Optional gamergates in the queenright ponerine ant Gnamptogenys striatula Mayr

Summary: Ponerine ants display a number of social structures to which particular behaviours are associated. In the ponerine ant species Gnamptogenys striatula Mayr, queens occur and queenright colonies are functionally polygynous. However, some workers are capable to mate and to produce their own worker offspring. These gamergates appear several days after the queens are experimentally removed, allowing some workers to adopt a sexual calling posture inside the nest. At that time, other workers get outside the nest to collect males in the arena and carry them back into the colony. There, these males are groomed before they can mate with sexual calling workers. As for queens, several gamergates may coexist in a same colony. The social profile of gamergates is similar to those of nurses and they stay closer to the egg piles.     

Functional Monogyny and Dominance Hierarchy in the Queenless Ponerine Ant Pachycondyla (=Bothroponera) sp. in West Java,Indonesia (Hymenoptera,Formicidae, Ponerinae)

Social organization of the queenless ponerine ant Pachycondyla (Bothroponera) sp. was studied in Bogor, West Java, Indonesia. Colonies had no morphologically distinct queens. Worker number per colony was 11 on average. Most colonies contained multiple mated workers (gamergates), however, only one of them laid eggs, whereas additional gamergates were sterile or ovipositing at a low rate, indicating that this ant species shows functional monogyny. Some virgin workers also developed their oocytes and actually laid eggs. Ritualized dominant — subordinate behavior was frequently observed and a linear dominance hierarchy was established among workers. The topranking individual was a fertile gamergate. Sterile gamergates positioned in lower ranks and they often engaged in foraging. Among virgin workers, callows and young workers monopolized higher ranks.     

Mating behaviour of Rhytidoponera sp. 12 ants inferred from microsatellite analysis

In the queenless ponerine ant Rhytidoponera sp. 12, all workers have a spermatheca and functional ovaries and are potentially able to mate and reproduce. Within a colony gamergates may either be full sisters to each other (Type 1 colony), or they may not be full sisters but still be significantly related to each other (Type 2 colony) due to daughter gamergates reproducing in their natal colonies after mating. Despite many studies the mating behaviour of R. sp. 12 has been poorly understood. In this study, we used microsatellite markers to investigate intracolony relatednesses of male mates to the gamergates (bmq) and between male mates (bmm), and mating frequencies and mating patterns, using gamergate DNA and sperm DNA isolated from the spermathecae of gamergates from five colonies. Average bmm and bmq estimates for all five colonies studied were not significantly different from zero, suggesting that on average, within colonies, mating males were unrelated both to each other and to the gamergates. A low frequency (3%) of multiple mating by gamergates was detected. Multiple mating by individual males with sister gamergates within Type 1 colonies was also detected at 3% and could give rise to half-sister nestmate workers. Polygamy in R. sp. 12 might indicate local female-biased operational sex ratios despite the expectation of overall male biases. Our results concur with previous reports that gamergates mate within the colony or nearby, but indicate more diversity in mating patterns than previously indicated for this polygynous ponerine ant species.     

Monogyny and regulation of worker mating in the queenless antDinoponera quadriceps

The morphologically specialized queen caste has been lost in various ponerine ants, and mated workers (‘gamergates’) reproduce instead of queens. Unlike previous reports in the literature, we found only one gamergate in each colony ofDinoponera quadriceps. We documented monogyny by dissecting ovaries and spermathecae in 914 workers from 15 colonies, and by observing mating in the laboratory. In colonies without a gamergate, aggressive interactions among some of the unmated nestmates led to the behavioural differentiation of a top-ranking worker (‘alpha’), which laid almost all the eggs. Only the alpha went outside the nest at night, and mated if foreign males were present (N=11 tests), thus becoming a gamergate. The alpha was sexually attractive even when her ovaries were not yet active. After intromission, the male remained linked to the alpha while she severed the end of his abdomen. Pieces of the male genitalia remained attached to her genital tract, and she removed them after 30±18 min (X±sdN=9). We interpret this to be a mating plug, preventing other males from fathering her offspring. None of these newly inseminated gamergates continued to go outside the nest, and, when tested, they never re-mated (N=4). Thus, gamergates ofD. quadricepsprobably mate only once. In queenless ant species, comparative evidence indicates that worker mating is often regulated in monogynous species, while unrestricted mating of young individuals is typical of polygynous species (oviposition is regulated subsequently). Furthermore, the occurrence of either monogyny or polygyny influences the mating strategies of males, and mating plugs have been reported only in some monogynous species.     

