Is there any evidence contrary to the pheromone theory of bee washboarding?

Is there any evidence contrary to the pheromone theory of bee washboarding?

We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

I've heard beekeepers say they "don't know" what the function of washboarding is.

James F. Taulman suggested in a 2017 study that the primary function of bee washboarding is the application of pheromones to assist returning foragers in locating the hive.

Is there any evidence contrary to Taulman's theory?

For those interested: Taulman has a youtube channel with some videos of washboarding observations at wild hives.

Division of labor in honey bee colonies is influenced by the foraging gene (Amfor), which encodes a cGMP-dependent protein kinase (PKG). Amfor upregulation in the bee brain is associated with the age-related transition from working in the hive to foraging for food outside, and cGMP treatment (which increases PKG activity)causes precocious foraging. We present two lines of evidence in support of the hypothesis that Amfor affects division of labor by modulating phototaxis. We first show that a subset of worker bees involved in the removal of corpses from the hive had forager-like brain levels of Amfor brain expression despite being middle aged age-matched food-handlers, who do not leave the hive to perform their job, had low levels of Amforexpression. This finding suggests that occupations that involve working outside the hive are associated with high levels of Amfor in brain. Secondly, foragers were much more positively phototactic than hive bees in a laboratory assay, and cGMP treatment caused a precocious onset of positive phototaxis. The cGMP effect was not due to a general increase in behavioral activity cGMP treatment had no effect on locomotor activity under either constant darkness or a light:dark regime. The cGMP effect also was not due to changes in circadian rhythmicity cGMP treatment had no effect on age at onset of locomotor circadian rhythmicity or the period of rhythmicity. The effects of Amfor on phototaxis are not related to peripheral processingelectroretinogram analysis revealed no effect of cGMP treatment on photoreceptor activity and no differences between untreated hive bees and foragers. The cAMP/PKA pathway does not appear to be playing a similar role to cGMP/PKG in the honey bee cAMP treatment did not affect phototaxis and gene expression analysis revealed task-related differences only for the gene encoding the regulatory subunit, but not the catalytic subunit, of PKA. Our findings implicate one neural process associated with honey bee division of labor that can be affected by naturally occurring changes in the expression of Amfor.

One challenge in the study of genes and behavior is to determine how a gene exerts its influence on neurons and neural systems to influence behavioral plasticity. Amfor is the ortholog of the Drosophila melanogaster foraging gene (Osborne et al.,1997) in the honey bee Apis mellifera, which encodes a cGMP-dependent protein kinase (PKG). In the honey bee, an age-related increase in Amfor expression in the brain during the life of a bee is associated with the onset of foraging behavior, and treatment with cGMP causes an increase in PKG activity and precocious foraging(Ben-Shahar et al., 2002).

The onset of foraging in honey bees is the culmination of the process of behavioral development that underlies colony division of labor(Robinson, 1992). A worker bee begins her adult life by progressing through a series of tasks in the beehive and then typically begins to forage at about 3 weeks of age. The timing of a bee's shift from hive to foraging duties is flexible, and depends on the needs of the colony. It is also associated with changes in metabolism, exocrine gland activity, hormone levels, brain structure, brain chemistry and gene expression in the brain (Robinson,2002).

PKG has numerous roles in a nervous system(Ruth, 1999 Wang and Robinson, 1997), but how it influences the shift from working in the hive to foraging in honey bees is not known. Ben-Shahar et al.(2002) suggested that perhaps the upregulation of PKG activity affects honey bee behavioral development via effects on the visual system, because they found strong expression of Amfor in the optic lobe lamina and in a subset of intrinsic cells of the mushroom bodies known to receive visual input(Ehmer and Gronenberg, 2002 Gronenberg, 2001). In addition, cGMP has been shown to have an important role in the development of the visual system in Drosophila(Gibbs et al., 2001). Flies carrying a mutation in a subunit of soluble guanylate cyclase (the enzyme that makes cGMP) show reduced photoreceptor response to light stimuli and altered phototactic behavior (Gibbs et al.,2001). Honey bee division of labor is known to involve maturational changes in responsive to olfactory task-related stimuli (e.g. Robinson, 1987a), but the role of vision in the control of honey bee behavioral development has not been studied.

Bees are extremely visual animals, with a large portion of their brain dedicated to visual processing(Gronenberg, 2001). Foragers perform well in different laboratory-based visual learning paradigms, no doubt because they rely considerably on visual abilities when foraging in the field(Zhang et al., 1999). Foragers use optic flow to measure distance (Esch et al., 2001), discriminate easily between different shapes(Horridge, 2000) and have well-developed color vision (Werner et al., 1988).

We tested the hypothesis that the effects of Amfor on honey bee behavioral development are due, at least in part, to an increase in positive phototaxis. We focused on this aspect of the visual system because honey bees experience a major change in exposure to light when they shift from working in the dark hive to foraging outside. Menzel and Greggers(1985) have shown that foragers are positively phototactic, but it is not known whether this behavior is developmentally regulated. Young bees do emerge from the hive to take brief defecation and orientation flights prior to the beginning of their foraging phase (Capaldi et al., 2000),but these are transient events. Perhaps more chronic increases in positive phototaxis occur in older pre-foraging bees, which then positions them closer to the hive entrance. There they may be induced to forage by exposure to olfactory and mechanical stimuli, such as communication by successful foragers via the dance language (Frisch,1967). A behaviorally related change in phototaxis has recently been reported for queen harvester ants (Messor pergandei) queens are positively phototactic as virgins but became negatively phototactic after mating (Julian and Gronenberg,2002).

We tested the hypothesis that the effects of Amfor on honey bee behavioral development are due, at least in part, to an increase in positive phototaxis by addressing three issues. First, we determined whether the previously reported increase in Amfor brain expression in foragers is also detectable in the brains of bees that are not foraging, but are nevertheless engaged in a task that requires leaving the hive. This was accomplished by comparing two groups of middle-aged bees: food handlers and corpse removers (undertakers). Although a majority of bees that are found outside the hive are foragers, other tasks such as undertaking occur outside as well. Undertakers are a subset group of bees that pick up corpses in the hive and then fly out to remove them(Visscher, 1983). Undertakers are younger than foragers, but they have high, forager-like titers of juvenile hormone (JH), which influences the pace of honey bee behavioral development(Huang et al., 1994). Second,we asked whether there is an ontogeny of phototaxis behavior in association with honey bee behavioral development, and if so, whether it can be accelerated by treatment that activates PKG. Third, we studied whether the observed treatment effects of cGMP on phototaxis are due to changes in overall levels of locomotor activity, the timing of the onset of locomotor circadian rhythmicity (Bloch and Robinson,2001 Moore et al.,1998 Toma et al.,2000), or general photoreceptor sensitivity.

In addition, we explored whether the cAMP/PKA pathway may also be playing a role similar to the cGMP/PKG pathway in the honey bee. These pathways are known to interact in other behavioral systems including in honey bees(Muller and Hildebrandt,2002). We determined the effects of cAMP treatment on phototaxis and measured the expression of genes encoding the regulatory and catalytic subunits of PKA in the brains of bees performing different behaviors.


The study of honey bee pheromones started in the 1960s and since then many advancements have been made in the knowledge of composition of pheromone blends, their glandular origin, and their colony target. But while we knew the effects of many of these pheromones, for a long time we could only speculate as to the neuronal mechanisms that mediate between pheromone and function. Only recently, with the development of molecular and genetic tools, some progress has been achieved in this direction. Thus, we are just starting to gain some awareness of the neurophysiological pathways of pheromone reception and processing in the bee brain, and only a few mechanisms have been entirely elucidated.

As previously reported, releaser and primer pheromones exert different effects on the receiver, the former being immediate and transitory and the latter delayed and long-termed. This difference suggests that two different mechanisms may exist by which pheromones influence the receiver: a direct effect on neural transmission for releaser pheromones against an effect on physiological processes (e.g., hormonal, metabolic, or genetic changes) for primer pheromones.

A common question regarding queen primer pheromones is their mode of action in regulating worker reproduction and behavioral development: is it by means of a controlling mechanism (queen pheromone as a suppressive agent) or a signaling one (queen pheromone as an “honest” signal) (Strauss et al. 2008)? In the 𠇌ontrol” hypothesis the queen pheromones manipulate workers coercively by inhibiting their ovarian and behavioral development. In the signal hypothesis, also called the cooperation hypothesis, queen pheromones simply act to signal to workers the queen presence and its egg-laying potential, rather than to manipulate worker behavior and/or physiology. In the presence of a strong and healthy queen, workers refrain themselves from reproducing and prevent nestmate workers from reproducing (worker policing) in order to maximize colony fitness. When workers perceive a decline in the fecundity of the queen, they can activate their ovaries to produce their own male offspring (Keller and Nonacs 1993 Kocher and Grozinger 2011 Le Conte and Hefetz 2008 Strauss et al. 2008).

Both hypotheses make sense from an evolutionary point of view, and several authors tried to collect evidence to support one or the other theory, but without giving a definite and unquestionable response. Both theories could explain the richness and variety of queen pheromones, whose components increase with increasing level of sociality, such as both theories could support the variability of response given by different workers to the colony pheromones (Kocher and Grozinger 2011 Strauss et al. 2008). In either case, the way the pheromone is detected and processed in the brain of different receiver workers seems to play a crucial role in the regulation mechanism.

Different pheromones use different ways of transmission from the producer to the receiver. Volatile substances, such as the alarm and Nasonov pheromones produced by the workers, and the components of the QMP that attract drones for mating and workers for swarm clusters, use a dispersal mechanism footprint pheromones, BP, and most components of QMP are transmitted primarily by contact, and the same is true for the esters produced in other queen glands (e.g., tergal and Dufour’s), which are delivered as an integral part of the queen signal together with QMP for this reason some authors have named them “passenger pheromones” (Keeling and Slessor 2005 Slessor et al. 2005).

Whatever the way of transmission of the pheromone, the reception process starts in the receiver olfactory system.

5.2.1. R eception of the P heromonal S ignal Olfactory Receptor Neurons

In insects, the peripheral odour detection starts in the peripheral chemosensory system with the detection of the chemical signal by olfactory receptor neurons (ORNs), which express specific olfactory receptors (ORs). ORs are seven-transmembrane domain proteins coupled to G proteins following the binding with odorant molecules cellular transduction cascades are activated, implicating the production of cAMP, leading to depolarization and action potentials.

These receptors are located mainly in the antennae, where they are organized in olfactory sensilla of various shapes the poreplate sensilla are the most frequent sensilla in the honey bee antennae (Figure 5.5). A poreplate sensillum is formed by an oval-shaped thin cuticular plate with numerous minute pores and is innervated by 5 to 35 ORNs with their corresponding ORs. Each poreplate contains the whole range of ORs and thus represents a whole miniature system (Brockmann and Brueckner 1995 Sandoz 2011).


Schematic representation of the reception pathway for general odors and social pheromones in the worker honey bee (left) and for sexual pheromones in the honey bee drone (right) from the antenna poreplate sensilla to the antennal lobe, and the following (more. )

Odorant molecules reach the dendrites of ORNs by diffusing through an extracellular fluid, called sensillum lymph, filling the sensillum cavity. In this fluid, odorant binding proteins (OBPs) transport the odorants to the ORNs (Sandoz 2011). While OBPs bind general odorants, a specific class of OBPs, the PBPs (pheromone binding proteins) are specialized in binding sexual pheromones and are present mainly in male insect sensilla (Laughlin et al. 2008 Leal 2005). OBPs and PBPs play an essential role in the detection of general odors and pheromonal molecules and in their transduction, passing the molecules to the sensory neuron membrane protein, which then delivers it to the olfactory receptor (Pesenti et al. 2008, 2009). Another class of soluble chemosensory proteins (CSPs), which shares no sequence homology with either PBPs or general OBPs, has been described in honey bees (Danty et al. 1998). However, in honey bees only 21 genes coding for OPBs and six coding for CSPs have been found in the genome, so that the relative importance of these molecules in the process of odor perception is still unclear (Forêt and Maleska 2006). The results of a recent study using a proteomic approach show that 12 of the 21 OBPs and 2 of the 6 CSPs predicted in the honey bee genome are present in the foragers’ antennae (Dani et al. 2010) and some OBPs are found to be more highly expressed in the mandibular glands of different honey bee castes, suggesting their involvement also in solubilization and release of semiochemicals (Iovinella et al. 2011). Three main subclasses of OBPs are defined in honey bees on the basis of antennal specific proteins (ASPs), namely ASP1, ASP2, and ASP3 (Danty et al. 1997, 1998). ASP1 is thought to be associated with QMP because of its higher abundance in drone sensilla and the ability to bind 9-ODA and 9-HDA, the most active components of the queen pheromone blend, while ASP2 and ASP3 bind general odorants (Danty et al. 1999). One of the CSPs, called ASP3c, specifically binds brood pheromone components and not general odorants or other pheromones (Briand et al. 2002).

