Appetitive reversal learning differences of two honey bee subspecies with different foraging behaviors

PEER-REVIEWED Zoological Science section https://peerj.com/articles/5918/ Research articleAnimal BehaviorEcologyEntomologyEvolutionary Studies Eddie Pérez Claudio1, Yoselyn Rodriguez-Cruz2, Okan Can Arslan3, Tugrul Giray​4, José Luis Agosto Rivera4, Meral Kence3, Harrington Wells5, Charles I. Abramson6 November 21, 2018 Note that a Preprint of this article also exists, first published October 11, 2018. Author and article information Abstract We aimed to examine mechanistically the observed foraging differences across two honey bee, Apis mellifera, subspecies using the proboscis extension response assay. Specifically, we compared differences in appetitive reversal learning ability between honey bee subspecies: Apis mellifera caucasica (Pollman), and Apis mellifera syriaca (Skorikov) in a “common garden” apiary. It was hypothesized that specific learning differences could explain previously observed foraging behavior differences of these subspecies: A.m. caucasica switches between different flower color morphs in response to reward variability, and A.m. syriaca does not switch. We suggest that flower constancy allows reduced exposure by minimizing search and handling time, whereas plasticity is important when maximizing harvest in preparation for long winter is at a premium. In the initial or Acquisition phase of the test we examined specifically discrimination learning, where bees were trained to respond to a paired conditioned stimulus with an unconditioned stimulus and not to respond to a second conditioned stimulus that is not followed by an unconditioned stimulus. We found no significant differences among the subspecies in the Acquisition phase in appetitive learning. During the second, Reversal phase of the experiment, where flexibility in association was tested, the paired and unpaired conditioned stimuli were reversed. During the Reversal phase A.m. syriaca showed a reduced ability to learn the reverse association in the appetitive learning task. This observation is consistent with the hypothesis that A.m. syriaca foragers cannot change the foraging choice because of lack of flexibility in appetitive associations under changing contingencies. Interestingly, both subspecies continued responding to the previously rewarded conditioned stimulus in the reversal phase. We discuss potential ecological correlates and molecular underpinnings of these differences in learning across the two subspecies. In addition, in a supplemental experiment we demonstrated that these differences in appetitive reversal learning do not occur in other learning contexts. Cite this as Pérez Claudio E, Rodriguez-Cruz Y, Arslan OC, Giray T, Agosto Rivera JL, Kence M, Wells H, Abramson CI. 2018. Appetitive reversal learning differences of two honey bee subspecies with different foraging behaviors. PeerJ 6:e5918 https://doi.org/10.7717/peerj.5918 Main article text Introduction A honey bee colony shifts its foraging effort as the floral resources come and go in the environment (see Seeley, 1995). This dynamic allocation of foragers is thought to be adaptive since resources are harvested maximally. The basis of this constant response to changes in floral resources is the preference and foraging decisions of individual honey bees. Several mechanisms involving learning have been shown to be important in decisions of individual foragers (Ferguson, Cobey & Smith, 2001). We examined whether plasticity in appetitive learning will differentiate bees of Apis mellifera caucasica subspecies that switch foraging preferences with ease from bees of Apis mellifera syriaca subspecies that do not switch even when reward contingencies change (see Çakmak et al., 2010). Both specialist strategy of A.m. syriaca, and generalist strategy of A.m. caucasica could be adaptive in their respective environments. The hypothesis is that specializing on a single flower type makes the bee faster both in finding the flower and in handling the flower, and thus decreases the time spent outside, at risk, or exposure to predators. Therefore, appetitive learning flexibility in the specialist subspecies, A.m. syriaca should be reduced to keep the bee focused on a single flower type. Alternately, in a low risk environment, a fully plastic foraging choice toward the most rewarding resources is the best solution, and favors greater learning plasticity in the generalist subspecies, A.m. caucasica. Then predation risk sets limits to plasticity in foraging choice (DeWitt, Sih & Wilson, 1998; Murren et al., 2015). Honey bees live in a wide range of habitats, extending from tropical to subarctic, either because of human intervention or because of evolutionary history of the populations (Whitfield et al., 2006; Wallberg et al., 2014). These genetically distinct populations are recognized as subspecies or races. Bringing members of different