Evolution of wingless reproductives in ants: weakly specialized ergatoid queen instead of gamergates in Platythyrea conradti

Platythyrea conradti is the only species in this genus with ergatoid (= permanently wingless) queens. Colonies lack gamergates (unlike other species in this genus), yet aggressive interactions among queen and workers define a hierarchy. A single fertile queen has the top rank and highranking workers do not lay eggs, except when the queen dies. Colonial reproduction by both alate queens (independent foundation) and gamergates (fission) seems the ancestral state in Platythyrea. Independent foundation can be selected against in some species, causing the loss of alate queens for economic reasons. Thus gamergates become the only reproductives, except in P. conradti in which queens became ergatoid. Gamergates and ergatoid queens are two mutually redundant reproductive phenotypes that allow colonial reproduction by fission. Received 1 July 2005; revised 10 November 2005; accepted 16 November 2005.     

Fecundity and the Behavioural Profile of Reproductive Workers in the Queenless Ant, Pachycondyla (= Ophthalmopone) berthoudi

Morphologically specialized queens are absent in Pachycondyla (=  Ophthalmopone ) berthoudi (ant subfamily Ponerinae). Instead, several of the workers mate and reproduce (gamergates). Gamergate proportion in nests commonly varies between nests and different times of the year. Individual fecundity of gamergates varies according to the number of these individuals in a nest, and we examined their behaviour in relation to fecundity in nests with different proportions of gamergates. In nests with high proportions of gamergates, they exhibited a diversity of behaviours inside the nest and in some cases could not be distinguished behaviourally from sterile workers. The fecundity of these gamergates was low and variable. In nests with low proportions, gamergates were relatively more fecund, and did not participate in colony labour. The behavioural profile of gamergates is therefore linked to their reproductive physiology, which is influenced by the number of mated individuals in the nest.     

Dominance behaviour and regulation of foraging in the primitively eusocial wasp Ropalidia marginata (Lep.) (Hymenoptera: Vespidae)

Ropalidia marginata is a primitively eusocial, polistine wasp widely distributed in peninsular India. In spite of its primitively eusocial status, queens of R. marginata are surprisingly docile and behaviourally non-dominant (except during the first week or so of their careers as queens). Yet they successfully maintain reproductive monopoly throughout their careers, probably through the use of pheromones. Workers exhibit dominance-subordinate interactions but these behaviours are not involved in regulating reproductive competition among the workers because workers with high dominance ranks are not necessarily the ones who replace lost queens. We have speculated and provided correlational evidence before that dominance-subordinate interactions among the workers have been co-opted in this species for the workers to regulate each other's foraging. Here, we provide experimental evidence in support of the speculation, by reducing demand for food and showing that this results in a significant decrease in the frequency of dominance-subordinate interactions among the workers.     