The wide range of pheromones described in honey bees, together with the great number of environmental odors they encounter, suggests a highly developed olfactory system that must be able to discriminate a large number of volatile substances. Indeed, the sequencing of the honey bee genome allowed the identification of an exceptionally high number of OR types (160�), compared to the already known ORs of Drosophila melanogaster (62 ORs) and Anopheles gambiae (79 ORs) (Robertson and Wanner 2006). This high number is evidently linked to the extraordinary olfactory abilities of honey bees, whose social life requires the perception of several pheromone blends as well as kin recognition signals and numerous floral odors.

It is presumable that different ORs are differentially expressed according to caste and function indeed, among the identified antennal ORs, the AmOR11, which is upregulated in drones, was recently demonstrated in male antennae to specifically detect 9-ODA and to respond to all the main QMP components (Wanner et al. 2007). On the contrary, a number of other receptors (OR63, OR81, OR109, OR150, OR151, OR152) are more highly expressed in worker bees than in drones and are probably linked to floral odorant reception, being differentially expressed in bees which live in different environments and thus experience diverse floral scents (Reinhard and Claudianos 2013). Antennal Lobes and the Glomeruli

The ORNs project their axons to a specific area of the deutocerebrum called antennal lobe, which is organized in densely packed nervous structures termed glomeruli (Figure 5.5). The axons of several ORNs converge to the glomeruli through four sensory tracts (T1–T4), which define four subpopulations of glomeruli, two containing about 70 glomeruli each (T1 and T3) and two with seven glomeruli each (T2 and T4). The ORNs of an individual poreplate project to all four glomerular subpopulations and are therefore distributed across the whole antennal lobes (Brockmann and Brueckner 1995 Flanagan and Mercer 1989 Kelber et al. 2006).

The arrangement and number of glomeruli are largely species-specific and vary from about 32 in the mosquito Aedes aegypti to more than 1000 in locusts and social wasps. In honey bees the workers possess 166 glomeruli and the drones 103. The latter also have four large glomerular complexes exclusively committed to processing sexual pheromones, probably with a functional specialization for a specific pheromone substance in each of the four complexes (Arnold et al. 1985).

It has to be noted that the number of glomeruli in the honey bee antennal lobe is almost equal to the number of ORN types expressing a given OR in the antennae, supporting the hypothesis of a linear relationship one-receptor/one-neuron/one-glomerulus.

Within the glomeruli the ORN axons synapse with two other kinds of neurons: the local neurones and the projection neurones (Figure 5.5). The former are mainly GABAergic neurons with an inhibitory output, while the latter are cholinergic neurons that show either excitatory or inhibitory responses to odors. The local neurons can be classified into two main types: the homogeneous local neurons, which innervate most if not all glomeruli in a uniform manner, and the heterogeneous local neurons, which innervate only a small subset of glomeruli with one dominant glomerulus that is densely innervated and a few others with very sparse processes (Flanagan and Mercer 1989 Sandoz 2011). The function of local interneurons is to interconnect the glomeruli and modulate the signal coming from ORs.

Projection neurons leave the antennal lobe via a variable number of pathways called antennocerebral tracts, connecting it with different areas of the protocerebrum, mostly the calyces of the mushroom bodies and the lateral protocerebrum (Hansson and Anton 2000 Kay and Stopfer 2006). Projection neurons can also be classified into two types: uniglomerular projection neurons branch in a single glomerulus within the antennal lobe and project to the mushroom body or the lateral horn through two major antennocerebral tracts, while multiglomerular projection neurons branch in most glomeruli and their axons follow three lesser antennocerebral tracts leading to other areas of the protocerebrum surrounding the α-lobe of the mushroom body or extending toward the lateral horn (Sandoz 2011, 2013). Pheromone Processing in the Glomeruli

Within the glomeruli the olfactory signal undergoes an important integration and encoding before being transmitted to the higher centers. Glomeruli are the anatomical and functional units of the antennal lobes and constitute sites of synaptic interaction between different neuron types. The activity patterns of antennal lobes in response to odors was studied in the honey bee by optical imaging techniques (Galizia et al. 1997, 1998). Axons of ORNs expressing the same odorant receptor or with similar odor specificities converge onto the same glomerulus. Considering that a single type of molecule interacts with a number of different ORNs, which activate a similar number of glomeruli, an odor blend is represented by the activation of a variable number of glomeruli, resulting in a spatial representation of the odor within the antennal lobe (Galizia et al. 1999 Joerges et al. 1997 Sachse and Galizia 2003 Sachse et al. 1999). This representation is variable in time and depends on the olfactory experience therefore, odorants are represented in the antennal lobe as changing spatiotemporal patterns of glomerular activity (Sandoz et al. 2003). The early olfactory learning during young adulthood enhances glomerular activity and modifies the spatiotemporal response patterns these changes affect neural activity until the time when bees initiate foraging activities (Arenas et al. 2009 Galizia and Vetter 2005).

Social (nonsexual) pheromones, like citral and geraniol (components of the Nasonov gland), IPA (the major component of the alarm pheromone associated with the sting apparatus), and the worker mandibular gland pheromone 2-heptanone, are coded in the antennal lobe as “general” odors since they elicit activity in the same brain region as environmental odors (Galizia et al. 1999 Joerges et al. 1997 Sachse et al. 1999). IPA elicits strong responses in several glomeruli that also exhibit strong responses to orange, clove oil, limonene, and several plant extracts (Galizia and Menzel 2001). Nevertheless, Sandoz et al. (2001) found that IPA and 2-heptanone, which share an alarm role but have a different chemical structure and source, induce a reciprocal generalization in olfactory conditioning tests, suggesting that a similarity in the neural representation of odor could rely not only on the chemical structure but also on their functional value (Sandoz et al. 2001).

Wang et al. (2008) investigated the neural activity elicited by eight components of the sting pheromone, compared with the whole bee sting apparatus, at the level of the antennal lobes of honey bee workers. They found that the sting preparation evokes a clearly distinct glomerular pattern compared to those of individual pheromone components (e.g., IPA-activated glomeruli in the medial part of the antennal lobes), whereas the stings activated the lateral dorsal part. It seems that the sting apparatus pheromone is processed in a similar way to general odors, since the main determinant of glomerular activation is its chemical structure rather than its pheromonal value. However, in contrast to the elemental strategy used for processing nonpheromonal mixtures, where the neural representation of mixtures of two to four nonpheromonal odors could be linearly predicted based on the neural representation of each component (Deisig et al. 2006), pheromonal blends do not follow such a linear representation, revealing more complex strategies for the processing of pheromonal mixtures in the honey bee antennal lobe (Wang et al. 2008). Sexual Communication: Drone Reception of QMP in Macroglomerular Complexes

Male insects, including honey bee drones, have a specialized olfactory subsystem to detect female sexual pheromones even at long distances. This subsystem is characterized by a large number of ORs and ORNs sensitive to the components of the female pheromones. Their axons converge to hypertrophied glomerular subunits called macroglomerular complexes that are located in the antennal lobes. In honey bees the sexual dimorphism of the reception system is evident (Figure 5.5) compared with worker bees, drones have larger antennae, with a flagellum surface twice as large as that of the workers, and about seven times as many poreplate sensilla (around 18,000 versus 2700) and ORNs (around 340,000 versus 65,000) (Esslen and Kaissling 1976). In addition, the female antennal lobe is composed of isomorphic glomeruli (about 160 in workers and 150 in queens), whereas in drones there is a reduced number of these “ordinary” isomorphic glomeruli (about 100) but there are four voluminous macroglomeruli (Arnold et al. 1985).

The sexual dimorphism of the reception system corresponds to different neuronal strategies to detect and respond to the pheromone signals. Electroantennographic studies showed that while worker antennae have a very similar response to the various QMP components, suggesting that there is no antennal specialization with regard to the number of sensory neurons, in contrast, drone antennae showed marked responses to 9-ODA and to synthetic QMP compared to the other QMP components. This high antennal response is characteristic of sexual pheromones that elicit a long-distance reaction and is attributed to a much higher number of sensory neurons in the male antennae (Brockmann et al. 1998).

These results confirm that worker antennae have a kind of generalized antennal tuning with no significant differences in the number of sensory neurons for individual mandibular pheromone components, while drone antennae are specialized in the perception of one component of the mandibular pheromone, 9-ODA. This scenario is in accordance with the above described differential morphology in the olfactory system between workers and drones, and confirmed by the finding that drones of the more primitive honey bee species, Apis florea, have only 1200 poreplate sensilla per antenna and only two macroglomeruli in their antennal lobes, corresponding to a minor complexity of the sex pheromone mixture in this species compared to A. mellifera (Brockmann and Bruchner 2001).

Wanner et al. (2007) identified four candidate sex pheromone ORs from the honey bee genome based on their higher expression in drone antennae compared to worker antennae. This number coincides with the number of macroglomeruli in the drone antennal lobe, but only one of them, the already cited AmOr11, specifically responds to 9-ODA, while the other three could not be linked to any queen pheromone component.

Further analysis of drone antennal lobes led to the discovery that the ventral part carries only ordinary glomeruli while the dorsal part shows two of the four macroglomeruli, one located dorsomedially (MG1) and the other on the dorso-lateral side (MG2). Optical imaging of the antennal lobe showed that floral odors and blend mixtures induced focal responses on the ventromedial side of the antennal lobe, a region rich in ordinary glomeruli. In contrast MG2 is clearly and specifically devoted to the reception of the QMP main component 9-ODA, which does not induce signals in regions other than MG2. Among the other QMP components, HOB and HVA induced activity mostly in two ordinary glomeruli in the center of the frontal region, which showed responses also to floral odorants 1-hexanol, limonene, clove oil, and orange oil blends, while 9-HDA and 10-HDA induced only very low and diffuse signals in ordinary glomeruli that could not be measured (Sandoz 2006, 2007). The fact that HVA and HOB are detected in drones by the general olfactory system and not by the pheromonal subsystem can be explained by their different pheromonal role: in fact they are produced mainly by mated queens and not (or very little) by virgin queens, suggesting that they are used for the induction of workers’ retinue behavior and not for drone attraction by virgin queens (Plettner et al. 1997). The role of 9-HDA and 10-HDA as sex attractants remains unclear.

The different organization of the olfactory system between workers and drones reflects their diverse role in the honey bee society: drones exhibit a clear olfactory specialization for the sexual pheromone 9-ODA consistent with their exclusive reproductive role in the hive, while workers show a broader range and less specific response for both pheromonal and nonpheromonal odors consistent with the use of these different signals in different behavioral contexts (Sandoz et al. 2007). Pheromone Processing in Higher Centers

Processed olfactory information leaves the antennal lobe by the projection neurons, towards higher-order brain centers, especially the mushroom bodies and the lateral horn (Figure 5.5).

Olfactory inputs project to a specific area of the mushroom bodies, the Kenyon cells, which form two cup-shaped regions called calyces in each brain hemisphere. The calyces are anatomically and functionally subdivided into the lip, the collar, and the basal ring. The lip region and the inner half of the basal ring receive olfactory input, whereas the collar and outer half of the basal ring receive visual input. The axons of Kenyon cells project in bundles into the midbrain, forming the peduncle and the vertical and horizontal lobes, also called α and β lobes (Strausfeld 2002).

