Chemical Communication and Reproduction Partitioning in Social Wasps
- 284 Downloads
Social wasps encompass species displaying diverse social organization regarding colony cycle, nest foundation, caste differences (from none to significant dimorphism) and number of reproductive queens. Current phylogenetic data suggests that sociality occured independently in the subfamily Stenogastrinae and in the Polistinae+Vespinae clade. In most species, including those with the simplest social organization, colony reproduction is monopolised by a single or few females. Since their nest mates can also develop ovaries and lay eggs, dominant females must somehow inhibit them from reproducing. Physical interactions in the form of open aggression or, usually, ritualised dominance by the fertile females contribute to fertility inhibition in several species, but it is unlikely to function in large colonies. In the latter case, reproduction within the colony is likely to be regulated through pheromones. Relatively little is known about these semiochemicals. Studies on all the three social wasp subfamilies, revealed that cuticular hydrocarbon components differ in abundance between egg-laying and not egg-laying females and that their composition depends on fertility status. In several species, females have been reported to manifestly react towards females with activated ovaries, but there is little evidence to support the hypothesis that fertile individuals are either recognized through their CHC composition, or that over-represented CHC constituents can inhibit fertility. Moreover, very little information exists on the possibility that exocrine glands release fertility signals or chemicals inhibiting fertility.
KeywordsReproduction partitioning Queen pheromone Oophagy Cuticular hydrocarbons Dufour gland Social parasites
Social wasps comprise a broad group of genera and species with very diverse social organizations, colony size and colony cycle. Since the first comprehensive review on pheromones and chemical communication in social wasps (Downing 1991), only a limited number of studies have investigated volatiles from exocrine glands and their role in communication (Bruschini et al. 2010). On the contrary, several researches have considered the role of cuticular hydrocarbons (CHCs) as recognition pheromones in different social contexts (Dani 2006; Kather and Martin 2015; Lorenzi 2006). The literature on CHCs in the context of dominance hierarchies in females of Polistes species has recently been reviewed (Jandt et al. 2014). Here we analyze the current knowledge on chemical communication in reproductive partitioning in social wasps and discuss the limits of the current state of art, addressing possible directions for future research.
An Introduction to Social Wasp Societies
Social wasps embrace about 1085 species, divided into three subfamilies, Stenogastrinae, Vespinae and Polistinae (Pickett and Carpenter 2010), the last of which is by far the largest and most diverse in social organization. Phylogenetic relationships among the Vespidae have been the subject of several studies, the results of which have impacted the hypothesis of social evolution within this clade. The most recent studies (Bank et al. 2017; Peters et al. 2017) confirmed the hypothesis of a common origin in the Polistinae and Vespinae as reported by Hines and co-workers (2007), distinct from that of the Stenogastrinae. This phylogeny contrasts the first proposal by Carpenter (1991) and later supported by Pickett and Carpenter (2010), whereby the Stenogastrinae is considered a sister group of the Polistinae + Vespinae clade. According to this latter hypothesis, sociality evolved only once in the Vespidae, whereas the discrete origin of Stenogastrinae implies that sociality evolved at least twice. Species belonging to the other subfamilies of the family Vespidae (Eumeninae, Masarinae and Euparaginae) that show some communal nesting behavior are not considered in the present review, since, to the best of our knowledge, no studies on chemical communication between temporarily associated females have been reported for these species.
Stenogastrinae, commonly named hover wasps, comprise 7 genera and 59 species whose distribution is limited to the South Oriental and Papuan Region (Turillazzi 2012). All build small nests with mud or vegetal material differing remarkably in shape between species and extraordinarily camouflaged against the vegetation. Colonies consist of two to twelve morphologically undifferentiated females with one, generally the oldest, laying most of the eggs. In Liostenogaster flavolineata and Parischnogaster mellyi, for instance, colonies are typically large, but in Anischnogaster and Metischnogaster they consist of only one or two females (Turillazzi 2012). Nest foundation can occur throughout the year and colonies are started by a single or, more rarely, more females. Usurpation and adoption of orphan nests have been reported as alternative strategies (see Turillazzi 2012). Dense aggregations of conspecific nests are common in L. flavolineata, L. vechtii and P. alternata and the females may often move from one colony to another (Samuel 1987; Turillazzi et al. 1997).