Ergatoid queens and intercastes in ants: Two distinct adult forms which look morphologically intermediate between workers and winged queens

Summary The term ergatogyne is used in ants to describe permanently-wingless female adults which are morphologically intermediate between workers and winged queens. This definition is ambiguous because there are two distinct categories of ergatogynes: ergatoid queens and intercastes. Both have an external appearance (ocelli and alitrunk structure) which combines traditional queen and worker characters, and thus can be confused if they both function as reproductives — however intercastes in most species cannot reproduce.Ergatoid queens have replaced winged queens in a substantial number of species. They are sometimes externally similar to conspecific workers, especially in various ponerine species which exhibit limited size dimorphism between castes. Ergatoid queens retain the specialized attributes of a reproductive caste, including larger ovaries, and they are always the functional egg-layers in a colony. In contrast, conspecific intercastes represent various graded stages in a series connecting workers and winged queens, and they occur together with the queens. These hybrid phenotypes result from deviations from the normal pattern of caste differentiation during larval development. Intercastes generally lack a spermatheca and have no reproductive function; however they can mate in a few leptothoracine ants, and then reproduce instead of winged queens in a proportion of colonies.     

Stress resistance and longevity are not directly linked to levels of enzymatic antioxidants in the ponerine ant Harpegnathos saltator

Background

The molecular mechanisms of variations in individual longevity are not well understood, even though longevity can be increased substantially by means of diverse experimental manipulations. One of the factors supposed to be involved in the increase of longevity is a higher stress resistance. To test this hypothesis in a natural system, eusocial insects such as bees or ants are ideally suited. In contrast to most other eusocial insects, ponerine ants show a peculiar life history that comprises the possibility to switch during adult life from a normal worker to a reproductive gamergate, therewith increasing their life expectancy significantly.

Results

We show that increased resistance against major stressors, such as reactive oxygen species and infection accompanies the switch from a life-history trait with normal lifespan to one with a longer life expectancy. A short period of social isolation was sufficient to enhance stress resistance of workers from the ponerine ant species Harpegnathos saltator significantly. All ant groups with increased stress resistances (reproducing gamergates and socially isolated workers) have lower catalase activities and glutathione levels than normal workers. Therewith, these ants resemble the characteristics of the youngest ants in the colony.

Conclusions

Social insects with their specific life history including a switch from normal workers to reproducing gamergates during adult life are well suited for ageing research. The regulation of stress resistance in gamergates seemed to be modified compared to foraging workers in an economic way. Interestingly, a switch towards more stress resistant animals can also be induced by a brief period of social isolation, which may already be associated with a shift to a reproductive trajectory. In Harpegnathos saltator, stress resistances are differently and potentially more economically regulated in reproductive individuals, highlighting the significance of reproduction for an increase in longevity in social insects. As already shown for other organisms with a long lifespan, this trait is not directly coupled to higher levels of enzymatic and non-enzymatic antioxidants.     

Population structure and mating biology of the polygynous ponerine ant Gnamptogenys striatula in Brazil

Gnamptogenys striatula is a polygynous ponerine ant, whose colonies contain either several differentiated queens or several gamergates. Population structure, queen mating frequency and deviation from random mating were investigated in a north-eastern Brazilian population. Eight workers from each of 33 queenright colonies and 17 queens and their progeny (20-40 offspring) were genotyped using eight variable microsatellite markers. Population differentiation tests indicated limited gene flow at the scale of several kilometres, and tests of isolation by distance revealed population viscosity at the scale of a few metres. This population structure, together with the frequent colony migrations and fissions observed in the field, suggest that new nests are founded by budding in G. striatula. Genetic data showed that 13 of our 17 queens were single-mated and four were double-mated. The estimation of the range of maximal frequency of double-mated queens in the population was 0.232-0.259, demonstrating that mating frequency is low in G. striatula. The low estimated mean relatedness between the 17 queens and their mates (-0. 04 +/- 0.49) indicated no evidence of inbreeding in G. striatula.     