The mushroom bodies receive not only olfactory and visual signals, but also mechanosensory and gustatory inputs. They play an important role in the process of associative learning of olfactory stimuli but also act as a multisensory integration center with a feedback and modulatory function (Mercer and Erber 1983). They are also involved in higher nervous functions, such as learning, memory, and cognitive processes.

In contrast, the processing of olfactory stimuli in the lateral horn are still mostly unknown, including the topography of neurons leaving the lateral horn and the descending pathways involved in behavioral output. In Drosophila this region is divided in two main subregions that separately process pheromones and fruit odors (Jefferis et al. 2007) since the honey bee’s lateral horn shows a specific compartmentalization with at least four subcompartments, an organization similar to that of Drosophila could exist in the honey bee, with a specific pheromone processing region in the lateral horn.

Given that no dedicated glomeruli have been found in workers for the processing of pheromones, Sandoz et al. (2007) hypothesized that specific recognition of pheromones, especially the social ones, may take place at higher processing levels downstream from the antennal lobes. It is conceivable that particular Kenyon cells could recognize specific combinations of activated projection neurons, which would indicate that the detected stimulus is a pheromone.

5.2.2. P rocessing and M odulation of the P heromonal S ignal

The reception and processing of pheromones lead to a response in the receiver that corresponds to a behavioral and physiological change. But how does this process work? The response to pheromones involves both environmental and physiological factors, since pheromones induce a behavioral plasticity in the receiver through a shift in neural response thresholds to environmental conditions.

Releaser pheromones act through a direct and unambiguous pathway in which one pheromone evokes one response in the receivers. In contrast, primer pheromones induce more deep and prolonged effects that can be modulated by the receiver to give a different behavioral response according to its physiological state. These different patterns suggest a different way of action for these two types of pheromones, but until now evidence suggests that the two pathways are partly overlapping and involve similar neuronal and physiological mechanisms.

Study of the mode of action of pheromones should first take into account that many factors affect their reception and processing. The same chemicals can be perceived and processed in a different manner according to the physiological state of the receiver, which in turn is influenced by both genetic and environmental factors correlated to the social environment and the individual developmental stage.

A well-known example is the response to QMP by workers of different ages: Pham-Delègue et al. (1993) demonstrated that there is an age-dependency and experience-dependency in the attraction effect of QMP toward workers. Furthermore, they showed that the olfactory environment experienced in the first day of adult life can strongly modify the functioning of the olfactory nervous system and thus worker behavioral responses (De Jong and Pham-Delègue 1991 Pham-Delègue et al. 1991). This was observed both for general olfactory sensitivity and for pheromonal stimuli, suggesting that age and experience induce different behavioral responses linked to the plasticity of the olfactory system at a peripheral or central level. The relationships between peripheral sensitivity, signal processing, and behavioral responses have only recently started to be elucidated.

The behavioral development from nurses to foragers is accompanied by a brain plasticity that involves in particular the antennal lobes and the mushroom bodies. This transition from life inside the hive to activities outside the hive is associated with a distinct increase in antennal lobe and mushroom body size: the volume of glomeruli changes with the shift to foraging duties, and forager bees have a larger mushroom body calyx than nurse bees of the same age (Brown et al. 2004 Farris et al. 2001 Maleszka et al. 2009). This increase is due to a growing number of neural connections, driven by the richer sensory experience of the outside life.

Another useful approach to uncover the physiological mechanism of pheromone effects exploits genetic differences in worker responses. For instance, some workers are highly responsive to QMP, while others respond poorly or not at all in laboratory bioassays (Kaminski et al. 1990 Pankiw et al. 1994, 1995). There may be genetic and physiological differences between high and low responding workers in receiving or responding to the queen pheromonal message and these differences could provide a powerful tool to dissect the neurochemical pathways of QMP effects (Winston and Slessor 1998).

Pheromones could act by modulating sensory response thresholds which affect the probability of workers performing certain behaviors, such as nursing, foraging, or defence. Besides QMP, alarm pheromones also show this modulating effect, for example on appetitive and aversive learning, which are important behaviors in forager and guard bee workers (Hunt 2007 Urlacher et al. 2010).

The different substances that are possibly involved in the neuromodulation of pheromone signals in the bee brain will be described in the following section, together with some interesting discovered cases of the pheromonal effect on specific functions. Modulation of the Signal: The Role of Biogenic Amines and Juvenile Hormones Brain Amines as Neuromodulators

In the honey bee brain several biogenic amines with potential modulatory function have been detected both in the central and peripheral nervous system. These molecules function as neurotransmitters, neuromodulators, and neurohormones, mediating a diversity of physiological and behavioral functions. In particular, dopamine (DA), serotonin (5-hydroxy-tryptamine, 5-HT) and octopamine (OA), which are all neurotransmitters and long-term brain modulators, seem to be involved in the modulation of behavior, which is functionally linked to pheromone activity (Mercer 1987 Mercer and Menzel 1982).

Biogenic amines in the honey bee brain are synthesized by a relatively small number of modulatory neurons, which often possess widespread projections. The mushroom body calyces in particular receive input from OA and DA neurons, which play an important role in associative learning (Bicker 1999).

DA neurons are present in most parts of the bee brain and in the subesophageal ganglion, representing about 0.1% of the entire neuronal population. Most are located in the mushroom bodies below the lateral calyx and in the anterior-ventral protocerebrum. DA neurons occupy large volumes of neuropil and DA fibers synapse onto the antennal lobes and the Kenyon cell bodies, suggesting a role in mediating distant rather than local neural interactions (Schaefer and Rehder 1989 Schuermann et al. 1989).

5-HT neurons are found in all areas of the brain, in particular the optic lobes, but 5-HT-immunoreactive fibers innervate the mushroom bodies outside the calyces, the antennal lobes, and almost all parts of the central body (Gauthier and Grünewald 2013). Antennal glomeruli contain 5-HT fibers restricted around the margin (Schuermann and Klemm 1984) and a large 5-HT interneuron interconnects the deutocerebral antennal and dorsal lobes with the subesophageal ganglion and descends into the ventral nerve chord (Rehder et al. 1987).

OA neurons are represented in most of the cerebral ganglion, but mainly in five brain regions: in the pars intercerebralis, mediodorsal to the antennal lobes, on both sides of the protocerebrum midline, between the lateral protocerebral lobes and the dorsal lobes, and on either side of the central body. Fine networks invade the antennal lobes, the calyces, and a small part of the α-lobes of the mushroom bodies, the protocerebrum, and all three optic ganglia (Kreissl et al. 1994). Another unpaired median cluster of OA neurons is located within the subesophageal ganglion, where the VUM neurons were identified (see Section

The level of these three biogenic amines (5-HT, DA, OA) in the honey bee brain has been shown to vary during worker development, namely active foragers had significantly higher levels of amines than younger bees working in the hive. These variations are age- and task-dependent and can be correlated to the behavioral development of workers (Schulz and Robinson 1999 Taylor et al. 1992 Wagener-Hulme et al. 1999). This variability thus reflects a differential responsiveness to stimuli associated respectively with brood care or with foraging, such as optical cues (nurse bees live in the dark while foragers need light to orientate), odorant signals (flower and environmental scent), and also learning and memory, since foraging tasks demand cognitive functions for orientation, flower handling, and communication. Furthermore, high levels of DA in the honey bee brain were found to be correlated with ovarian development (Sasaki and Nagao 2001) and the dietary administration of dopamine is able to activate ovaries in queenless workers, suggesting a role of dopamine in the regulation of the reproductive status of honey bee workers (Dombroski et al. 2003).

The levels of amines can vary also independent of age: a different level of DA and 5-HT was found in the optic lobes of nectar foragers and pollen foragers, behaviors that are typically performed at similar ages (Taylor et al. 1992), and between food storers and comb builders, the former having significantly lower levels of DA (Wagener-Hulme et al. 1999). This non-age-dependent difference can be correlated to a differential development of specific brain functions correlated to the performed tasks. There is a different modulation of amine levels in the two brain regions involved in the division of labor, the optical lobes, and the mushroom bodies. In the optical lobes the amounts of DA, 5-HT and OA vary significantly with worker age, but not with task, whereas in mushroom bodies they vary significantly with worker behavior, but not with age (Schulz and Robinson 1999).

Among the three amines, OA is the one that exhibits the most robust association with behavior: foragers had significantly higher brain levels of OA compared to bees performing in-hive tasks, such as nursing or food storing, independent of age (Schulz and Robinson 1999 Wagener-Hulme et al. 1999). The strong correlation between OA concentration in the antennal lobes and worker task suggests that it plays a causal role in the regulation of honey bee behavioral development. In particular, its increase in the antennal lobes seems to be involved in the release and maintenance of the foraging state since the administration of OA to workers at the foraging age results in an earlier onset of foraging, but when administered to younger workers it produces no effects (Schulz and Robinson 2001 Schulz et al. 2002a).

The influence of OA on foraging behavior probably acts through the regulation of response to foraging-related stimuli that involve learning and memory. This is supported both by anatomical and experimental findings: OA fibers were found in all neuropils that contain pathways for proboscis extension learning (Kreissl et al. 1994) OA administration enhances worker responsiveness to unconditioned olfactory stimuli, probably producing a central excitatory state in which the effectiveness of sensory stimuli is improved (Mercer and Menzel 1982) furthermore, while both DA and 5-HT injected into the bee brain reduce the response to a conditioned olfactory stimulus, OA-treated bees do not have a reduced response. The application of DA in the mushroom body causes a reduction of potentials after antennal stimulation that can account for the reduced response (Mercer and Erber 1983). Further studies confirmed the role of OA in appetitive olfactory learning in bees: injections of this amine in the honey bee brain provide a substitute for sucrose reward and induce olfactory learning (Hammer and Menzel 1998) last, blocking OA receptors disrupts olfactory conditioning (Farooqui et al. 2003). Recent research has examined in depth the role of DA neurons in aversive learning and of OA neurons in appetitive learning (see Sections and Brain Amines and Pheromones

It is known that queen pheromones act as typical tranquillizer signals, suppressing perception and stabilizing emotional agitation especially of young worker bees (Lipinski 2006). For instance, workers in queenless colonies tend to be agitated, nervous, and aggressive it seems that queen pheromones act on workers as a sort of social peacemaker. This effect is achieved through different physiological and hormonal mechanisms. In queenright colonies young workers have significantly lower levels of all three main biogenic amines and JH titers compared to queenless colonies: the calming effect is probably exerted by lowering the level of neurotransmitters and by decreasing the excitation of corpora allata, which results in a reduced arousal to external stimuli. A similar calming effect is exerted by brood pheromones and by mandibular pheromones of older workers (Lipinski 2006).

To understand the role of brain amines in the modulation of pheromonal signals, the relationship between their level in the worker brain and the worker response to pheromones was investigated in several studies. For example, Harris and Woodring (1999) found that in honey bees the ingestion of 5-hydroxytryptophan, a precursor of 5-HT, causes a reduction of the worker response to IPA, measured as buzzing response. On the contrary, the ingestion of L-DOPA, precursor of DA, has no effect on the buzzing response stimulated by IPA, suggesting that response to alarm pheromone in honey bees is regulated only by 5-HT metabolism, while it is known that DA and 5-HT are both involved in the neuromodulation of aggressive behavior in many vertebrates and invertebrates (Hunt 2007).