Although physical dominance interactions between females are infrequent, subordinates tend to avoid the dominant female when she walks over the nest. In physical interactions, the dominant females antennate their subordinates, and solicit them to offer a trophallactic drop of liquid. Openly aggressive behavior can also occur between nestmates, but it is more often directed towards foreign wasps attempting to land on the nest (Turillazzi 2012). A linear hierarchy has been reported between colony members of L. flavolineata; the dominant female has more developed ovaries and spends more time on the nest, while subordinates leave more frequently to forage. Subordinate females in several species can also develop ovaries, albeit to a lesser extent than the dominant, can be fertilized (Turillazzi 2012) and thus are ready to lay eggs. A high level of female nestmate relatedness has been found in L. flavolineata (Sumner et al. 2002) and P. alternate, where the oldest female monopolizes colony reproduction. The subordinates, however, may obtain indirect fitness by helping to raise offspring whilst waiting to inherit the nest (Bolton et al. 2006).
In contrast to the Stenogastrinae, in most Vespinae species the fertile females (often referred to as gynes in the pre-foundation phase and queens in the colonial phase) are generally larger than workers and caste dimorphism may exist for some characters (Jeanne and Suryanarayanan 2011; O’Donnell 1998). Since gyne larvae are reared in larger cells and receive more food than worker larvae, nutrition probably acts as a proximate factor for pre-imaginal caste determination, although environmental and social factors may also be involved (Jeanne and Suryanarayanan 2011).
Vespine nests share a common pattern: individual combs are built each attached to the previous one to form a stack surrounded by an envelope with access to the exterior via one or more opening. The holoarctic and oriental region distribution of these wasps (Pickett and Carpenter 2010) only partially mirrors the nest-founding modalities. Regardless of their distribution, most species form annual colonies which are founded by a single queen. The main exception is species of the tropical genus Provespa (Matsuura 1991) where founding is through swarming, and some perennial colonies of Vespula species in tropical or subtropical areas (Greene 1991).
Queens are reported to physically dominate workers in the small-colony species of Dolichovespula and Vespula rufa (Greene 1991). This does not occur in other vespine species where queens are often surrounded by workers touching them with their antennae and mouth parts, as in honeybees (Matsuura 1991). Workers in the queen’s retinue could assess queen fertility, probably by checking for pheromonal secretions, a behavior which could lay at the basis of reproduction partitioning.
Although reproduction is exclusive to queens, workers may lay eggs in orphaned nests and contribute to male production in queen-led colonies. A relatively high worker contribution has been reported for some Dolichovespula species (Foster et al. 2001), whereas male production is monopolized by queens in Vespa crabro and Vespula vulgaris (Foster et al. 2000; Foster and Ratnieks 2001). In the last two mentioned species, fertile workers are rare in queen-led colonies and eggs laid by workers are removed by nestmate workers. A similar control over fellow worker reproduction, referred to as policing, is also known in other social insects, including the honeybee (Ratnieks and Visscher 1989). In the smaller colonies of most Dolichovespula species and Vespula rufa, the queen destroys worker-laid eggs (Wenseleers et al. 2005). Colony kin structure has been suggested as an ultimate factor influencing queen-worker conflicts over male production. According to this hypothesis, high relatedness due to single mating by the queen favors worker reproduction, relaxes worker policing and leads to a more rigid queen policing. On the contrary, moderate kinship, due to multiple queen matings, e.g. several Dolichovespula species, favors male production by the queen and active worker policing (Foster et al. 2001; Wenseleers et al. 2005). However, a more recent study on D. saxonica (Bonckaert et al. 2011) found that the percent of worker-born males increases with colony development, therefore worker reproduction may depend more on colony stage and size rather than intracolonial relatedness, for which reliable proximate cues may be lacking. Moreover, worker-laid eggs appeared more frequently in those areas of the nest which are least visited by the queen (Bonckaert et al. 2011).
Polistinae, the largest subfamily (26 genera and 958 species, according to Pickett and Carpenter 2010) comprises species with very diverse social organization and colony foundation modalities. Four different tribes have been recognized, with Ropalidini limited to the Oriental and African Region, Mischocyttarini to the New World, Epiponini inhabiting the tropical and subtropical areas of the New World and Polistini being cosmopolitan.
Polistini, often referred to as paper wasps, only include the ubiquitous genus Polistes (206 species). Some Polistes species, and in particular P. dominula (P. dominulus and P. gallicus in earlier literature) are considered model organisms in sociobiology. Leo Pardi was the first to describe the inhibitory effect of physical dominance on fertility in Polistes females cooperating in nest foundation (Pardi 1946), later recognized as a widespread mechanism in small colony social insects. Frequency and intensity of dominance interactions vary between Polistes species. Frequent dominance is generally associated with evident task and reproduction partitioning (high reproductive skew), while lower dominance is associated with a more equitable reproduction sharing among associated females (Reeve 1991).
