae consumed most of the food, and they started depositing uric acid crystals on the dorsal side of the body. When uric acid crystals were observed, the larvae were removed from the feeding dishes, weighed, and transferred to a 100 × 15 mm Petri dishes lined with Kimwipes® tissue paper (Figure 1C).
The next day the old Kimwipes® tissue paper containing feces was removed and the larvae were gently transferred onto a clean tissue paper and kept at 34° C and 70% RH in the humidity chamber. At the end of the defecation stage larvae started spinning cocoons, and this was recorded as the spinning stage (Figure 1D). Dead or undeveloped pupae were removed from the Petri dishes, and pupae (Figure 1E) were kept in the humidity chamber until they completed development and became adults (Figure 1F). Bees were removed from the Petri dishes as soon as they become adults, weighed, inspected under a stereo microscope, and dissected to count the ovarioles.
The adult bees were classified as queen phenotypes if they completed the development in 15–16 days, had notches on the mandibles, curved stings, large spermathecae (1mm in diameter), and had no corbiculae; as worker phenotypes if they completed the development in 21–22 days, had rows of corbicular hairs, straight stings with barbs, and had mandibles without the notches; and as intercastes if they completed development between 17–20 days, had small notches on the mandibles, and/or undeveloped corbiculae.
Hive-reared A. mellifera served as controls. The brood comb from which larvae were grafted was removed from the colony 20 days after caging the queen and placed in an incubator. The next day newly-emerged bees were sampled, weighed, and dissected for ovariole counts.

A single healthy and well-established honey bee colony, headed by a Carniolan Apis mellifera carnica queen, was used as a source for all worker samples of this study. Using sister bees originating from a single colony helps minimize variability in hive conditions and the genetic make-up of the workers. A total of eight capped worker brood frames ready to hatch were removed from this hive and placed in an incubator at 35 °C with 50–60% relative humidity. The following day, several thousand one-day-old sister bees were collected into a sterile plastic box for further use.

Hazard quotients (HQ) in wax were determined as the sum of all acaricide residues detected in wax (µg/kg) divided by their respective contact LD₅₀ (µg/bee) in each beeswax sample (OECD, 2018 ). LD₅₀ values were taken from Sánchez-Bayo and Goka (2014 ) and PPDB/VSDB (Pesticide Properties DataBase /Veterinary Substances DataBase ). The risk to honeybees and brood was evaluated by comparing the LD₅₀ with the residue levels for each acaricide found in bees and brood. Contamination of honey was assessed by comparing the acaricide levels in honey with their MRLs.