Experimental Investigation of the Mechanism of Reproductive Differentiation in the Queenless Ant,Diacamma sp., from Japan

We test two hypotheses about regulation of the reproductive division of labour in the permanently queenless ponerine ant, Diacamma sp., from Japan. All workers emerge with gemmae (tiny innervated thoracic appendages), but only one individual keeps them in each colony, and she is the only mated reproductive worker (gamergate). The gemmae of all other workers are mutilated by the gamergate soon after their emergence, and they can never mate. In the presence of gamergate, mutilated workers have inactive ovaries and do not behave aggressively. Two possible consequences of mutilation are: 1. olfactory signal — a pheromone inhibiting the oogenesis of mutilated workers is no longer released by the gemmae; and 2. endocrine degeneration of its afferent neuronal connections interferes reproductive physiology of a gamergate. Gemmae of gamergates were coated with shellac (to prevent pheromone emission) or removed, and over three weeks we studied any changes in ovarian activity of the gamergates as well as nestmate workers. Coating of gemmae did not elicit worker oviposition, suggesting that gemmae pheromones do not have a regulatory function. Experimental mutilation of gamergates resulted in a slight increase in both the frequency of dominance interactions and the ovarian activity of mutilated workers, but this effect was much lower than in colonies where the gamergate was removed. This contrasts with the immediate change in the behaviour (aggressive to timid) of newly emerged workers following mutilation.     

Seasonal cycle of colony structure in the Ponerine ant Pachycondyla chinensis in western Japan (Hymenoptera, Formicidae)

Investigation of reproductive strategies of ants in the subfamily Ponerinae is important for understanding of the evolution of social structure and of the significance of caste dimorphisms. The biology of species with mated and egg-laying workers (gamergates) has been studied for many species, however, little attention has been paid to species that reproduce via alate queens only. We investigated the seasonal cycle of changes in the colony structure of Pachycondyla chinensis reproduced by alate queens in western Japan, and found the following novel biological characteristics of this species. P. chinensis showed a remarkable caste dimorphism in ovariole numbers: workers had no ovaries while queens had 18 to 36 ovarioles in their ovaries. The nesting system seemed to be polydomous: 266 of 400 nests collected were queenless. The number of queenless nests increased during the reproductive season. Among the 134 queenright nests, 38 had several mated-queens without significant differences in ovary activation and the remaining 96 nests were monogynous. During winter to early spring, most nests were polygynous. After alate production, most of the old queens seemed to die or be expelled and replaced by new queens. Virgin dealated queens were often found and they seemed to have laid eggs. Received 3 August 2007; revised 19 December 2007; accepted 20 December 2007     

Reproduction,dominance, and caste: endocrine profiles of queens and workers of the ant <Emphasis Type="Italic">Harpegnathos saltator</Emphasis>

The regulation of reproduction within insect societies is a key component of the evolution of eusociality. Differential patterns of hormone levels often underlie the reproductive division of labor observed among colony members, and further task partitioning among workers is also often correlated with differences in juvenile hormone (JH) and ecdysteroid content. We measured JH and ecdysteroid content of workers and queens of the ant Harpegnathos saltator. In this species, new colonies are founded by a single queen, but after she dies workers compete in an elaborate dominance tournament to decide a new group of reproductives termed “gamergates.” Our comparisons revealed that queens, gamergates, and inside workers (non-reproductive) did not differ in levels of JH or ecdysteroids. However, increased JH and decreased ecdysteroid content was observed in outside workers exhibiting foraging behavior. Application of a JH analog to virgin queens of H. saltator, although effective at inducing dealation, failed to promote egg production. Together, these results support the hypothesis that JH has lost its reproductive function in H. saltator to regulate foraging among the worker caste.     

Bonding between workers and queens in the ant genus Myrmica

Bond formation between young workers and queens is described for three species of Myrmica. This caste bond inhibits aggression and develops cooperation. Workers have a bias towards queens of their own species and do not discriminate between a queen of their own colony and a novel conspecific one. They are able to develop bonds with queens of other species of the genus provided they meet them first and well before their state of queenless aggressiveness sets in. A bispecific group, in situation contrived to enable them to meet gradually, will fight a normal group. They will then develop their bond to include the conspecific queen without becoming hostile to the earlier substitute. Evidently a queen of another species is able to provide an outline set of stimuli that is later built up by the conspecific queen, a characteristic of the imprinting development process. Allospecific queens can establish new microsocieties in a way reminiscent of temporary social parasites. The fact that allospecifically bonded workers will fight their one-time litter-mates that have been normally bonded suggests that the two groups have acquired different smells from their specifically distinct queens.     