Foundation of colonies by one or more potentially fertile female, whose number can vary from two to several dozen, occurs in all species of Polistes and Myschocyttarus studied so far, three out of the four Ropalidini genera, i.e. Belonogaster, Parapolybia and several species in the genus Ropalidia (Gadagkar 1991). Such species are often referred to as independent-founding; dominance behavior and reproductive skew among associated females have been reported for species belonging to all five genera (Gadagkar 1991, 1996; Turillazzi 1996).
In contrast, the other species of Ropalidia, the entire Polybioides genus and species of all genera in the Epiponini tribe found their nests by swarming, where swarms consist of groups of workers accompanying, generally, more than one gyne. All swarm-founding Polistinae occur in tropical areas (Jeanne 1991). Swarming species generally form larger colonies than those founding independently. Colonies of different genera and even species of the same genus vary considerably in size (Jeanne 1991).
Despite the clear reproductive specialization in queens, caste dimorphism is not common to all swarming species. While some species exhibit remarkable differences in size and allometry, with generally larger queens, in others no differences exist and newly emerged females can be reproductively flexible, depending on the state of the colony (Jeanne 1991; Noll et al. 2004). Aggressive interactions between egg-laying females are rare, but in certain species, workers can challenge and even kill some queens, consequently reducing the number of matrilines in the colony (West-Eberhard 1978). In some species, workers can also contribute to male production (Noll and Wenzel 2008).
Queen Control and/or Honest Fertility Signals
Reproductive partitioning in social animals implies that non-reproductive individuals have either reduced direct fitness or do not reproduce at all. In social hymenopteran species where workers, at least for some time in their lives, can develop their ovaries and reproduce (mainly by laying unfertilized eggs), proximate factors such as behavior, signals or cues by queens may reduce worker fertility. Two partially alternative hypotheses have been proposed (see Smith and Liebig 2017 for a recent review): (i) queens can control co-foundresses or worker fertility (ii) queens emit signals or bear features correlated to their fertility so co-foundresses and/or workers can either limit their own fertility or reproduce following the strategy that maximizes their own fitness.
In both cases, proximate factors can differ in nature, although control or signaling based on direct physical interactions or requiring spatial proximity are unlikely to be efficient in large-sized colonies.
In some Polistes species, individuals permanently exhibit different color patterns and it has been demonstrated that features and facial color can be cues appraised by wasps during dominance interactions (Tibbetts and Dale 2004). However, social signals are unlikely to act as visual stimuli in species where most interactions occur in the dark, e.g. those nesting underground or whose nests are surrounded by envelopes typical of all Vespinae and several swarming Polistinae.
Physical interactions by egg-laying females manifesting domination and requests for trophallactic fluids from subordinates and workers are common in small colonies, for example some Stenogastrinae, independent-founding Polistinae, some swarming Polistinae and Dolichovespula. They are, however, rare in species which form large colonies.
Vibratory movements can be observed in several wasp species and have mostly been studied in Polistes where they are performed more by females of high hierarchic status (Brennan 2007). Despite the association with dominance, such behavior is considered to be a signal directed to larvae rather than to other adult females (Brennan 2007).
Among Polistes dominula cofoundresses, dominance behavior is often performed by the larger females with more developed ovaries, although body size is not an absolute predictor of hierarchical status (Turillazzi and Pardi 1977). Pioneer studies by Röseler and co-workers (1984; 1985) discovered that even before hierarchy has been established, size of the corpora allata differs between foundresses, where these endocrine glands are smaller in females of low hierarchic status. In adult females, the juvenile hormone (JH) stimulates the gonads to produce ecdysteroids, which in turn activate vitellogenin production. JH is therefore associated with both ovarian development and dominance behavior. In insects, JH and ecdysteroid production are stimulated by two pathways: target-of-rapamycin (TOR) and insulin/insulin-like signaling (IIS), whose activation is influenced by nutritional conditions (see Kapheim 2017).
Dominant alpha foundresses still retain dominance and other behavioral traits after ovariectomy, although they can no longer inhibit reproduction in other females (Röseler and Röseler 1989). Despite the link between fertility and behavior, this implies that even in small colonies dominant behaviour by alpha females cannot by itself inhibit fertility. Using the same species, P. dominula, Tibbetts and Izzo (2009) found that supplementing the JH analogue methoprene stimulates ovary development and dominance behaviour more in larger foundresses with facial markings signaling high competition quality than in smaller foundresses with no high quality markings. Permanent visual features could thus denote fertility or fertility potentiality, contributing to fertility signaling. However, a more recent study revealed no differences in visual facial markings between females of different hierarchical and fertility status (Kelstrup et al. 2017).
