Honey bee workers that are pollen stressed as larvae become poor foragers and waggle dancers as adults

Abstract

The negative effects on adult behavior of juvenile undernourishment are well documented in vertebrates, but relatively poorly understood in invertebrates. We examined the effects of larval nutritional stress on the foraging and recruitment behavior of an economically important model invertebrate, the honey bee (Apis mellifera). Pollen, which supplies essential nutrients to developing workers, can become limited in colonies because of seasonal dearths, loss of foraging habitat, or intensive management. However, the functional consequences of being reared by pollen-stressed nestmates remain unclear, despite growing concern that poor nutrition interacts with other stressors to exacerbate colony decline. We manipulated nurse bees' access to pollen and then assessed differences in weight, longevity, foraging activity, and waggle-dance behavior of the workers that they reared (who were co-fostered as adults). Pollen stress during larval development had far-reaching physical and behavioral effects on adult workers. Workers reared in pollen-stressed colonies were lighter and shorter lived than nestmates reared with adequate access to pollen. Proportionally fewer stressed workers were observed foraging and those who did forage started foraging sooner, foraged for fewer days, and were more likely to die after only a single day of foraging. Pollen-stressed workers were also less likely to waggle dance than their unstressed counterparts and, if they danced, the information they conveyed about the location of food was less precise. These performance deficits may escalate if long-term pollen limitation prevents stressed foragers from providing sufficiently for developing workers. Furthermore, the effects of brief pollen shortages reported here mirror the effects of other environmental stressors that limit worker access to nutrients, suggesting the likelihood of their synergistic interaction. Honey bees often experience the level of stress that we created, thus our findings underscore the importance of adequate nutrition for supporting worker performance and their potential contribution to colony productivity and quality pollination services.

Fig 1. Weight, longevity, and foraging activity of adult workers were reduced when access to pollen was limited during larval development.

Mean (± SEM) A) fresh weight of focal workers at adult emergence, B) longevity of focal workers, C) age at onset of foraging for focal workers who foraged, and D) number of days observed foraging for focal workers who foraged. Focal workers originated from source colonies that were split into colony subunits that had either limited or abundant supplies of pollen when focal workers were reared as larvae. Subunits were either confined to a cool incubator to prevent further pollen foraging (pollen-limited or confined controls) or allowed to forage freely (unconfined controls). When development was complete, focal workers were co-fostered as adults in an unrelated host colony. The experiment was replicated in three separate trials that used different source and host colonies. Means comparisons were made within trials wherever treatment effects were significant after a Bonferroni correction; significant differences between treatments are indicated by letters.

Fig 2. Survivorship of adult workers was lowest when access to pollen was limited during larval development.

Workers were either reared in colonies with limited pollen (and confined to prevent further foraging) or reared in colonies with abundant pollen (either confined or allowed to continue to forage; controls). In each trial, focal workers were introduced into an observation hive after adult emergence; only those workers present 24 hours later were included in the survivorship analysis. The gray area in trial 3 indicates the period over which waggle-dance recruitment was monitored. Raw data are depicted, pooled across colonies within a treatment per trial, rather than Kaplan-Meier estimates of survival function (see Methods). Significant differences in survival among treatments within a trial are indicated by different letters.

Fig 3. Waggle-dance behavior of adult workers was not affected by access to pollen when focal workers were larvae.

Provided are mean per dancer measures of dance performance (± SEM) as focal individuals foraged at unknown food sources in trial 3. All workers were uniquely tagged and individually identifiable. Waggle-dance activity was monitored for 1–2 h/day (as weather and foraging permitted) from the time that workers were min. 12 days to max. 45 days of age (see gray box in Fig. 2). Means were calculated considering only those workers who danced (i.e., zero values were not included for non-foraging or non-dancing focal workers; n = 9 workers reared in pollen-limited, confined colonies; n = 66 workers reared in abundantly supplied, confined controls; n = 41 workers reared in abundantly supplied, unconfined controls). Mean waggle-run duration (number of frames at 30 frames per second, a proxy for distance to advertised food source) was estimated for the first dance performed by each pollen-limited worker and compared to means for the first dances performed by control workers during the same hours of videotape (n = 9 workers reared in pollen-limited, confined colony subunits; n = 20 workers reared in abundantly supplied, confined controls; n = 13 workers reared in abundantly supplied, unconfined controls).

Fig 4. Workers reared in colonies with limited pollen performed waggle dances with greater directional imprecision as adults.

Variability in the A) direction and B) distance components of waggle dances were estimated for workers who were either reared in colonies with limited pollen (and confined to prevent further foraging) or reared in colonies with abundant pollen (either confined or allowed to continue foraging; controls). All dances were performed for a sucrose-solution feeder that was maintained at a fixed location from the observation hive in trial 3. Feeder dances were performed by workers who had treatment-specific marks (paint marks and tags), but individuals were not always uniquely identifiable, so each dance was treated as an independent record. Standard deviations (SD) of the angle and the duration of the waggle runs for each dance were calculated to compare directional precision among treatments. Differences between treatments are indicated by letters where significant treatment effects were found.