OA has been shown to be quite strictly involved in the response to pheromones linked to behavioral development, which we know to be regulated by the demographic composition of the colony and by the presence of brood, through worker and brood pheromones. Barron et al. (2002) showed that OA is able to enhance worker responsiveness to brood pheromones and to decrease responsiveness to social inhibition exerted by adult bees. OA thus acts as a modulator of pheromonal communication by regulating the response thresholds to worker and brood pheromones. However, the modulation of brood pheromone response is selective for the foraging stimuli, since other functions regulated by this pheromone are not enhanced by OA, like capping behavior (Barron and Robinson 2005). Furthermore, OA does not enhance the response to other pheromonal signals, like retinue response to QMP. The specific mechanism by which OA achieves these results is not yet clear it may act by modulating ORNs in the antennal lobes or by modulating the neuronal circuits involved in the processing of the olfactory stimulus within the mushroom bodies (Schulz et al. 2002a). Neurons of the octopaminergic VUM family may be involved in this modulating function: the VUM mx1 neuron projects from the subesophageal ganglion, where it gets gustatory input from sucrose receptors, to the brain, meeting the olfactory pathway in three areas: the antennal lobes, the mushroom bodies calyces, and the lateral horn thus it may act by combining olfactory and gustatory stimuli with higher functions (Hammer 1993 Schröter et al. 2006). Juvenile Hormone and Pheromones

Similar to brain amines, the level of JH is functionally correlated to worker behavioral development: JH levels are higher in foragers compared to nurses, and treatment with JH or JH analogues results in precocious foraging (Huang et al. 1991 Robinson 1987a). It has been demonstrated that QMP is able to reduce the titer of JH in workers (Kaatz et al. 1992 Pankiw et al. 1998), which results in the lower level of JH in nurse bees, which are in strict contact with the queen and thus with QMP, compared to foragers.

There is a strict relation between the level of OA in the honey bee brain and the level of JH in the hemolymph: OA stimulates production of JH in vitro (Kaatz et al. 1994) and treatment with the JH analog methoprene results in increased forager-like levels of OA in the antennal lobes of preforager workers (Schulz et al. 2002b). The regulation of foraging behavior probably passes through an increase in OA levels in the brain, since allatectomized bees (no JH production) can still initiate foraging after an OA treatment. The timing of OA and JH presence is consistent with the hypothesis that JH acts earlier in the process of forager development as a trigger factor, while OA acts later but more rapidly as a releaser factor of foraging behavior (Schulz et al. 2002b). These findings suggest that the variability in JH and OA levels between workers of different age and task are a key factor in modulating the worker behavioral response to pheromones, but it is not fully established whether JH and OA act through the same or different neural pathways.

The hypothesis that JH influences age-dependent olfaction was tested by examining the effect of the JH analog methoprene on alarm pheromone perception (Robinson 1987b). Worker sensitivity to alarm pheromone increases with age (Collins 1980) and with increasing group size (Southwick and Moritz 1985), indicating a strong influence of the social context on pheromone processing. Methoprene strengthens the behavioral response to alarm pheromone at every age, but is strongest between 5 and 8 days of age. Contrary to behavioral assays, the electroantennographic response to alarm pheromone did not increase in workers after day 5 and was not affected by methoprene: this shows that the honey bee peripheral olfactory system reaches full maturity 4 days after adult emergence and suggests that hormonal modulating effects on pheromone perception occur in the central nervous system (Masson and Arnold 1984 Robinson 1987b). Direct Modulation of Worker Behavior: HVA Mimic of Dopamine

Another interesting cue in the study of pheromone processing in the bee brain came from the observation that one of the components of QMP, HVA, has a similar structure to DA, one of the biogenic amines that plays a role in honey bee behavioral regulation (Beggs et al. 2007). The presence of this compound within the pheromonal blend suggested that exposure to HVA might affect DA function, modulating dopaminergic neural pathways.

Three DA receptor genes Amdop1, Amdop2, and Amdop3 were identified in the honey bee mushroom bodies the receptor density, their gene transcript, and levels of gene expression have been found to change during the lifetime of the adult worker bee (Humphries et al. 2003 Kokay et al. 1999 Kurshan et al. 2003 Mustard et al. 2003). Beggs and Mercer (2009) demonstrated that HVA selectively activates the D2-like DA receptor Amdop3.

The application of QMP to worker honey bees alters DA receptor gene expression, mainly lowering Amdop1 transcript levels consistently, the DA-evoked response, measured as intracellular cAMP level, is lower in mushroom bodies of workers exposed to QMP or HVA (Beggs et al. 2007). This finding is in agreement with the hypothesis that HVA plays a direct role on the modulation of DA levels in the brain. Further confirmation came from an experiment in which workers exposed to a mated queen (which produces higher levels of HVA) showed significantly lower brain DA levels than workers of the same age exposed to a virgin queen (low or absent HVA production) HOB, the other QMP component produced by mated and virgin queens, showed no effect on DA levels of worker brain. Finally, activity levels in bees treated with QMP are reduced, but this effect can be reversed by a treatment with L-dopa, a precursor of DA (Beggs et al. 2007). Taken together, all these results confirm that HVA alone is able to mimic the effects of QMP on DA levels in the honey bee brain and that DA pathways are not affected by other components of the QMP blend.

Another possible role of HVA in the QMP blend focuses on the inhibition of ovarian reproduction in workers: since the treatment of queenless workers with dopamine enhances ovarian development in workers (Dombroski et al. 2003), HVA may inhibit ovarian activation by acting agonistically on the dopamine pathway. However, a direct effect of HVA on ovarian development has not yet been confirmed. Worker Attraction and Aversion: The Role of Pheromones on Appetitive and Defense Behavior

The appetitive learning conditioning in honey bees is a well-known experimental technique in which bees rewarded with sucrose on particular stimuli become able to respond to the same stimulus or to a similar one even without sucrose reward the response is typical and measurable, consisting in the proboscis extension reflex (PER) (Giurfa 2007). On the contrary, the aversive learning conditioning consists in training bees to a defensive response, namely the SER, in response to potentially noxious stimuli. This is achieved through a modified protocol for the PER, in which the stimulus is not associated with a sucrose reward, but to a mild electric shock (Carcaud et al. 2009). The PER and SER tests were used to reveal the modulating role of some pheromones on worker appetitive and aversive learning. QMP and Queen Attraction

Vergoz et al. (2007a) demonstrated that while OA mediates appetitive learning, as already shown by other authors, aversive learning in honey bees is mediated by DA in fact it is suppressed by blocking of DA, but not OA, receptors. Since it has been demonstrated that HVA can mimic DA function, Vergoz et al. (2007a) postulated that QMP, through its component HVA, is responsible for blocking aversive learning in young workers. This hypothesis was proved in a further study (Vergoz et al. 2007b), which showed that QMP does block aversive learning in young bees while leaving appetitive learning intact. The authors postulate that QMP production by the mated queen gives her an advantage by preventing young workers, which are in close contact with her and on which she depends for feeding, form an aversion to her pheromonal bouquet.

During their studies on appetitive and aversive behavior, Vergoz et al. (2009) observed that worker responsiveness to QMP is strongly age-dependent, since 2-day-old workers are more strongly attracted to QMP than 6-day-old ones, while foragers are even repelled by QMP. They also showed that this behavior is likely to be modulated by receptors in honey bee antennae: those of 2-day-old workers strongly attracted by QMP have a higher expression level of OA receptor Amoa1 and of DA receptor Amdop3 compared to 6-day-old workers the level of Amdop3 transcript decreases during the first week of adult life, together with the attraction towards QMP. However, this pattern is true only for bees that have been exposed to QMP since adult emergence, while young bees that have not been exposed previously to QMP are not attracted to it and show a higher expression level of the DA receptor Amdop1. Thus it seems that the queen possesses several ways to modulate worker behavior through QMP at the level of the antennal sensory neurons: by suppressing avoidance behavior (by blocking the DA signal) and by enhancing the attractiveness of her pheromone (by increasing the OA signal). This is supported by the fact that high expression levels of OA receptor gene Amoa1 and DA receptor Amdop3 in the antennae augment the attractive qualities of QMP, while suppression of DA receptor Amdop1 also enhances attraction to QMP by reducing worker sensitivity to unattractive components of the pheromone.

Similar results were found by McQuillan et al. (2012), who analyzed OA and DA antennal receptors in workers of different age and task commitment. The expression levels of the receptors Amoa1 and Amdop2 show an increase with age, being higher in older workers, while the opposite trend is shown for Amdop3 expression levels, which clearly decrease with age. Furthermore, expression levels of Amoa1 are higher in same-age pollen foragers than in nurses, consistently with the higher OA brain level in foragers (Schulz and Robinson 1999 Wagener-Hulme et al. 1999). Although the physiological significance of this variability in receptor gene expression has not been fully determined, the dynamics of gene expression in the antennae are indicative of a functional role of the periphery in the behavioral changes of honey bee workers. Alarm Pheromones and Defense Behavior

Several studies have shown that alarm pheromones, besides their important role in triggering bee defense behavior, can act as modulators of the sensitivity to environmental stimuli.

Stress-induced analgesia is a mechanism that increases the threshold of responsiveness to external stimuli that elicit innate defensive responses by activating endogenous opioid pathways. In honey bees the threshold of stinging response (the main defense behavior) was artificially increased with injection of morphine, and this effect was antagonized by naloxone, demonstrating the presence of an endogenous opioid system in the honey bee and its involvement in the modulation of the stinging response (Nu༞z et al. 1983). The exposure of workers to IPA causes a reduction in the responsiveness to a nociceptive stimulus (electrical stimulation) by increasing the threshold of responsiveness. This effect is antagonized by naloxone, indicating the involvement of an opioid system, as a typical opioid analgesia is induced. The social meaning of this analgesic effect is to increase the worker defensive efficiency by reducing the probability of withdrawal when facing an enemy (Nu༞z et al. 1998).

In the experiments of Balderrama et al. (2002) IPA exposure led to a decrease in responsiveness to sucrose and an increase of responsiveness to a noxious stimulus (i.e., an electric shock). In a followup study the exposure to alarm pheromones or IPA showed a strong effect on appetitive learning by decreasing the learning success of exposed bees (Urlacher et al. 2010). These effects are not in contrast with the main hypothesized role of alarm pheromones, as the depression of foraging activity, through the decrease in sucrose responsiveness and the appetitive learning, allows workers to freely contrast a potential danger or enemies signaled by the release of alarm pheromones. This can strengthen worker’s commitment to their role in colony guarding and defense.

The physiological mechanism subtending this modulating effect could involve biogenic amines, which are known to regulate aversive and appetitive learning, respectively, through DA and OA pathways (Giurfa 2007 Vergoz et al. 2007a). Alternatively, the activation of an opioid-like system, which was shown to be affected by this pheromone, could lead to a general learning impairment for its analgesic effects (Nu༞z et al. 1998). Modulation of Worker Metabolism: The Effect of Pheromones on Nutrient Stores

We saw that two pheromones have a prevalent role in the regulation of worker development by slowing the worker transition from nurse to foragers: QMP and brood pheromone. In the previous sections we showed that QMP acts through a central or peripheral modulation of brain amines, which influences the subsequent behavioral and physiological pathways, including the reception level of the pheromone itself. Moreover, OA modulates worker responsiveness to brood pheromone by regulating worker response thresholds.

Another way of action of these two pheromones seems to be the regulation of worker metabolism. Nurse bees have higher lipid stores than foragers and isolated worker bees have lower lipid levels than bees kept in a colony, regardless of food availability (Toth and Robinson 2005 Toth et al. 2005) thus pheromones may partly exert their effects by regulating workers’ nutrient storage. Moreover, among worker proteins, vitellogenin (Vg), an egg yolk protein, is produced in higher levels by the fat bodies of nurse bees than forager bees (Fluri et al. 1982) and thus can serve as a molecular marker for the nurselike physiological state.

In an experiment by Fischer and Grozinger (2008), the administration of QMP on caged workers increased protein and Vg level in the fat bodies. According to the authors, this effect could be achieved by behavioral, physiological, or molecular mechanisms: QMP modulates feeding behavior, inducing treated bees to consume more food or to reduce activity it decreases level of JH, which is known to increase metabolism (Sullivan et al. 2003), and this reduction in turn increases Vg levels and potential lipid storage finally, it can modulate metabolic pathways through regulation of the genes involved in the insulin signaling pathway, which is associated with nutrient storage (Fischer and Grozinger 2008).