Although P. dominula represents a good model to study the effect JH has on dominance behavior and ovarian development, results differ in both independent and swarm-founding Polistinae. For example, Kelstrup et al. (2017) found no link between JH titre, dominance and ovarian development in foundresses of P. smithii. In the epiponine Synoeca surinama, castes are not dimorphic; young females in queen-less colonies can develop ovaries and eventually become fertile. Established fertile females dominate these individuals by approaching them and bending their gaster. JH titre in dominant individuals were not high (Kelstrup et al. 2014b).
As reported in the literature and discussed below, fertility is often reflected in the composition of cuticular hydrocarbons (CHCs). Studies on solitary and social insects have found that the relative abundance of CHC components is influenced by gonadotropins (see below), making it unlikely that the CHC profile acts as a dishonest fertility signal. However, Van Oystaeyen et al. (2014) assaying Vespula vulgaris, the bumblebee Bombus terrestris and the ant Cataglyphis iberica, found that treating queen-less colonies with the most abundant single CHC component in queens lowered the odds of ovarian development in workers. Fertility inhibition by a semiochemical in the absence of a queen suggests pheromonal control. Similar experiments on B. impatiens (Amsalem et al. 2015), however, found no evidence that the application of the single most abundant hydrocarbons in queens modulated worker reproduction.
Only a few semiochemicals of non-hydrocarbon nature have been reported to be secreted by queens and workers in social wasps (see below). No studies so far have investigated whether these semiochemicals are affected by gonadotropins and represent honest signals.
Cuticular Hydrocarbons and Fertility Signals
The insect cuticle is covered by a complex mixture of long-chained, low volatile lipids of which hydrocarbons are the most important constituents in number of compounds and abundance (Howard and Blomquist 2005). The link between fertility and the composition of cuticular hydrocarbons (CHCs) was first reported in non-social insects. In the house-fly Mosca domestica, the long chain alkene muscalure ((Z)-9-tricosene) acts as a sexual pheromone and is found on the cuticle of females with developed ovaries, together with the corresponding C23 epoxide and a ketone of the same chain length ((Z)-14-tricosene-10-one) (Blomquist et al. 1984). Synthesis of this alkene is stimulated by ecdysteroids and especially ovarian ecdysone. Treatment with ecdysteroids inhibits elongation of fatty acids, favoring synthesis of (Z)-9-tricosene in spite of (Z)-9-heptacosene, while methoprene favors the release of muscalure on the cuticle. Thus, hormones involved in ovarian development can affect both the synthesis and transport of CHCs. Moreover, production of muscalure is higher in sugar fed flies than protein fed controls, indicating that diet can also influence pheromone biosynthesis (Adams et al. 1995).
The composition of epicuticular hydrocarbon mixtures varies greatly between insect species, but the majority of components are often identical or differ only in chain length, since the biosynthetic pathways of these compounds is common to all insects (Blomquist 2010). Therefore, the components of social wasp CHCs are the same as those reported for several other insect species. CHC synthesis occurs in oenocytes and fat bodies and involves a network of fatty acid synthases (FAS), elongases, desaturases, reductases and a decarbonylase (Blomquist 2010). Different fatty acid synthases incorporating different Acyl-CoA are probably involved in the biosynthesis of linear or methyl branched alkanes as well as alkanes bearing methyl branching in a different position (Finck et al. 2016; Howard and Blomquist 2005). Insect genomes entail several genes of the fatty acid synthase and elongase families. In principle, species-specific CHC profiles result from mutation of these genes (Finck et al. 2016).
In social insects, large body of evidence shows that CHCs are chemical cues allowing nestmate recognition (Howard and Blomquist 2005). Lower variability in CHC mixtures at the colony level is probably due to similar expression of the genes involved in hydrocarbon biosynthesis in kin females, as well as permanence on a common nest, which, in the case of wasps, is impregnated with the same cuticular hydrocarbons (Dani 2006). Physical contact between nestmates and trophallaxis with HC-containing secretions, as observed in the postpharyngeal gland secretion in ants (Soroker et al. 1995), leads to homogenization of colony members.
Despite the uniformity of colony odor, some HC components differ in concentration in members of different castes and even subcastes in several species of social hymenopterans (see Howard and Blomquist 2005). How such intra-colony differences are compatible with a colonial CHC profile is not clear.
Although differences between colonies or castes can be analyzed from the entire CHC profile as well as single compounds through various statistical approaches, information is still limited regarding the compounds or series of homolog compounds relevant to nestmate recognition. In Formica exsecta, for example, ratios of the different compounds in the (Z)-9-alkenes homologue series were found to be colony specific and not affected by task-specific changes in the CHC mixture (Martin and Drijfhout 2009). On the contrary, in F. fusca and in F. argentea, positional isomers of methyl branched hydrocarbons of the same chain length are conserved within colonies and may provide sufficient information for nestmate recognition (Krasnec and Breed 2013; Martin et al. 2008).
