Sa-challenged bees were injected for a second time with ringer (Sa-R, n = 32), viable Pl (Sa- Pl, n = 27), viable Ec (Sa- Ec, n = 34), or viable Sa Saa- Sa, n = 38).
Ec-challenged bees were injected for a second time with ringer (Ec-R, n = 41), viable Pl (Ec- Pl, n = 28), viable Ec (Ec- Ec, n = 31), or viable Sa (Ec- Sa, n = 38).
Pl-challenged bees were injected for a second time with ringer (Pl-R, n = 71), viable Pl (Pl- Pl, n = 129), viable Ec (Pl- Ec, n = 26), or viable Sa (Pl- Sa, n = 47).
At day 8 after the challenge, bees were injected for a second time with either 2 μL of ringer saline solution, 2 μL of viable Pl bacteria (2 × 10 cells/mL), Ec (1 × 10 cells/mL), or Sa (1 × 10 cells/mL).
Final experimental groups were as follows: ringer-challenged bees were injected for a second time with ringer as a control (R-R, n = 83), viable Pl (R- Pl, n = 129), Ec (R- Ec, n = 51), or Sa (R- Sa, n = 47).
Obviously, this feedback is activated when bees were injected, irrespective whether they received bee ringer solution or E. coli.
To control for injection, another set of bees (n = 25) were injected with 5 μl of Ringer solution, a saline solution regularly used in insect physiology.
