Effects of Imidacloprid Soil Drench Applications on Nesting Blue Orchard Mason Bees (Osmia lignaria)

Objective 1

Eleven hundred Osmia lignaria (eastern subspecies) adult cocoons were obtained from a commercial source (Mason Bee Company, Deweyville, UT) in February 2019 and have been kept in cold storage at 4^◦ Celsius until ready for emergence.  Three hundred cocoons were emerged in the laboratory between March 11 and March 14, which included 120 females and 180 males.  The females were individually marked and treated with contact exposure to soil for one hour in petri dishes at four different concentrations [< 1, 50, 390 and 780 parts per billion (ppb)], then released into six 6 x 6 x 2 m flight cages each enclosing 5 to 6 flowering rabbiteye blueberry shrubs at Durham horticulture research farm in Watkinsville, Georgia (Figure 1A).  Each tent contained a Styrofoam nest block with 36 nesting holes filled with paper tubes (Figure 1B,C).  Five females at each treatment level were released in each tent, for a total of 20 females per tent, and 30 untreated males were released per tent.  Males were not treated to isolate the effects to females which contact the soil regularly for nest building.  Bees were released in the flight cages on March 20, 2019 and behavior is being observed daily for 20 minutes per cage to ascertain effects on nesting behavior. 

Cell production rate was measured by tracking dates of female nest initiation to completion.  Nests were monitored daily to look for signs of females initiating nesting activity, including entering nests with pollen in the scopa and/or bringing mud to nest holes.  A total of 20 hrs of formal observation was conducted over 14 days during the peak of blueberry bloom with each tent receiving 3 hrs and 20 mins of observation in 11 separate 20 min observation periods. 

When nesting was completed, nest tubes were removed for examination.  Reproductive output was measured by total number of cells produced per nest, numbers of pre-pupal cocoons per nest, weight of pre-pupal cocoons, and gender ratio of F1 generation adults. An otoscope was used to determine number of cells in each nest tube.  Number of cells in combination with the amount of time from nest initiation to nest completion was used to assess nest cell production rate.  During 2-5 June 2019, pre-pupal cocoons were removed from the nest tubes, number of pre-pupal cocoons per nest were counted, and cocoons were individually weighed to the nearest 0.0001 gm with a microbalance scale.  F1 generation adults were cut from their cocoons in mid-April 2020 and sexed to determine gender ratio.

Another group of 180 adult male and female O. lignaria were exposed to treated soil in petri dishes at the same four concentrations (< 1, 50, 390, and 780 ppb), with three with replicates of each concentration at 20% soil moisture and three with replicates of each concentration at 40% soil moisture.  Treated bees were released into small insect cages (60 x 30 x 30 cm) (Figure 1D).  Seven females and eight males were released each cage, and mortality was being assessed daily for three days.  

Objective 2:

To determine effects on larval development, we utilized specialized well plates which were developed for conducting experiments on grafted Osmia embryos in laboratory settings (Boyle and Pitts-Singer 2018).  The well plates were printed on a MakerBot 3-D printer with PLA filament (resin based), fused deposition modeling, 0.3 mm filament thickness (each layer).  The well plates contain 24 wells measuring 8 mm in diameter and 1.5 cm depth (Fig 2A).  Each well was filled with 0.25 g of damp soil placed in the bottom of the well, using the same soil treatment levels (0, 50, 390, and 780 ppb).  Nine well plates were utilized for the experiment, each well plate containing six replicates of each treatment level, for a total of 54 replicates of each treatment level.  Recently capped reed nests containing embryonic O. lignaria were obtained from commercial distributors in Florida and Washington state.  Reeds were mailed by the distributors in batches as soon as they were capped by females in the field (capping the end of the reed with mud is an indication the female has completed the nest).  Immediately upon receipt, reeds were opened (Fig 2B), and individual larvae were randomly selected and removed along with their native pollen provision.  The pollen provisions were weighed, and both the pollen provision and larvae were grafted into the wells containing 0.25g of treated soil.  Pollen provisions from larvae not selected for the experiment were analyzed for background levels of imidacloprid contamination by liquid chromatography.  No background levels were detected in any of the samples. 

Once grafting was complete, well plates were placed in an incubator set to 27^◦C.  Well plates were observed daily for progress, and dates of transition to 2^nd instar, 5^th instar, start of silk spinning, and completed cocoon were logged for each individual.  Signs of life for each larvae were checked daily, including movement, eating, observation of pollen moving through guts, or movement of hemolymph under skin.  If no signs of life noted and common signs of death appeared (change in color, darkening, dull appearance and no response to stimulus), then the larvae was marked as dead.  Completed cocoons were removed from well plate after three weeks and weighed in the pre-pupal stage.

Objective 3:

Two laboratory choice experiments were designed to examine whether females avoid or select soils containing imidacloprid residue.  For the first experiment, mated and fed four-day-old females were placed in choice arenas in the following soil treatments:  choice of 0 and 50, 0 and 390, 0 and 780, 50 and 390, 50 and 780, 390 and 780 ppb, with a full spectrum light source over the choice arena.  Eighteen females per choice treatment were be observed in choice chambers for five minutes. Number of entries into each chamber and time spent in each choice chamber were logged.

For the second experiment, female and male O. lignaria were emerged in small test cages, allowed to mate and feed on a sugar/water/pollen mixture for five days.  At the end of the fifth day post-emergence, a well plate containing pollen provisions was added to the test cages (Figure 3A).  Well plates utilized for this experiment were developed for conducting experiments on grafted Osmia embryos in laboratory settings (Boyle and Pitts-Singer 2018).  Honeybee pollen from a local beekeeper in Athens, Georgia was crushed and mixed with a 50:50 sugar and water solution to make pollen provisions.  Mated 5-day-old females (5 per test cage) were given a choice of two soil treatments.  One soil treatment contained imidacloprid concentration of either 100 (N = 6) or 1,000 (N = 6) ppb, the other was uncontaminated (0 ppb). Each soil treatment initially weighed 25g (Figure 3B).  Paper nest tubes were provided in the test cages.  Females laid eggs on the pollen provisions within the wells and utilized soil to cap the wells (Figure 3C), some females also utilized soil to cap ends of paper tubes (Figure 3D), although tubes were otherwise empty.  After a period of two weeks of female activity, the two treatments of soil were air dried overnight and weighed.  The dry weight of the remaining treated and untreated soil in each replicate was used to determine which soil was more heavily selected by the females.