Differences between the CHC profile of egg-laying foundresses and workers in social wasps were reported in the first studies on cuticle lipids, based on Polistes colonies, before they had been recognized in other social hymenopterans. In P. dominula, Bonavita-Cougourdan and co-workers (1991) found that dominant foundresses in monogynous colonies differ from their descendent daughters mostly in the concentration of some methyl branched alkanes. In polygynous colonies, however, variation in CHC composition among associated foundresses of different hierarchic status was weak soon after colony foundation, but stronger at worker emergence, when most alfa females had more developed ovaries than their subordinates. When dominant females were removed from the colonies, most beta females acquired an alfa-like profile (Sledge et al. 2001a). However, under experimental conditions where both alfa and beta females could lay eggs on the same nest and fully develop their ovaries, the beta females did not acquire the same CHC profile as alfa females (Dapporto et al. 2007), suggesting that hierarchy influences CHC profile as well as fertility. The same experiment revealed differences in the HCs covering eggs laid by alfas and betas (Dapporto et al. 2007). Egg hydrocarbons are likely to derive from the Dufour gland, which also contains the same HCs, but differing in relative abundance, with respect to cuticle, (Dani et al. 1996). However, the oocytes, dissected from ovarioles were already covered with HCs (Dani, pers. observation), indicating that the Dufour gland is not the only source of these compounds. Dufour gland secretion has been demonstrated to be involved in egg discrimination and differential oophagy by dominant females in P. fuscatus (Downing 1991). Since very few reports mention compounds other than HCs for this secretion (Dani 2006), HC profile seems to be the most probable recognition cue for differential oophagy.
In Polistes wasps, the abdominal sternal glands are an additional source of HCs and, together with the Dufour gland, are probably one of the main sources of the HCs which dominant females apply to the nest by vigorously stroking their abdomen over the surface (Dani et al. 1992). In experimental colonies of the monogynous species P. gallicus, where queens were prevented from interacting with workers, frequent stroking behaviour delayed worker ovarian development, suggesting that cues applied to the nest by the queens may indeed contribute to fertility inhibition (Dapporto et al. 2007). Very different results have been found for P. smithii, where CHCs were not found to be associated with social status or fertility (Kelstrup et al. 2017).
In the Ropalidini, cuticular and Dufour gland HCs have been extensively studied in Ropalidia marginata. In this species, unlike several Polistes, egg laying foundresses rarely dominate or interact aggressively with other females, but if there is no fertile foundress, some females (potential queens) may become very aggressive for a few days, until they develop ovaries and gain control over colony reproduction (Bhadra et al. 2010). As in Polistes, cuticular and Dufour gland HCs only differ in the relative amount of their components (Mitra and Gadagkar 2014). Both cuticular and Dufour gland HCs shift from a worker-like to queen-like profile in potential queens developing their ovaries (Mitra and Gadagkar 2012). Since Dufour gland secretion from established queens inhibits aggression in other females (Bhadra et al. 2010), potential queens seem to rely on open aggressive behaviours only until they can chemically signal their fertility (Mitra and Gadagkar 2012). Interestingly, queens transferred to a foreign colony will be attacked, but inhibition of aggressive behaviour by Dufour gland secretion is independent of the origin of the queen. The fertility signal conveyed by the gland therefore seems distinct from nestmate recognition pheromones (Mitra et al. 2011).
Only a few studies have analyzed the correlation between CHCs and fertility in swarming Polistinae. In the epiponine Polybia micans, where colonies are small and female reproductive fate depends strongly on the social environment, queens and workers exhibit remarkable differences in their CHC profiles, with two compounds (3-methyl pentacosane and n-pentacosane) being far more abundant in queens. These compounds increase in females during ovarian development and with high levels of juvenile hormone. These individuals are easily recognized and violently attacked by nestmates (Kelstrup et al. 2014a). Worker and queen profiles are also distinct in Synoeca septentrionalis (Santos et al. 2018) and Synoeca surinama (Kelstrup et al. 2014b). In the latter, females treated with the JH analogue methoprene develop ovaries and shift to a queen-like profile (Kelstrup et al. 2014b).
Differences in females depending on ovarian development have also been reported in four species of Stenogatrinae (Turillazzi et al. 2004), but the role of CHCs as a fertility signal has never been demonstrated. In all hover wasps, CHCs comprise far fewer compounds than in Polistes and alkenes, rather than methyl-alkanes, occur in several species (Cervo et al. 2002; Turillazzi et al. 2004). Moreover, long chain primary alcohols are also found on the cuticle of Liostenogaster flavolineata (Cervo et al. 2002). The relative abundance of several alkenes is often negatively correlated with ovarian development, while a positive correlation exists with paraffines, the only exception being L. vechti (Cervo et al. 2002). All hover wasps have very developed Dufour glands whose dense secretion serves as a substrate for rearing the larvae. Long chain hydrocarbons have also been reported for the Dufour gland (Keegans et al. 1993). In L. flavolineata, Dufour gland secretion and the secretion deposited as substrate both contain two long chained 2-alkoxyethanols which could act as emulsifying agents for deposited larval food (Keegans et al. 1993). Any possible role of the Dufour gland secretion as a fertility signal has not been investigated.