A confirmation of the role of pheromones in regulating worker metabolism comes from the researches of Smedal et al. (2009), which demonstrated that BP also regulates Vg accumulation in the fat body. Beside its role in oogenesis Vg is utilized by workers for food production and is involved in the regulation of foraging behavior and the enhancing of worker lifespan, possibly by scavenging free radicals and enhancing honey bee immunity (Amdam et al. 2003, 2004, 2005 Nelson et al. 2007 Seehuus et al. 2006). Exposure to synthetic BP blend causes an increase in the amount of Vg in the fat bodies of young bees (3𠄴 days old) and a decrease in older workers (23� days old). This is consistent with the results of Pankiw et al. (2008), who showed that brood pheromone stimulates pollen consumption, leading to an increase of protein content in hypopharyngeal glands, but also showed that workers of different ages are affected in an opposite manner by the pheromone, confirming the differential perception of pheromones according to worker age and task. In this case brood pheromone acts on young workers by enhancing their capacity to produce brood food and to store a surplus from Vg synthesis, and on older workers by inhibiting an extensive Vg storage, ensuring that more protein remains free in the hemolymph to be converted into brood food (Smedal et al. 2009).

5.2.3. F rom S ignal to B ehavior: P heromones and G ene E xpression

Pheromone processing in the bee brain leads to neurophysiological changes that result in the production of a specific behavior or to changes in sensory thresholds that result in altered behavior under different contexts. In either case, the molecular mechanisms by which pheromones are transduced in the brain to influence behavior are only beginning to be understood. A great breakthrough was made with the completion of the honey bee genome (Honey Bee Genome Sequencing Consortium 2006) and the development of a genome-wide honey bee microarray, which enabled to search for differences associated with variation in responsiveness to pheromones.

A number of authors found that worker division of labor is based, in addition to the already mentioned age and environmental factors, on genetic differences among workers performing different tasks. Thus the probability of performing a particular task within a specific age caste would be determined not only by the endogenous and exogenous environment, but also by the genotype of the worker (Calderone and Page 1988 Frumhoff and Baker 1988). These genetic differences could influence, for example, the probability of a worker to become a guard, a nectar forager, a pollen forager, or a nest-site scout (Robinson and Page 1988, 1989).

The natural variation in honey bee pheromone response, observed by several authors (Pankiw et al. 1994) may be potentially adaptive, because it creates variability in task performance that supports colony plasticity and thus productivity. Kocher et al. (2010) found variability in worker attraction to QMP and consequently in the retinue response of adult workers, which appears to be associated with brain gene expression patterns and linked to the reproductive potential in honey bees. The authors found 960 gene transcripts that are differentially expressed between high and low responder workers, and a negative correlation between individual retinue response and ovariole number, a trait strongly linked to reproductive potential (Makert et al. 2006). This indicates that workers with the highest reproductive potential (e.g., the greatest number of ovarioles) avoid the queen, while those with lower reproductive potential are attracted to her. Under queenless condition workers with high reproductive potential would activate their ovaries, whereas the ones with low reproductive potential would be in charge of rearing a new queen (Kocher et al. 2010). This would confirm the observations by Moritz et al. (2002) that in A.m. capensis, workers that are likely to become reproductively active are indeed more likely to avoid the queen.

One way that a pheromone can influence behavior is by orchestrating large-scale changes in brain gene expression. In recent years several authors demonstrated that a differential gene expression exists between workers performing different tasks (Whitfield et al. 2003, 2006) and that exposure of honey bee workers to pheromones causes changes in brain gene expression that are associated with downstream changes in behavior. Therefore, it should be possible to investigate the mode of action of pheromones by correlating the changes in gene expression and the resulting behavioral expression. The first attempt in this direction was made by Grozinger et al. (2003) with QMP. Insights into the Pheromone-Mediated Genetic Mechanism Underlying Worker Behavioral Development

We know that QMP has a delaying effect on the transition from hive tasks to foraging in workers. Several genes have been identified as correlated to nursing or foraging conditions (Whitfield et al. 2003) and the exposure of young honey bee workers to QMP was found to activate genes associated with nursing and to repress genes associated with foragers. In the study by Grozinger et al. (2003) the gene that was more robustly and chronically regulated was found to be an ortholog of the Drosophila transcription factor krüppel homolog 1 (Kr-h1). This gene encodes for a zinc finger transcription factor that plays an important role in orchestrating development and cell differentiation. Although the different components of QMP taken individually were thought to elicit limited responses, two of them, 9-HDA and 9-ODA, were both able to produce a strong QMP-like gene activation. In particular, they were able to downregulate expression of Kr-h1, suggesting that 9-ODA and 9-HDA are the QMP components that influence the timing of the transition from hive work to foraging (Grozinger et al. 2007).

From all the reported observations about the role of pheromones in modulating worker behavior, it is interesting to investigate the functional relation between QMP, which regulates the transition from nurse to forager, and OA and JH, which have levels with strong correlation to these behavioral stages. Grozinger and Robinson (2007) studied the effects of these three factors on the modulation of the gene Kr-h1. JH analog, methoprene, or OA are unable, alone, to modulate Kr-h1 expression, demonstrating that these molecules do not have a direct influence on the gene expression. Conversely, methoprene, but not OA, significant reduces the effect of QMP on Kr-h1 brain expression in young bees, suggesting that high JH titers, typical of foragers, prevent downregulation of Kr-h1 expression by QMP in older bees (Grozinger and Robinson 2007). The authors’ interpretation is that QMP affects workers’ transition to foragers partly via JH regulation, since the pheromone is able to lower JH levels, and JH levels in turn modulate pheromone response, but other mechanisms must be involved, since a JH analog is not able to affect gene (namely Kr-h1) expression.

Together with QMP, BP is responsible for the regulation of worker behavioral development, delaying the transition of workers from nurses to foragers. Its way of action seems to be even more complex than QMP, since it has a dose- and age-dependent effect, and in addition to a primer effect on behavioral maturation, it acts as a releaser, stimulating the foraging activity of older bees that are competent to forage (Pankiw 2004c Pankiw and Page 2001).

Alaux et al. (2009) showed that BP effect on foraging ontogeny is linked to a variation in gene expression, since BP treatment upregulates brain genes that are highly expressed in workers specialized in brood care, and downregulates genes that are highly expressed in foragers. According to its age-related effect, the exposure to BP for 5 days caused a brain gene expression profile similar to the profile of nurse bees, while this similarity to nurse bees was absent in bees exposed to BP for 15 days. In fact, although there was a significant overlap between the gene sets controlled by BP in young and old bees, many were regulated in opposite directions. For example, the gene malvolio (mvl), which is activated in precocious foragers (Ben-Shahar et al. 2004), was upregulated by BP in 15-day-old bees, but not in 5-day-old bees, suggesting that mvl represents a key component of the regulation of foraging behavior by BP. This differential effect on brain gene expression of 15-day-old bees is consistent with the role of BP as releaser pheromone, triggering foraging behavior in older bees (Alaux et al. 2009).

Comparing these results with those obtained by Grozinger et al. (2003) with QMP, which exerts a similar effect on behavioral development, it emerges that some genes are regulated by both BP and QMP, probably because of the different chemical composition of the two pheromones, which are also found to use different peripheral receptors (Robertson and Wanner 2006 Wanner et al. 2007). Alarm Pheromone and the Expression of Immediate Early Genes

It has been demonstrated that primer pheromones exert their effects partly by causing changes in brain gene expression (Alaux et al. 2009 Grozinger et al. 2003). Releaser pheromones, which cause immediate and short-term responses, are thought to act through more direct neurophysiologic modulation systems. Today, the study of the mode of action of these two kinds of pheromones has changed this rigid distinction. For example, the exposure of honey bee colonies to IPA, originally classified as a releaser pheromone, caused a significant increase in expression of the gene c-Jun (Alaux and Robinson 2007). C-Jun belongs to the group of immediate early genes (IEGs), which are activated transiently and rapidly in response to a wide variety of stimuli. They are activated at the transcription level before any new proteins are synthesized and are known as early regulators of cell growth and differentiation signals, but are also involved in synaptic plasticity. The correlation between IPA exposure and c-Jun expression in honey bees blurs the long-standing distinction between primer and releaser pheromones and highlights the importance of brain gene expression in social regulation (Robinson et al. 2005).


D -galactose treatment, lifespan analysis and MDA measurement

The mean lifespan of laboratory-caged foragers decreased (P<0.05, Mantel–Cox test) when their food was supplemented with 10% d -galactose (Fig. 1A). The median survival time for control bees and bees fed d -galactose was 12 and 7 days, respectively. The experiment was discontinued after the death of all bees fed d -galactose. At this point, 50% of the control bees were still alive. Survivorship between the two groups was similar during the first 5 days of the experiment, but sharply declined in bees fed d -galactose after this point.

Galactose treatment decreases lifespan and increases malondialdehyde (MDA) content. (A) Circles represent forager bees fed 50% sucrose (w/v) supplemented with 10% d -galactose. Squares represent forager bees fed only 50% sucrose (w/v). MDA was quantified by using the thiobarbituricacid reactive substances (TBARS) method in (B) forager heads, (C) forager thoraces, (D) nurse heads and (E) nurse thoraces. White bars represent 15- to 17-day-old bees fed 10% d -galactose+50% sucrose or 50% sucrose for 5 days post capture. Gray bars represent 15- to 17-day-old nurse bees (D,E) or forager bees that were either given free flight or were restricted from taking flights (B,C). Bars (means±s.e.m., n=5–7 per bar) not connected by the same letters are statistically different from one another.

Galactose treatment decreases lifespan and increases malondialdehyde (MDA) content. (A) Circles represent forager bees fed 50% sucrose (w/v) supplemented with 10% d -galactose. Squares represent forager bees fed only 50% sucrose (w/v). MDA was quantified by using the thiobarbituricacid reactive substances (TBARS) method in (B) forager heads, (C) forager thoraces, (D) nurse heads and (E) nurse thoraces. White bars represent 15- to 17-day-old bees fed 10% d -galactose+50% sucrose or 50% sucrose for 5 days post capture. Gray bars represent 15- to 17-day-old nurse bees (D,E) or forager bees that were either given free flight or were restricted from taking flights (B,C). Bars (means±s.e.m., n=5–7 per bar) not connected by the same letters are statistically different from one another.

Comparing lab-held versus hive-collected bees suggests that the caging procedure increases stress as evidenced by increased MDA. Feeding d -galactose to laboratory-housed bees increased MDA in all tissues, and sucrose-fed bees showed a moderate increase in MDA relative to bees housed in a hive (Fig. 1B–E). This pattern suggests an effect of caging alone that increases MDA, which is then further increased by d -galactose feeding. Precocious foragers captured and fed d -galactose in the laboratory showed higher MDA levels in the head than counterparts fed only sucrose. These bees fed d -galactose also displayed higher MDA levels than foragers caged at 10 days of age (i.e. flight restricted) or allowed to forage freely since the initiation of foraging.

DNA oxidative damage and lipid peroxidation

In general, larger differences in 8-OHdG accumulation were observed when comparing high amounts of flight (>14 days) to low amounts of flight (<3 days), whereas incremental increases in flight showed little difference (Fig. 2A,B). In flight muscle, 8-OHdG increased in foragers with the most flight experience compared with all other groups, suggesting that intense flight increases DNA oxidative damage in this tissue.

Flight activity and age are associated with increased oxidative damage. 8-Hydroxy-2′-deoxyguanosine (8-OHdG) was quantified using ELISA in forager (A) brain tissue and (B) flight muscle. In A and B, white bars represent foragers that had unrestricted access to flight and gray bars represent foragers that were restricted from taking flights after 3 days. The x-axis describes ages (DO: days old) and flight experience [<3, 7–9 or >14 days flight (Flt) experiences]. The x-axis represents each group's respective age at the time of collection. MDA was quantified by ELISA in the (C) brain tissue and (D) flight muscle of 10-, 20- and 40-day-old foragers (white bars) and nurses (gray bars). Bars (means±s.e.m., n=5–7 per bar) not connected by the same letter are statistically different from one another.