Morphology and reproductive behaviour of intercastes in the ponerine antPachycondyla obscuricornis

Summary Queens are usually the functional reproductives inPachycondyla obscuricornis, but none were found in one colony collected near Brasilia. In this colony, three adults differed morphologically from workers; although wingless, their thorax was more or less like that of winged queens (scutum and scutellum distinct; metanotum larger than that of workers). Another 11 such individuals were found by opening cocoons. We consider that these are intercastes (sensu Peeters, 1991). Two adult intercastes were mated and laid eggs. In contrast, all the workers were virgin, but some laid distinct trophic eggs which were fed to the two reproductives. Following experimental removal of the intercastes, dominance interactions began among the workers, and reproductive eggs were then laid. Intercastes with a reproductive function have not been reported previously in the Ponerinae. They are distinct from ergatoid queens (permanently wingless reproductive caste) or gamergates (mated egglaying workers).     

Policing behaviour towards virgin egg layers in a polygynous ponerine ant

The majority of colonies of Gnamptogenys menadensis in Sulawesi lack queens and several workers ('gamergates') mate and reproduce instead. Virgin workers lay morphologically specialized trophic eggs which are fed to larvae. Some of these virgins switch to male eggs when gamergates are experimentally removed. Three distinct patterns of oogenesis thus result in: (1) trophic eggs; (2) reproductive eggs (unfertilized) laid by virgin workers; and (3) reproductive eggs laid by gamergates, whose ovarioles are always longer than those of virgin workers. We investigated the behavioural regulation of ovarian activity in virgin workers by temporarily excluding gamergates. In 12 groups of 35-45 virgins, a few workers became dominant and started to lay reproductive eggs. Once gamergates were reintroduced, sterile workers attacked and immobilized workers with enlarged ovaries (confirmed by dissection of 173 individuals), which often died as a result. Gamergates were never aggressive towards new egg layers. Aggression was not triggered by divergence in colony odours, as it was absent in control experiments in which six colonies were divided in half, with each part containing gamergates, and reunited after 50 days. Our results show that sterile workers discriminate against new egg layers, given that their ovaries are not as developed as those of gamergates. Olfactory detection of different levels of ovarian activity thus appears possible. Mesh experiments indicated that the putative pheromones are nonvolatile and require physical contact for transmission. Aggressive behaviour directed at reproducing workers can be interpreted as worker policing. In G. menadensis, worker policing results in virgins laying only trophic eggs. Copyright 1999 The Association for the Study of Animal Behaviour.     

Independent colony founding by ergatoid queens in the ant genus Pogonomyrmex: queen foraging provides an alternative to dependent colony founding

Ant queens exhibit two primary strategies to initiate nests, independent colony founding (ICF) by solitary queens and dependent colony founding (DCF) when the queen starts a nest with a group of workers that disperse on foot from the parent nest. Numerous ant species have wingless (ergatoid) queens, and it is generally assumed that these species exhibit obligate DCF because their lack of wing musculature provides them with few resources to divert towards producing their first brood of workers. Thus, ICF by ergatoid queens is viewed as maladaptive because these queens need to take additional dangerous foraging trips to garner sufficient food to rear their first brood of workers. Contrary to this prediction, I document ICF by ergatoid queens for three species of harvester ants in the genus Pogonomyrmex (subfamily Myrmicinae), P. cunicularius cunicularius, P. cunicularius pencosensis, and P. huachucanus. Queens of P. huachucanus were obligate foragers, i.e., no minim workers could be produced without external food, and one queen of P. cunicularius pencosensis was observed foraging in the field. Abundant and/or predictable food resources likely select for the evolution of semi-claustral nest founding and ICF by these ergatoid queens. Under these conditions, foraging time would be minimized and the number and size of minim workers would be maximized. These benefits should increase founding success, which could compensate for loss of long-range dispersal. Overall, this study demonstrates that care should be taken before concluding that ant colonies employ DCF based solely on queen morphology.     