CHC composition differs among castes in several vespine species (Butts et al. 1991; Butts et al. 1995) and recent studies have questioned whether such differences could be at the basis of policing behaviors. Experiments on Vespula vulgaris by Van Oystaeyen et al. (2014) found that in queen-less colonies, workers treated with the most abundant compounds in queen CHCs, namely n-heptacosane, n-nonacosane and 3-methyl nonacosane, were less likely to develop ovaries than their counterparts in control colonies. Together with its homologue 3-methyl pentacosane, 3-methyl nonacosane is also more abundant on queen eggs; worker-laid eggs treated with this compound are less frequently removed by workers than untreated ones (Oi et al. 2015). Consequently, in this species at least one of the components of the sterility-inducing pheromone is also a recognition cue used in worker policing. Treating worker eggs with a blend of linear and branched hydrocarbons, found in higher amounts in queen CHCs, gave similar results in Dolichovespula saxonica (Oi et al. 2016).
Since alkanes have been found to be the most abundant component in queens of several species of social Hymenoptera, Van Oystaeyen et al. (2014) argue that an alkane fertility signal already existed in the solitary common ancestor of Aculeata and has been retained in most clades, while signals of a different chemical nature are derived states found in a few groups. Regarding Polistes species, differences in alkane constituents between foundresses or between foundresses and workers are not surprising, given that in all species so far analyzed methyl branched and n-alkanes far outnumber alkenes, found in only a few species (Dani 2006). In the vespine clade, however, where all species present olefins (Kather and Martin 2015), alkanes are over-represented in the queens of most species (Van Oystaeyen et al. 2014).
In conclusion, correlation between CHC composition and ovarian development has been demonstrated in all social wasps so far studied, including species with clear caste morphological differences. Moreover, the shift to a fertile or queen-like profile in females developing their ovaries indicates that CHC composition is a very plastic and labile trait, influenced by the hormones regulating fertility. Among the over-represented alkanes in fertile females, only a few (mainly n-nonacosane) are consistently more abundant in a number of wasp species (Van Oystaeyen et al. 2014), suggesting that hormonal fertility regulation may affect biosynthetic processes diversely, with a different output on the composition of cuticular and possibly glandular secretion hydrocarbons.
Several studies on CHCs as fertility signals have found that the shift towards a fertile profile is often associated with behavioral challenges by nestmates. Yet, since hormones could affect exocrine glands producing other compounds, the link between CHC profile and fertile female recognition by nestmates is merely correlative, and can only be confirmed through supplementation of specific compounds, as bioassays on some ant species have demonstrated (Smith et al. 2009; Smith et al. 2012).
Cuticular Hydrocarbons and Social Parasites
Wasps reproducing as obligate social parasites occur both in the genera Polistes and Dolichovespula. The former has been studied in depth to understand the strategies employed for integration into host colonies and exploitation of the work force (Cervo and Dani 1996; Lorenzi 2006). The female of the social parasite Polistes sulcifer enters the host colonies aggressively. Prior to usurpation, the CHC profile of parasite females is much simpler than that of their host, P. dominula. However, within a few days of usurpation, the parasite profile shifts to match that of the hosts (Turillazzi et al. 2000) acquiring, possibly through chemical mimicry, a colony specific pattern. This coincides with a marked fall in aggressive behaviour towards the host (Sledge et al. 2001b). Although the parasites vigorously stroke their abdomens over the nests, like dominant host foundresses, there is no change in nest lipid composition, apart from one dimethyl alkane which is highly concentrated in pre-usurpation parasite CHCs (Turillazzi et al. 2000). Further studies have shown that some days after usurpation, the parasites develop an alfa-like host CHC profile (Dapporto et al. 2004). Although the parasites monopolize most of the colony reproduction, control over ovarian development in their hosts is lower than in queen-right colonies (Cini et al. 2014), suggesting that matching the host fertility signal does not imply total reproductive control.
Exocrine Glands and Queen Pheromones
Queen retinue in vespine wasps is believed to be induced by queen-released pheromones, but the nature of these compounds has proved elusive. A lactone, namely δ-n-hexadecalactone has been isolated from the head of Vespa orientalis queens and reported to induce cell construction for future reproductive brood (Ikan et al. 1969; Ishay et al. 1965). No similar compounds have been reported for other species.