Flight activity and age are associated with increased oxidative damage. 8-Hydroxy-2′-deoxyguanosine (8-OHdG) was quantified using ELISA in forager (A) brain tissue and (B) flight muscle. In A and B, white bars represent foragers that had unrestricted access to flight and gray bars represent foragers that were restricted from taking flights after 3 days. The x-axis describes ages (DO: days old) and flight experience [<3, 7–9 or >14 days flight (Flt) experiences]. The x-axis represents each group's respective age at the time of collection. MDA was quantified by ELISA in the (C) brain tissue and (D) flight muscle of 10-, 20- and 40-day-old foragers (white bars) and nurses (gray bars). Bars (means±s.e.m., n=5–7 per bar) not connected by the same letter are statistically different from one another.

When comparing hive-collected foragers and nurse bees, the pattern of lipid peroxidation is both tissue- and behavior-specific. In foragers, brain MDA levels were independent of age, but flight muscle MDA levels significantly increased in 40-day-old individuals (Fig. 2C,D). Alternatively, in nurses, brain MDA increased with age, but flight muscle MDA did not (Fig. 2C,D).

ROS: levels of H2O2, ●OH and SO −

Patterns of H2O2 and ●OH accumulation were tissue- and behavior-specific. Levels of H2O2 and ●OH increased in the thoraces, and additional flights led to further accumulation. In forager heads and thoraces, levels of H2O2 and ●OH increased with the number of flights taken (Fig. 3). Restricting forager flight increased H2O2 and ●OH levels in heads (Fig. 3A) but decreased H2O2 and ●OH levels in thoraces (Fig. 3B). The presence or absence of flight caused the largest effect in superoxide (SO – ) accumulation. In forager brain tissue, superoxide levels were lowest in middle-aged (15–17 days old) flight-restricted foragers (Fig. 4A). In forager thoraces, levels of superoxide were independent of flight activity (Fig. 4B).

Reactive oxygen species (ROS) accumulate with flight activity. Using a probe [2′,7′-dichlorodihydrofluorescein diacetate (CM-H2DCFDA)] that reacts with hydroxyl radicals (●OH) and hydrogen peroxide (H2O2), ROS accumulation associated with flight was measured in the heads (A) and thoraces (B) of 9- to 11-, 15- to 17-, or 24- to 26-day-old (DO) foragers that were allowed free access to flight [white bars: <3, 7–9 or >14 days flight (Flt) experience] or restricted to the hive (gray bars: <3 days flight experience). Values are represented as percentages of positive controls (100 μmol l −1 hypoxanthine, 5 mU ml −1 xanthine oxidase and 0.2 U ml −1 horseradish peroxidase). Bars (means±s.e.m., n=5–7 per bar) not connected by the same lowercase letter are significantly different (P<0.05, mixed model ANOVA and Tukey's HSD).

Reactive oxygen species (ROS) accumulate with flight activity. Using a probe [2′,7′-dichlorodihydrofluorescein diacetate (CM-H2DCFDA)] that reacts with hydroxyl radicals (●OH) and hydrogen peroxide (H2O2), ROS accumulation associated with flight was measured in the heads (A) and thoraces (B) of 9- to 11-, 15- to 17-, or 24- to 26-day-old (DO) foragers that were allowed free access to flight [white bars: <3, 7–9 or >14 days flight (Flt) experience] or restricted to the hive (gray bars: <3 days flight experience). Values are represented as percentages of positive controls (100 μmol l −1 hypoxanthine, 5 mU ml −1 xanthine oxidase and 0.2 U ml −1 horseradish peroxidase). Bars (means±s.e.m., n=5–7 per bar) not connected by the same lowercase letter are significantly different (P<0.05, mixed model ANOVA and Tukey's HSD).

Superoxide accumulates in the brain and flight muscle. Superoxide accumulation (measured with MitoSOX) associated with flight was measured in the heads (A) and thoraces (B) of 9- to 11-, 15- to 17-, or 24- to 26-day-old (DO) foragers that were allowed free access to flight [white bars: <3, 7–9 or >14 days flight (Flt) experience] or restricted to the hive (gray bars: <3 days flight experience). Values are represented as percentages of positive controls (100 μmol l −1 hypoxanthine, 5 mU ml −1 xanthine oxidase). Bars (means±s.e.m., n=5–7 per bar) not connected by the same lowercase letter are significantly different (P<0.05, mixed model ANOVA and Tukey's HSD).

Superoxide accumulates in the brain and flight muscle. Superoxide accumulation (measured with MitoSOX) associated with flight was measured in the heads (A) and thoraces (B) of 9- to 11-, 15- to 17-, or 24- to 26-day-old (DO) foragers that were allowed free access to flight [white bars: <3, 7–9 or >14 days flight (Flt) experience] or restricted to the hive (gray bars: <3 days flight experience). Values are represented as percentages of positive controls (100 μmol l −1 hypoxanthine, 5 mU ml −1 xanthine oxidase). Bars (means±s.e.m., n=5–7 per bar) not connected by the same lowercase letter are significantly different (P<0.05, mixed model ANOVA and Tukey's HSD).

GPDH activity

Generally, GPDH activity was dependent on flight activity. The lowest GPDH levels occurred in bees with the least flights taken (nurses and flight-restricted foragers) (Fig. 5). GPDH levels increased in aged foragers compared with aged nurse bees (Fig. 5C,D).

Glycerol-3-phosphate dehydrogenase (GPDH) activity is dependent on flight activity. To assess GPDH levels associated with flight, GPDH enzyme activity was measured in the heads (A) and thoraces (B) of 9- to 11-, 15- to 17-, or 24- to 26-day-old (DO) foragers that were allowed free access to flight [white bars: <3, 7–9, or >14 days flight (Flt) experience] or restricted to the hive (gray bars: <3 days flight experience). GPDH activity associated with age and behavioral differences in heads (C) and thoraces (D) was measured in forager bees (gray bars: 9–11, 15–17 or 24–26 days old) and nurse bees (white bars: 9–11, 15–17 or 24–26 days old). Values are represented as GPDH activity in U mg −1 . Bars (means±s.e.m., n=5–7 per bar) not connected by the same lowercase letter are significantly different (P<0.05, mixed model ANOVA and Tukey's HSD).

Glycerol-3-phosphate dehydrogenase (GPDH) activity is dependent on flight activity. To assess GPDH levels associated with flight, GPDH enzyme activity was measured in the heads (A) and thoraces (B) of 9- to 11-, 15- to 17-, or 24- to 26-day-old (DO) foragers that were allowed free access to flight [white bars: <3, 7–9, or >14 days flight (Flt) experience] or restricted to the hive (gray bars: <3 days flight experience). GPDH activity associated with age and behavioral differences in heads (C) and thoraces (D) was measured in forager bees (gray bars: 9–11, 15–17 or 24–26 days old) and nurse bees (white bars: 9–11, 15–17 or 24–26 days old). Values are represented as GPDH activity in U mg −1 . Bars (means±s.e.m., n=5–7 per bar) not connected by the same lowercase letter are significantly different (P<0.05, mixed model ANOVA and Tukey's HSD).

Catalase and SOD activity

Although levels of H2O2 and ●OH increased with flights taken, a similar increase in catalase activity was absent. In forager brains, catalase activity increased in middle-aged foragers (15–17 days old) relative to flight-restricted foragers of the same age (Fig. 6A). In flight muscle, catalase activity was independent of age and flight experience (Fig. 6B).

Catalase (CAT) activity is independent of age and behavior. To assess CAT levels associated with flight, catalase enzyme activity was measured in the heads (A) and thoraces (B) of 9- to 11-, 15- to 17-, or 24- to 26-day-old (DO) foragers that were allowed free access to flight [white bars: <3, 7–9 or >14 days flight (Flt) experience] or restricted to the hive (gray bars: <3 days flight experience). Values are represented as nmol formaldehyde produced min −1 ml −1 . Bars (means±s.e.m., n=5–7 per bar) not connected by the same lowercase letter are significantly different (P<0.05, mixed model ANOVA and Tukey's HSD).

Catalase (CAT) activity is independent of age and behavior. To assess CAT levels associated with flight, catalase enzyme activity was measured in the heads (A) and thoraces (B) of 9- to 11-, 15- to 17-, or 24- to 26-day-old (DO) foragers that were allowed free access to flight [white bars: <3, 7–9 or >14 days flight (Flt) experience] or restricted to the hive (gray bars: <3 days flight experience). Values are represented as nmol formaldehyde produced min −1 ml −1 . Bars (means±s.e.m., n=5–7 per bar) not connected by the same lowercase letter are significantly different (P<0.05, mixed model ANOVA and Tukey's HSD).

SOD activity was both tissue and behavior dependent. In forager brains, SOD activity was independent of flight experience (Fig. 7A). In forager flight muscle, SOD activity decreased in experienced foragers (<14 days) relative to their caged counterparts restricted from flight (Fig. 7B).

Superoxide dismutase (SOD) activity is independent of age and behavior. To access SOD levels associated with flight, SOD enzyme activity was measured in the heads (A) and thoraces (B) of 9- to 11-, 15- to 17-, or 24- to 26-day-old (DO) foragers that were allowed free access to flight [white bars: <3, 7–9 or >14 days flight (Flt) experience] or restricted to the hive (gray bars: <3 days flight experience). Values are represented as SOD activity in U ml −1 . Bars (means±s.e.m., n=5–7 per bar) not connected by the same lowercase letter are significantly different (P<0.05, mixed model ANOVA and Tukey's HSD).

Superoxide dismutase (SOD) activity is independent of age and behavior. To access SOD levels associated with flight, SOD enzyme activity was measured in the heads (A) and thoraces (B) of 9- to 11-, 15- to 17-, or 24- to 26-day-old (DO) foragers that were allowed free access to flight [white bars: <3, 7–9 or >14 days flight (Flt) experience] or restricted to the hive (gray bars: <3 days flight experience). Values are represented as SOD activity in U ml −1 . Bars (means±s.e.m., n=5–7 per bar) not connected by the same lowercase letter are significantly different (P<0.05, mixed model ANOVA and Tukey's HSD).

Instinct vs. Intelligence

These words were written 11 centuries ago, after Byzantine emperor Constantine VII ordered scholars to compile centuries of information on farming, going back to the Greeks and Romans. The resulting work, Geoponika: Farm Work, has recently been translated for the modern reader.

Modern day researchers also recognize the European honey bee is one of the world’s most sophisticated and interesting creatures. Through a combination of individual and colony behavior, it can learn, adapt, communicate, divide up its labor, utilize senses of direction, sight, and smell, help propagate its colony, and engage in skills far in excess of its relative brain capacity. A bee has one million neurons, compared to a human’s 100 billion neurons. Research has shown that they have the ability to learn from their experiences in spite of their brain size. Inquiring minds want to know, “Are their skills hard-wired, i.e. instinctual, or are they capable of learning and abstract thought that borders on intelligence?” If we interpret their actions correctly, we may see a level of intelligence and creativity far in excess of their brain size and their place on our planet as a lowly insect.

Primitive Consciousness

Andrew Barron and Colin Klein from Macquarie University in Sydney, Australia, propose that insects have the capacity for consciousness, or awareness, due to the “intricate circuitry” of their brains. It’s doubtful that they fly around contemplating the meaning of life, but they do have the capacity to feel something, even on a primitive level.

Instinct vs. Intelligence is a dichotomy that we simple bee fans will continue to disagree on till the cows come home. The experts don’t agree which is the dominant force, so don’t expect to find a conclusive verdict here. You will have to decide for yourself, or like most of us, just embrace the mystery of it.

We can enjoy tending to our bees on a regular basis, even though we do not live in a research lab or even give a hoot about the scientific method. We see the consistent patterns of brood, the way they tend for their young, guard their home against predators, bring home the bacon as hard as it may be to find sometimes, build comb, clean their hives, turn nectar into liquid gold through cooperation with other bees, prepare for Winter even though they have not lived long enough to know what Winter is, and honor her royal highness. How could they be anything but smart? At this point, some of you may wish to shout, “On the contrary, you’ve just described the definition of honey bee instinct!”

