Reproductive Senescence in Drones of the Honey Bee (Apis mellifera)

Abstract

In the face of high proportions of yearly colony losses, queen health and fecundity has been a major focus of industry and research. Much of the reproductive quality of the queen, though, is a function of the mating success and quality of the drones (males). Many environmental factors can negatively impact drone semen quality, but little is known about factors that impact the drones' ability to successfully mate and deliver that semen, or how widely drones vary. In our study, we observed the daily variation in honey bee drone reproductive quality over time, along with a number of morphological traits. Drones were reared in cages in bank colonies, and 20 individuals were dissected and measured daily. The number of viable spermatozoa in the seminal vesicles was zero at emergence and reached an average maximum of 7.39 ± 0.19 million around 20 days of life. Decline in spermatozoa count occurred after day 30, though viability was constant throughout life, when controlling for count. Older drones had smaller wet weights, head widths, and wing lengths. We predict that this is likely due to sampling bias due to a differential lifespan among larger, more reproductively developed drones. Our study shows that drones are more highly variable than previously suggested and that they have a significant variation in reproductive physiology as a function of age.

Keywords: Apis mellifera; aging; drone; honey bee; reproduction; senescence; sperm viability; spermatozoa.

Figure 1

(a) measurement of drone heads (H) was perpendicular to the body axis at the widest point of the eyes and measurement of thorax width was measured as the distance between the tips of the two tegulae (T); (b) length of the dissected forewings (W) was taken as the length from the base of the costa to the tip of the wing, crossing through the intersection of the 2nd and 3rd sub-marginal cells with the 2nd marginal cell. The average of both wings was analyzed; (c) length of each mucus gland (MG) along the central axis was measured from the distal bulb to the ejaculatory duct. Lengths were averaged for analysis. Length of each seminal vesicle (SV) was measured along the central axis from the base of the vas deferens to the mucus gland and an average of both was used for analysis.

Figure 2

Newly emerged drones have no spermatozoa in their seminal vesicles. Counts of both total (diamond) and viable (triangle) spermatozoa significantly differ among differently aged drones (p < 0.0001). Total count was naturally higher than viable count, but there was no interaction effect between count type (p > 0.05). Peak counts occurred around day 19.

Figure 3

Whole body wet weights (a) was significantly smaller in older drones (p < 0.0001). There appears to be an initial increase in mass and a sudden drop after 6 days, perhaps the time when drones begin physiological preparation for flight. Thoracic mass (b) is also smaller in older drones (p < 0.05), though this trend is steadier.

Figure 4

Forewing length (a) was significantly shorter in older drones (p < 0.001) as was head width (b) (p < 0.005). Thorax width (c) did not vary among differently aged drones (p > 0.05). Means ± standard errors are reported.

Figure 5

Mucus glands (a) visually increased in size over the first several days of life but did not significantly vary thereafter (p > 0.05) and was not smaller in older drones. Seminal vesicles (b) were significantly smaller in older drone (p < 0.05). Means ± standard errors are reported.

Figure 6

Drones aged 11+ d were analyzed for correlation. Viable spermatozoa count (C), seminal vesicle length (SV), mucus gland length (MG), whole body mass (M), thoracic mass (m), head width (H), forewing length (W), thoracic width (T). All measures were significantly positively correlated (p < 0.05) excepting C:H and C:T (p > 0.05). Two clusters were observed: (1) reproductive development (C, SV, & MG) and (2) body size (M, m, H, W, & T).

Figure 7

(a) Reproductive development, composed of the variables C, SV & MG, was significantly correlated with body size, composed of the variables M, m, H, T & W (p < 0.0001). Though the coefficients for the principal components were generated from flight-aged drones (11 + d), this correlation holds over the entire lifespan of drones; (b) Both principal components vary significantly by age (p < 0.05). Body size (black circle) varied in a relatively linear fashion, with older drones being smaller than younger drones. Reproductive development (yellow diamond) varied along a quadratic function, with drones aged 16 days being the most reproductively developed.