Very recently, two oxyacids, already described in the honey bee mandibular pheromone (4-oxo-octanoic acid and 4-oxo-decanoic acid) have been identified in the sternal gland secretion in virgin gynes of Vespa velutina, and shown to attract males under natural conditions (Wen et al. 2017). Since these oxyacids were not found in gynes after the mating season, they are unlikely to play a role as a queen pheromone. Interestingly these are the first compounds identified with a clear function of a sex pheromone in social wasps.
Reproductive differences among colony members is a key feature of eusociality. Proximate mechanisms by which individuals affect fertility of their colony members or regulate that of their companions is an interesting aspect of the biology of social insect species, where most females in the colony are potentially fertile. In social wasps, aggressive and ritualized behaviors by fertile females contribute to the formation of linear hierarchies in small colonies, where dominant females reproduce more than their subordinates. However, in most species studied so far, the composition of cuticular hydrocarbons differs between egg-laying and unfertile females, since some constituents are more abundant, although not exclusive, to fertile females. The source of cuticular hydrocarbons are enocytes, the fat body and the Dufour gland. Composition of cuticular hydrocarbons is influenced by hormones, juvenile hormone and ecdysteroids, which also induce the ovarioles and oocyte development. In other insects, these hormones affect specific pathways of hydrocarbon biosynthesis, but no studies on social hymenopterans have so far been performed.
Exocrine gland secretions have not been fully examined and no clear differences between egg-laying and unfertile females have been reported (see Bruschini et al. 2010). The only exception to this scenario is a δ-lactone extracted from the head of Vespa orientalis queens (Ikan et al. 1969; Ishay et al. 1965), whose effect on the colonies has not been investigated in depth.
Although egg-laying and unfertile females differ in cuticular hydrocarbon composition, only a few experiments have demonstrated the effect that the over-represented compounds have on colony reproductive partitioning. Nothing is known about how they are dispersed around the nest, nor perceived by individuals in large colonies. A primer effect of worker fertility inhibition and a releaser effect as an egg recognition pheromone, promoting worker policing over laid eggs, has only been reported in two vespine species (Oi et al. 2015; Van Oystaeyen et al. 2014).
Queen pheromone fertility signals and/or worker fertility control seems highly probable in large colony species where the rarity of interactions makes physical control difficult. Queen retinues, similar to those found in honey bees, have been reported for vespine species (Matsuura 1991) and could possibly be a way of spreading queen pheromones around the colony. No information on possible semiochemicals is available to date.
The disproportion between current knowledge on cuticular hydrocarbon composition and chemicals of a different nature reflects the relative ease of sample preparation and analysis of these mixtures, compared with the methods required for unknown constituents of glandular secretions. Yet a full understanding of the semiochemicals contributing to reproduction regulation in social wasp colonies cannot be reached without more facts on exocrine gland secretions. A deeper knowledge of the chemical composition and social function of the secretion of the most important exocrine glands at least is therefore mandatory. This implies a renewed approach to the problem, more detailed chemical studies and accurate bioassays to test the effect these secretions and their components have on behaviour and reproduction. Finally, mechanisms by which semiochemicals influence hormonal fertility regulation is a field still completely unexplored in the biology of all social insects, one which needs to be addressed by scientists interested in chemical ecology, neurobiology and insect physiology.
Authors are grateful to Dr. Christina Coster-Longmann for her help in revising the manuscript.