So let’s look more at the issue of instinct to understand honey bee behavior. In other words, what are they pre-programmed to do? The female workers’ tasks change throughout their short life, as nurse bees, house bees, scout bees, guard bees, and foragers. We know that the queen’s pheromones have a great influence on their behavior, but where do we draw the line between instinct, queen influence, and learned intelligence? Research is conducted along very specific, narrow lines, to weed out variables that could skew results. The cold eye of science can be very accurate and insightful, but can also ignore the wonderment of nature. To me, this is what makes beekeeping so fascinating. I don’t view my bees with the detachment of a scientist, but with the awe of a fan. I view their abilities, whether intelligence or instinct, with curiosity, appreciative that they share their existence, and gifts of honey and wax with us.

The argument for instinct

Wikipedia describes instinct as “the inherent inclination of a living organism towards a particular complex behavior.” They cite nineteenth century French entomologist Jean-Henri Fabre, who described instinct as behavior that does not require cognition or consciousness.

Does that describe honey bee behavior, or is there more to it? In The Hive and the Honey Bee, (1992 edition), Norman Gary wrote the chapter titled, “Activity and Behavior of Honey Bees.” He makes a strong argument that the behavior of honey bees is hard-wired, or instinctual. The bees unconsciously see a need within the hive and respond to it.

He cautions us against the tendency to “anthropomorphize” our understanding of bees, meaning to attribute human feelings, activities, and characteristics to non-human beings. Anthropomorphizing is a game we often engage in, for example when we think of a certain mouse who wears clothes, sings, walks on two legs, has a girlfriend named Daisy, and a dog named Pluto. However, there is a difference between anthropomorphizing for fun and trying to understand true honey bee behavior.

As an example, Gary says we typically think from a human standpoint that bees forage intensively to build up their honey reserves for the long, cold Winter that lies ahead. But with a life span of six weeks, a foraging bee wouldn’t know anything about the upcoming Winter. And most of them wouldn’t have to worry about it anyway because they’d be dead before then.

There are many words that we ascribe to bees, in an attempt to make sense of their behavior from a human point of view. They have “duties,” “responsibilities,” and “division of labor.” We say they are “angry” when they sting, “clever” when they build comb, and “ambitious” when they work overtime.

Gary says their activities are not nearly as brilliant as they seem to be. Nor do they involve a master plan for the betterment of themselves or their colony. Each bee responds to a stimulus nearby, without awareness of the big picture.

Their age, genetic composition, internal and external factors, sensitivity to stimulus around them, and chemical and physiological changes that develop with age, are key factors that influence their activities.

Virtually all insect behavior is genetically programmed from the moment an egg is fertilized. They are basically biological robots, performing activities that favor the survival of the colony. Each bee’s response at any given moment is dependent on its condition. A bee near a cell containing larva responds differently than a bee near a queen. She might sense the stimulus of a hungry larva, and reflexively feed it. A bee carrying a particle of wax would be stimulated by a cell that needs to be capped. The more it responds to a particular stimulus, the more it learns and duplicates that behavior next time.

Do they know what they are doing? In all probability no, says Gary, even though their activity appears highly organized. All of their activities can be accounted for by reflexive or mechanistic responses. Thus, another anthropomorphic term such as “division of labor,” is not applicable, because it’s not done on purpose. As they age, their activities become more varied because of physical ability and learning. He summarizes his thoughts on the subject with, “We must conclude that intuitive thinking or cognition, akin to that of humans, is probably non-existent in bees, or at least inconsequential.”

The Argument for Intelligence

The Byzantine Geoponika goes on to explain why scholars from the Middle Ages feel bees are intelligent:

“… its works is truly divine and of the greatest use to mankind. Its social life resembles that of the best regulated cities. In their excursions, bees follow a leader and obey instructions …”

Their appreciation of bees is impressive, and it is enlightening to know this knowledge existed in ancient times. But is advanced evolutionary behavior by insects proof of consciousness?

Jumping ahead several centuries again, researchers at Australian National University stated their research shows that “honey bees have a remarkably robust and flexible working memory, in spite of having a very small brain, and much fewer neural connections than the average vertebrate. … It even hints at the emergence of a primitive intelligence from a small brain.”

We know that honey bee skills include color recognition, including ultraviolet light, and a strong sense of smell. They utilize these physiological traits to learn which colors, and which odors, are most rewarding for finding nectar. Among their other cognitive skills are the use of abstract thought and symbolic language, to distinguish landscape and the way home after foraging for distances of perhaps several miles. Learning through experience is a form of intelligence.

One of the greatest discoveries of abstract thinking among honey bees goes to Karl von Frisch, an Austrian who won the Nobel Prize in physiology and medicine in 1973. He and his team unlocked the key to the Waggle Dance, the elaborate pattern of movements in which bees communicate to others in their hive the location of such things as a good source of nectar, or a place to build a new hive after a swarm. He discovered that the “dance” (another anthropomorphic term), includes time and distance to a source, relative to the position of the sun, including any obstacles that may be in the way of their travel. And all this is communicated in the dark!

Other researchers have argued that visual dance isn’t significant because of the dark. It may be the odor of the plants that lingers on the foraging bee after returning from the field that contributes to finding the source of the nectar. And still other researchers claim that sounds from the dancing bees are the key. In any event, the mystery and magic of the waggle dance are an ingenious form of communication.

Another school of thought has become popular lately that says consciousness lies not with the individual bees, but with the colony as a whole. Proponents use the term “superorganism.” The Merriam Webster Dictionary defines superorganism as, “A complex structure of interdependent and subordinate elements whose relations and properties are largely determined by their function of the whole.” Using this definition, the behavior of the colony dictates the behavior of the individual bees. Colonies are a complex structure of interdependent and subordinate elements–the bees. Social animals like bees cannot live alone, and their behavior only makes sense when viewed as colony consciousness. Through their collective behavior and short life span, they keep the colony alive and it is the colony which determines behavior. Jamie Ellis from the University of Florida writes, “Put a group of honey bees together and you get behaviors, attributes, characteristics, etc. that are otherwise absent in individual honey bees.” A colony is more than the sum of its parts.

Another study was conducted with bumble bees (alas not honey bees), led by Lars Chittka from London’s Queen Mary University. They showed that bees could learn to pull on a piece of string to release nectar. When new bees watched experienced bees pull the string, the new bees learned at a more rapid rate, successfully pulling the string 60 percent of the time.

In a recent issue of Science Magazine, some of the same researchers described how they trained bumble bees to move a ball to a target location to receive a treat. Chittka summarized, “…there is more and more evidence, both from experiments on small-brained insects and computational neuroscience, that small circuits can deal with exceptionally complex challenges.” This creative problem solving shows how new behaviors can develop as the need arises.

New Technology Unlocks Clues

So we started out with a strictly instinctual, pre-programmed, robotic explanation of bee behavior. Then we mixed in learning from experience, communication through the waggle dance, and expanded it through superorganism behavior. Do these factors themselves conclude honey bee intelligence?

Newer technologies have enabled a level of research that was never available in the days of von Frisch. For example, today micro-electronic devices can be placed on bees to track their movement, and miniature video cameras can see inside dark hives.

In 2010, Thomas Seeley of Cornell University published his book Honeybee Democracy, which follows his adventures tackling honey bee communication and intelligence. In one experiment that he conducted on an isolated island, several swarms of bees had to choose their new home site from among several empty hives that he set up in different areas. Every box except one had problems as a home site, whether it was capacity, location, size of opening, etc. Each swarm had approximately 70 scout bees searching for their new home. The scouts advocated for different new homes based on their waggle dance. Through a process of elimination, persuasion, and promoting the advantages of their preferred choice, the swarms ultimately selected their new homes. They made the correct decision nine times out of 10 trials. With electronic trackers, Seeley was able to display graphically how the scouts walked through the interior of each hive they visited, checking out its suitability. He proved through group decision-making and brain power how the swarm scouts ultimately reached their final decisions. Quite a remarkable feat on the part of the bees.

The arguments (and emotions) for instinct vs. intelligence are strong on both sides. Like so many factors in life, it is not a black or white issue. Do honey bees in fact perform their acts of brilliance, such as the waggle dance or selecting the best home site, without thought? If so, then perhaps the definition of intelligence, including consciousness and creativity, needs to be re-evaluated.

Anthropologic descriptions of honey bees allow us to romanticize them perhaps more than they deserve, but is that so bad? We cannot help but smile at the ancients (in Geoponika) who wrote that bees are “pleased by a good tune,” “hate laziness,” demonstrate “mechanical skill and near-logical understanding,” and “honor their king.”

Honey bees’ placement on the consciousness scale of 1-100 will probably continue to move back and forth as new information comes to light. When and if we ever reach an ultimate conclusion regarding their range of instinct vs. intelligence, it will probably not change how we manage our six-legged “livestock,” or how much honey we manage to harvest. If anything, the ongoing mystery just adds to the fascination of these complicated, sometimes ornery, creatures of nature.


Our study challenges the theoretical expectation that selection should favor an optimal signal match between the mimics and their specific models, and that mimetic imperfection negatively affects the model's attractiveness to the dupes in a specialized mimicry system.

On the contrary, our bioassays show that the pollinator bees actively prefer the floral scent of the “imperfect” mimic, which can be explained by a sensory preference or “receiver bias” toward “novel” signals over more commonly encountered ones (17, 18). As shown in a previous study on C. cunicularius, females use population-specific ratios of their key sex pheromone compounds, and males are more attracted by sex pheromone “dialects” from allopatric populations than from their own (13). Because C. cunicularius usually forms dense “island” aggregations and both sexes can only travel across short distances for mates, foraging resources, or nesting sites, we suggest that populations are probably subjected to inbreeding and male preferences for novel signals might therefore be adaptive and promote outbreeding, e.g., by avoiding sibling mating should the opportunity arise. Such preferences for novel signals are regarded as a common emerging feature in animal cognitive processes, and studies have demonstrated that this phenomenon can be an important driving force behind signal evolution (19, 20). In our orchid-pollinator system, the pollinators' preferences can impose selection on floral odor of the orchids, because the orchids' reproductive success is primarily pollinator-limited (21 ⇓ –23). Specifically, results from a recent study support the hypothesis that the patterns of pollinator-attracting compounds of the orchids' floral odor are under pollinator-mediated selection (22). Pollinator-driven evolution in the orchids does not, however, automatically imply male-driven evolution in female bees. Females of many solitary bee species, including C. cunicularius, are thought to mate only once, and, unlike the orchids, the emerging female bees are presumably not limited in their reproductive success by access to males, because patrolling males typically outnumber the virgin female bees during their reproductive period. We therefore suggest that the mating preferences of C. cunicularius males are unlikely to drive a shift in the females' sex pheromone (13), whereas the pollinators' preference for novel signals (Figs. 5 and 6) have the potential to drive the evolution of the orchids' floral scent (22). Therefore, we hypothesize that this marked difference in the selective pressures acting on the signals of the models versus the mimics has fueled the evolutionary divergence in pollinator-attracting odor signals, which, over time, has resulted in the imperfect chemical mimicry observed between the orchid mimics and their sympatric model species (Figs. 1, 2, 3B, and 4).

Attempts at explaining the evolution of imperfect mimicry have until now mostly involved the “multimodel hypothesis,” postulating that a partial resemblance to multiple models can be beneficial to a mimic (24). In sexually deceptive orchids, however, most species specifically mimic one single model taxon (4, 5). Accordingly, in O. exaltata, C. cunicularius males are the only known pollinator species. Alternatively, several authors (25, 26) have advocated that what appear to be “poor” mimics to human senses might in fact be accurate mimics for the operator's perception. Again, this hypothesis is not relevant for our study, as we compared only the patterns of biologically active odor compounds of orchids and bees that have been shown to attract the males (14). Finally, physiological constraints such as the costs of signal production have been proposed as a mechanism that might impede further improvement of mimic-model resemblance (27). Constraints are, however, also unlikely to play an important role in signal evolution of Ophrys, because the production of the attractive signal, the cuticular wax compounds, and changes in relative rather than absolute amounts of a bouquet imply relatively low costs (10).