subspecies together for experiments revealed many genetic differences in behavior and its regulation (Giray et al., 2000; Brillet et al., 2002; Alaux et al., 2009; Çakmak et al., 2009, 2010; Kence et al., 2013; Büchler et al., 2014). Foraging choice differences across two subspecies from Turkey provides the ideal situation to test the underlying learning plasticity differences across specialists and generalists. Previously, A.m. syriaca and A.m. caucasica bees have been studied for genetic, colony and behavioral differences (genetics: Bodur, Kence & Kence, 2007; foraging behavior: Çakmak et al., 2009; colony traits: Çakmak et al., 2010; Kence et al., 2013). The bees from the subspecies A.m. syriaca inhabit southeast Anatolia, a generally dry habitat with longer seasonal foraging periods constrained by periodic blooms of one or few flowers (Kandemir, Kence & Kence, 2000; Kandemir et al., 2006). For foraging A.m. syriaca bees, minimizing predation risk is important. In this region, there is a predatory wasp that can capture foraging honey bees, and bees of this region are demonstrated to have specific behavioral adaptations against this Vespa species, such as reducing foraging activity (Ishay, Bytinski-Salz & Shulov, 1967; Butler, 1974; Ruttner, 1988; Roubik, 1992; Çakmak, Wells & Firatli, 1998). This response is absent in A.m. mellifera (Matsuura & Sakagami, 1973). In contrast, the bees from the subspecies A.m. caucasica inhabit temperate deciduous forests in the northeast of Anatolia and the eastern Black Sea coast regions of Turkey. Weather in these regions limits foraging to a short, 3-month seasonal period, making it important to maximize collection rate. One specific type of plasticity in learning, reversal learning, has been examined because of its potential relevance to tracking changing foraging resources (Ferguson, Cobey & Smith, 2001). The bees learn to associate a stimulus (a floral odor) with a reward and learn to discriminate this from a second odor not associated with reward. Later bees are asked to switch the odor associations. Reversal learning measures behavioral flexibility, and either single or multiple reversions, and either two or more choices are utilized to examine the extent of flexibility (Izquierdo et al., 2017). In comparison of bees of different ages (Ben-Shahar et al., 2000), selected lines (Ferguson, Cobey & Smith, 2001), and subspecies (Abramson et al., 2015), rate of reversal appears to differ, albeit the shape of reversal appears to remain similar (see Fig. S1). In the context of foraging behavior, reversal learning is similar to when a bee visits one flower providing nectar at that time, and later in the day switch to a different flower, that is, providing nectar then (Wagner et al., 2013). In addition, the response of bees to variability in nectar availability is similar to the response of other organisms such as vertebrates to variable reward or resources under experimental or natural conditions (Commons, Kacelnik & Shettleworth, 1987). For instance, if the constant forage rate would provide energetic needs, organisms are likely to abandon variable reward for constant reward (Caraco, 1981; Zalocusky et al., 2016). In previous work, we have demonstrated that bees from the temperate subspecies A.m. caucasica a more likely to switch to a different flower color morph. In contrast, bees from the subtropical subspecies A.m. syriaca are not sensitive to variability in reward, and continue to visit the same flower morph even when rate of reward is one in three visits (Çakmak et al., 2010; Fig. 1). Foraging visits of bees from two subspecies to alternate flowers when preferred flower provides constant or variable amounts of nectar reward. Figure 1: Foraging visits of bees from two subspecies to alternate flowers when preferred flower provides constant or variable amounts of nectar reward. Average percent visits to alternate flower color was significantly less for A.m. syriaca than caucasica. Bees first visited blue, white or yellow flowers. Later they visited alternates or initial preferred flowers with either constant reward (two μl 1M sucrose) or variable reward (only one of three flowers with six μl reward). Sample size: six colonies/subspecies, 30–35 bees/colony, 30–40 choices/bee. Error bars = SE. Factorial ANOVA indicated significant subspecies differences. Groups with different letters above bars are different at p < 0.05 (Çakmak et al., 2010). Download full-size imageDOI: 10.7717/peerj.5918/fig-1 We hypothesized that flower constancy even when faced with variable reward could be due to learning and memory differences of A.m. syriaca bees from other bees, including A.m. caucasica. We used the proboscis extension response (PER) conditioning (Abramson et al., 2015) assay to examine differences in appetitive learning behavior across bees from colonies of both subspecies maintained in a “common garden” apiary (Kence et al., 2013).