- Bank S, Sann M, Mayer C, Meusemann K, Donath A, Podsiadlowski L, Kozlov A, Petersen M, Krogmann L, Meier R, Rosa P, Schmitt T, Wurdack M, Liu S, Xin Z, Misof B, Peters RS, Niehuis O (2017) Transcriptome and target DNA enrichment sequence data provide new insights into the phylogeny of vespid wasps (Hymenoptera: Aculeata: Vespidae). Mol Phylogenet Evol 116:213–226CrossRefPubMedGoogle Scholar
- Bruschini C, Cervo R, Turillazzi S (2010) Pheromones in social wasps. In: Gerald L (ed) Pheromones. Academic Press, London, pp 447–492Google Scholar
- Butts DP, Camann MA, Espelie KE (1995) Workers and queens of the European hornet Vespa crabro L. have colony-specific cuticular hydrocarbon profile (Hymenoptera, Vespidae). Comp Biochem Physiol 42:45–55Google Scholar
- Butts DP, Espelie KE, Hermann HR (1991) Cuticular hydrocarbons of four species of social wasps in the subfamily Vespinae: Vespa crabro L., Dolichovespula maculata (L.), Vespula squamosa (Drury) and Vespula maculifrons (Buysson). Comp Biochem Physiol 99B:87–91Google Scholar
- Carpenter JM (1991) Phylogenetic relationships and the origin of social behavior in the Vespidae. In: Ross KG, Matthews RW (eds) The social biology of wasps. Comstock Publishing Associate, Ithaca and London, pp 7–32Google Scholar
- Cervo R, Dani FR (1996) Social parasitism and its evolution in Polistes. In: Turillazzi S, West-Eberhard MJ (eds). Oxford University Press, Oxford, pp 98–112Google Scholar
- Dani FR (2006) Cuticular lipids as semiochemicals in paper wasps and other social insects. Ann Zool Fenn 43:500–514Google Scholar
- Gadagkar R (1991) Belonogaster, Mischocyttarus, Parapolybia, and Indendent-founding Ropalidia. In: Ross KG, Matthews RW (eds) The social biology of wasps. Comstock Publishing Associate, Ithaca, pp 149–190Google Scholar
- Gadagkar R (1996) The evolution of eusociality, including a review of the social status of Ropalidia marginata. In: Turillazzi S, West-Eberhard MJ (eds). Oxford University Press, Oxford, pp 248–271Google Scholar
- Greene A (1991) Dolichovespula and Vespula. In: Ross KG, Matthews RW (eds) The social biology of wasps. Comstock Publishing Associate, Ithaca, pp 263–305Google Scholar
- Jandt JM, Tibbetts EA, Toth AL (2014) Polistes paper wasps: a model genus for the study of social dominance hierarchies. Insect Soc 61:11–27Google Scholar
- Jeanne RL (1991) The swarm-founding Polistinae. In: Ross KG, Matthews RW (eds) The social biology of wasps. Comstock Publishing Associate, Ithaca, pp 191–231Google Scholar
- Lorenzi MC (2006) The result of an arms race: the chemical strategies of Polistes social parasites. Ann Zool 43:550–563Google Scholar
- Matsuura (1991) Vespa and Provespa. In: Ross KG, Matthews RW (eds) The social biology of wasps. Comstock Publishing Associate, Ithaca and London, pp 232–262Google Scholar
- Mitra A, Gadagkar R (2014) The Dufour's gland and the cuticle in the social wasp Ropalidia marginata contain the same hydrocarbons in similar proportions. J Insect Sci:9Google Scholar
- Pardi L (1946) Ricerche sui Polistini VII. La “dominazione” ed il ciclo ovarico annuale di Polistes gallicus (L.). Boll Ist Entomol Univ Bologna 15:25–84Google Scholar
- Peters RS, Krogmann L, Mayer C, Donath A, Gunkel S, Meusemann K, Kozlov A, Podsiadlowski L, Petersen M, Lanfear R, Diez PA, Heraty J, Kjer KM, Klopfstein S, Meier R, Polidori C, Schmitt T, Liu S, Zhou X, Wappler T, Rust J, Misof B, Niehuis O (2017) Evolutionary history of the Hymenoptera. Curr Biol 27:1013–1018CrossRefPubMedGoogle Scholar
- Pickett KM, Carpenter JM (2010) Simultaneous analysis and the origin of eusociality in the Vespidae (Insecta: Hymenoptera). Arthropod Syst Phylogeny 68:3–33Google Scholar
- Reeve HK (1991) Polistes. In: Ross KG, Matthews RW (eds) The social biology of wasps. Comstock Publishing Associate, Ithaca, pp 99–148Google Scholar
- Röseler PF, Röseler I, Strambi A (1985) Role of ovaries and ecdysteroids in dominance hierarchy establishment among foundresses of the primitively social wasp, Polistes gallicus. Behav Ecol Sociobiol 18:9–13Google Scholar
- Samuel CT (1987) Factor affecting colony size in the stenogastrine wasp Liostenogaster flavolineata. University of Malaya, Kuala Lumpur, Ph.D. dissertationGoogle Scholar
- Santos EB, Shemilt S, de Carvalho CAL, Martin SJ (2018) Life history and chemical ecology of the warrior wasp Synoeca septentrionalis (Hymenoptera: Vespidae, Epiponini). PLoS One https://doi.org/10.1371/journal.pone.0194689
- Turillazzi S (1996) Polistes in perspective: comparative social biology and evolution in Belonogaster and Stenogastrinae. In: West-Eberhard MJ (ed) Turillazzi S. Oxford University Press, Oxford, pp 235–247Google Scholar
- Turillazzi S, Pardi L (1977) Body size and hierarchy in polygynic nests of Polistes gallicus (L.) (Hymenoptera, Vespidae). Monit Zool Ital (NS) 11:101–112Google Scholar