Our study reports on an alternative mechanism of pollinators selecting for signal divergence between model and mimic in a specialized floral mimicry system. Because such mechanisms may be more widespread, our study calls for a re-evaluation of the role of signal similarity between mimics and their model species, by paying particular attention to sensory and behavioral ecology of the operators. Such research will lead to a better understanding of patterns of signal evolution in mimicry, especially in cases where imperfect mimicry is actually adaptive.

Materials and Methods


All odorants were purchased from Sigma–Aldrich, J&K, or Chemsky, or were synthesized in-house. Deuterium incorporation into compounds 14, 8, and 9 was accomplished by methods reported in the literature from undeuterated or deuterated commercially available starting materials, as described in SI Appendix, with full characterization of all compounds, following purification by chromatography and recrystallization to a constant melting point (when possible), by 1 H and 13 C NMR, IR spectroscopy, and GC-MS. Spectra and GC-MS traces are included in SI Appendix. The chemicals were dissolved in DMSO or ethanol and diluted further into working concentrations before experiments.

Heterologous Expression of ORs.

A HEK 293T-derived Hana3A cell line was grown in Minimum Essential Medium (HyClone) containing 10% (vol/vol) FBS at 37 °C with 5% (vol/vol) CO2. Lipofectamine 2000 (Invitrogen) was used for transfection. Luciferase assays were performed as previously described. After 18–24 h, OR, the accessory OR protein, mRTP1S, and constructs for firefly luciferase and Renilla luciferase expression were transfected into cells. Twenty-four hours after transfection, the cells were stimulated with odorants [plus 30 μM Cu 2+ ions when the ligands were MTMT-, bis(methylthiomethyl) disulfide, and dimethyl sulfide-dissolved in CD293 (Invitrogen)]. We used the Dual-Glo Luciferase Assay System (Promega) and followed the manufacturer’s instructions for measuring chemiluminescence.

Statistical Analyses.

One-way ANOVA or an unpaired Student’s t test was used to compare the 95% confidence interval logEC50 values among isotopomers for each receptor/odorant pair in Figs. 3 and 4. The level of significance was *P < 0.05. An F test was used to compare the best-fit values of EC50, Hill slope, and top of the dose–response curves between the original hydrogenated odorant and its isotopomers in Figs. 3 and 4 and SI Appendix, Fig. S3.2. Bonferroni correction was applied to the F tests to account for multiple comparisons. The level of significance was *P < 0.00076 before correction and *P < 0.05 after correction.

The authors declare that they have no conflict of interest.

T.C. and M.A. involved in the conceptualization. T.C. and M.A. involved in the methodology. C.S. and T.C. involved in the investigation. M.A. contributed to the resources. T.C. wrote the original manuscript. T.C., C.S. and M.A. wrote, reviewed, and edited the manuscript. T.C. and C.S. involved in the visualization.

Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Evolution Midterm

Environmental variation is based on the differences produced when same genotypes are exposed to different environments. This usually happens because certain environments might alter gene expression. For example one genotype in a species of plants, often produces different sizes of mature plants, when grown in the different altitudes or with different soil nutrients. Similarly, dark pigmentation on the tips of ears and paws of Siamese cats develops in certain latitudes with colder temperatures. Many additional examples of traits that could vary a great deal under environmental changes are found in quantitative traits.

An insertion or a deletion would cause a frame-shift. For example, a deletion in DNA: TGACTAACGGCT, leaves mRNA sequence as: ACGAUUGCCGA and the polypeptide would be:
Threonine-Isoleucine-Cysteine. Duplication/insertion of a single base also causes a frameshift, since each codon has three bases. This also affects the rest of the sequence.

If the polypeptide change remains the same due to a change where DNA has one base-pair change, yet same amino acid remains in its place, we use the term synonymous substitution. For example ACU codes for Threonine, but if there is a silent substitution of ACU to ACC, ACG or ACA, the amino acid will remain in its place. The protein will function the same, with such mutation. A silent site mutation does not change the amino acid specified by a codon a replacement mutation does.

If there is a substitution of one amino acid to another, due to the base-pair change in DNA and RNA, we use the term non-synonymous substitution.

Since we only have 10 individuals (genotypes) in the population, we cannot make good conclusions about the advantageous genotypes. If we had a larger sample, in the similar ratios: 6Aa: 3aa: 1AA, we could say that heterozygotes are the most advantageous.

The unequal number of chromosomes in the gametes from polyploidy species and gametes from normal diploid plants would probably result in the reproductive isolation.

On the contrary, large scale mutations (chromosomal structure changes), such as inversions, duplications, translocations and deletions affect larger portions off genome and they have more drastic effects on phenotypes, as well as on fitness. In addition, they could be associated with divergent populations.

This hypothesis is also logical, because all of the lineages in the Old World have two pigment genes on the X chromosome, while all of the lineages in the New World have only one.

B) Migration followed by natural selection. The frequency of the T allele may increase, but only slowly, and perhaps not at all, due to the rarity of the T allele and the weakness of selection.

C) Frequency-dependent selection. The frequency of small males is likely to gravitate toward a stable equilibrium frequency, at which small and large males have identical fitness.

D) Underdominance. One allele will very likely go to fixation, and the other allele will be lost. Which allele is lost is likely to depend on where the initial allele frequencies are, relative to the unstable equilibrium point.

B) Decrease in CF allele frequency, due to reduced heterozygote superiority.

C) Decrease in CF allele frequency, due to (voluntarily) decreased reproductive success.

B) Natural selection. The frequency of the Aat120 allele declined over several generations toward the original frequency seen in the intertidal population, likely due to selective differences in the two environments.

C) Genetic drift. Random events can cause random deviations in allele frequencies from the expected course of evolution. This effect is stronger in small populations.

Migration: Greater effects in small populations, because each new migrant represents a greater proportion of total population size

Genetic drift: Greater effects in small populations

Inbreeding: Greater effects in small populations

New mutations per individual: Similar in populations of all sizes

New mutations per generation in the whole population: Fewer in small populations

New mutations per year in the whole population: Fewer in small populations

Probability that a new mutation will be effectively neutral: Similar in populations of all sizes

The frequency of any particular allele at one locus does not vary with the alleles present at the other locus,

The frequency of any chromosome haplotype can be calculated by multiplying the frequencies of the separate alleles, and:

B) The chestnut allele will probably decrease because half of the chestnut alleles in this population are linked to the R allele, which carries a selective disadvantage (RR foals die). However, random genetic drift might interfere with this process 20 horses is a small population, particularly considering only some of them can be mares. If the population were even smaller (e.g., 10 horses), genetic drift would have even greater effects.

B) In birds, the W chromosome never has an opportunity to exchange genes with any other W chromosomes. Like the Y in mammals, it is vulnerable to Muller's ratchet and selective sweeps, and has accumulated a large number of deleterious mutations. The same phenomenon has occurred in virtually all organisms that have sex chromosomes (except see below).

B) Yes, because different levels of childhood maltreatment (an environmental condition) are associated with different levels of antisocial behavior. If there were no effect of environment, both lines would have a slope of 0 (parallel to the x-axis).

C) Men with different genotypes respond in the same direction: for men in both MAOA categories, increased maltreatment is associated with heightened antisocial behavior. However, the strength of this effect is different in the two groups. Men with the low MAOA activity genotype appear more strongly influenced by environment than are men with the high MAOA activity genotype. The line for men with low MAOA activity has a greater (steeper) slope than the line for men with high MAOA activity, indicating a stronger effect of the environment factor on the x-axis.

D) Yes. This means simply that some of the variation in antisocial behavior is attributable to genotype.

B) No. Like most quantitative traits, neuroticism is almost certainly influenced by many genes, only some of which have been discovered, and only some of which vary in genotype. (Variation in serotonin transporter genotype explains only about 5% of variation in neuroticism. Many other QTLs associated with neuroticism have since been found.)

Directional selection occurs when a value to one side of the population average (higher or lower, but not both) has highest fitness. This trims one tail off the population distribution and expands the other tail, shifting the population mean. Variation tends to reduce (because one tail is trimmed off) but not very much (because the other tail tends to lengthen, depending on available genetic variation).

An observational study is one in which researchers simply observe the patterns that occur in nature, such as Huey et al.'s study of rock selection in garter snakes. (Sometimes, observational studies may compare two groups of animals that occur in nature. However, the researchers do not assign individuals to the different groups rather, the individuals have "assigned themselves" to the different groups, which can introduce considerable confounds.)

A comparative study is one that compares different taxa of organisms, often studying the distribution of a trait on a phylogeny, and seeking to understand why some clades evolved the trait and others did not. Examples include Hosken's study of testis size in bats, and Futuyma et al.'s study of host shifts in leaf beetles. A comparative approach is very useful when many taxa have evolved a similar trait.

b. Only by cutting wings were they able to test the effects of wing-markings and wing-waving independently.

A second prediction is that differences in temperature and rainfall do not cause the difference in leaf toxins. This could be tested by growing Island A and Island B shrubs under controlled conditions mimicking those of both islands. If, under both sets of conditions, Island A shrubs continue to produce toxins and Island B shrubs do not, we can eliminate the hypothesis that environmental conditions themselves do not lead to toxin production.

A third prediction is that herbivorous insects are not the selective force favoring toxin production. This could be tested by performing tests of the preference of insects to either Island A or Island B shrubs in a controlled setting. If insects exhibit no preference, we can eliminate them as a selective force.

b. Douglas fir trees: Height is likely selected for because the advantage it confers in competition for sunlight but it may be selected against because of the energy investment required to build and maintain the "extra" tissues required to grow very tall, particularly the impressive investment in woody support tissue and water- and food-transporting tissue.

c. Termite gut symbionts: Cellulose-digesting organisms are favored because of the nutrients they provide the termites. However, a high abundance of these microorganisms may be selected against because they limit the amount of food a termite can store, or because they use much of the energy of the cellulose for their own metabolism, leaving the termite with less energy.

d. Maple trees: Loss of leaves in fall can be favored because otherwise, leaves may be destroyed (i.e., by heavy snowfall) or frozen. However, the complete loss of leaves requires an enormous investment in new leaf growth every spring. (Note that conifers have a different strategy - they retain their leaves, or needles. But each needle is very skinny and small, presumably to reduce risk of damage during snowfall, and does not offer much photosynthetic leaf area during summertime.)

e. Male moths: Large antennae may be favored because they make individuals more sensitive to female pheromone, allowing them to find mates more efficiently. They might be selected against because of the energy demands they represent, or because they are fragile and easily damaged.

Population genetic methods for assessing colony densities

During the reproductive season (spring–autumn), honey bee colonies produce large numbers (500+) of males (drones). When they are about 2 weeks old, males commence daily mating flights. Large numbers of males from many colonies gather at drone congregation areas (DCAs Loper et al., 1992 Koeniger & Koeniger, 2000 Galindo-Cardona et al., 2012 ). The time of mating flights and the location of the congregation areas are species-specific (Koeniger & Wijayagunasekera, 1976 Koeniger et al., 1988 Rinderer et al., 1993b Hadisoesilo & Otis, 1996 Koeniger & Koeniger, 2000 Otis et al., 2000 Oldroyd & Wongsiri, 2006 ). Mating takes place on the wing. Typically, a queen mates on one or two afternoons in her life, with 10–30 males on each occasion (Palmer & Oldroyd, 2000 ). Males are attracted to a queen by her shape, movement, and the sex pheromone she secretes from her mandibular glands, which has 9-oxo-2-decanoic acid (9-ODA) as a major component (Butler et al., 1962 Gary, 1962 ). Drones fly to DCAs along flyways that follow major features in the landscape such as treelines (Loper et al., 1987 , 1992 ).

Aspects of this reproductive biology can be exploited to obtain estimates of colony density. Males can be sampled from an area either by an aerial trap baited with a pheromone lure (Kraus et al., 2005b Moritz et al., 2007 Jaffé et al., 2010 Arundel et al., 2012 Hinson et al., 2015 ) (see the Sampling methods section below and Fig. 1) or by sampling the worker progeny of queens that were allowed to mate at the site of interest (Jaffé et al., 2010 Arundel et al., 2014 ). In both cases, rather than searching for colonies, colonies are identified by inferring the minimum number of colonies that could generate the observed genotypes of the sampled males.