Queen lifespan and colony longevity in the ant Harpegnathos saltator

Abstract.  1. The longevity of field colonies was investigated in the ponerine ant Harpegnathos saltator (Jerdon) in which either reproductive workers (gamergates) or a single queen reproduce.
2. Data from 3 years were used to calculate the ratio between queen-right ( n  = 50) and gamergate ( n  = 12) colonies that can be used to derive the colony mortality of gamergate colonies. Using the survival rates of queens in the laboratory, extrinsic and intrinsic mortality rates of queen-right colonies were calculated.
3. No significant differences in the sizes of queen-right and gamergate colonies above 14 workers was found, suggesting that mortality of established colonies is not size related.
4. The mortality of gamergate colonies is 4.17 times higher than the intrinsic mortality of queen-right colonies.
5. In the laboratory, mean survival time of queens in colonies of more than 14 workers was 1.79 years.
6. Estimated mean survival time of queen-right and gamergate colonies in the field varies between 0.78 and 0.43 years respectively, when no costs of conflict during the replacement of queens occur; however, when the latter costs increase colony mortality to a level similar to extrinsic mortality, the calculated longevity of queen-right colonies would increase to 1.02 years.     

Worker policing limits the number of reproductives in a ponerine ant

Reproductive division of labour is an essential feature of insect sociality, but the regulation of sterility among colony members remains incompletely understood. Ant workers and queens are morphologically divergent and workers are only capable of producing males in a colony, although they usually do not do so. Worker policing is one mechanism proposed for their infertility and it can be expressed as either aggressive inhibition of ovarian activity among workers or destruction of worker-laid eggs. A few studies have shown that workers with developed ovaries are preferentially attacked by nest-mates, but adequate demonstration of worker policing also requires evidence that these attacks result in the suppression of ovarian activity or death. We investigated worker policing in the ponerine ant Harpegnathos saltator in which workers are able to mate and replace the founding queen. Five colonies were each divided into two groups, one of which consisted exclusively of infertile workers. Some individuals in the orphaned groups began laying eggs during the three-week separation and upon reunification these were vigorously attacked by infertile workers of the other groups. The ovarian activity of these new egg layers became inhibited, as revealed by subsequent dissection of marked individuals. Worker policing in H. saltator appears to function primarily in preventing an excess of reproductive workers.     

CONFLICTS OF INTEREST IN SOCIAL INSECTS: MALE PRODUCTION IN TWO SPECIES OF POLISTES

In social insects, reproduction is often monopolized by queens even though in many species are workers capable of laying male eggs. Because it is difficult to see how one or a few queens can suppress the much more numerous workers, collective worker control, or policing, offers an attractive solution. When workers are less related to other workers than they are to queens, workers should be selected to suppress each other in favor of the queen's male offspring, if other things are equal. Otherwise, they should allow each other to lay male eggs. For two species of Polistes, we used DNA microsatellites to estimate these two relatednesses, to determine the sex of brood, and to determine whether male brood was produced by queens or workers. Workers were significantly more related to each other (0.63 and 0.73 for P. bellicosus and P. dorsalis, respectively) than they were to queens (0.40 and 0.54, respectively) so they were predicted to allow each other to lay the male eggs. However, workers did not lay male-destined eggs in either species, so the results do not support collective worker control. There are two possible explanations for this result. Queens may be able to physically dominate in these small colonies. Alternatively, this may be a conventional settlement that minimizes conflict and the attendant costs.