Research projects

This section describes most of the research of the Liebig lab. Any references to the research field can be found in the respective papers of the Liebig lab that are listed in Publications.

Fertility signaling in ants

The most fundamental property of eusocial insects is reproductive division of labor. In ants, typically a mated queen lays eggs that are raised by workers. The latter generally have ovaries and the ability to lay their own eggs,  but they usually do not do so in the presence of the queen. What is the mechanism that tells the workers when they should reproduce? Workers need information about the presence and fertility of the queen. If she is absent or her fertility declines workers are expected to activate their ovaries and try to produce males from unfertilized eggs. Workers perceive this information through fertility signals in the form of specific cuticular hydrocarbons which can be visualized as chromatographic chemical profiles (Peeters and Liebig 2009, Liebig 2010).

For example, workers of the ant Harpegnathos saltator can mate and replace the queen as reproductive (see our model systems). By doing so, they become so-called gamergates. In the process of becoming gamergates, the workers develop a hydrocarbon profile that is only shown by reproductive individuals including queens (Liebig et al 2000).

In the ant Camponotus floridanus, queens develop a reproductive-specific hydrocarbon profile when they increase their ovarian activity along with increasing colony size (Endler et al 2006). This profile is  generally also present on the surface of the eggs of the respective egg-layer. Since founding queens and egg-laying workers have low ovarian activity, their cuticular profile, as well as the surface hydrocarbons of their eggs, do not show reproductive-specific hydrocarbons (Endler et al 2006, Endler et al 2007). Workers are able to discriminate such eggs from those laid by highly fertile queens by their hydrocarbon profiles. A transfer of queen-specific hydrocarbons onto worker-laid eggs makes them “appear” to have been laid by highly fertile queens which leads to their partial acceptance (Endler et al 2004).

The exposure of queenless worker groups to eggs laid by highly fertile queens is sufficient to lead to a 100% inhibition of worker egg-laying indicating that the hydrocarbon profile of these eggs functions as a signal of the queen indicating her fertility and presence (Endler et al 2004).

Inhibition of worker egg laying by the presence of queen-laid eggs. Worker groups that received eggs laid by highly fertile queens and larvae did not lay eggs within 160 days of isolation while worker groups that received only larvae or no brood started to lay after 60 days.

The fertility level of the queen is recognized by workers. In fact, it even interferes with nestmate recognition. When foreign highly fertile queens were exposed to workers from a colony with a highly fertile queen, no attacks were initiated by the workers. However, foreign workers and foreign founding queens were immediately attacked in the same experiment (Moore and Liebig 2010a). This indicates that a fertility signal of the queen interferes with nestmate recognition. The fertility signal is most likely represented by the same hydrocarbons that are responsible for egg discrimination and worker ovarian inhibition.

Introductions of foreign workers (FW) and foreign, lowfertility queens (FLFQ) were significantly more likely to end in aggression than introductions of nestmate queens (NQ), nestmate workers (NW) or foreign, high-fertility queens (FHFQ). Different letters indicate significant pairwise differences

Further support for the presence of a fertility signal that varies with ovarian activity comes from an experiment with incipient colonies. In the same experimental setup as above, workers from incipient colonies attack highly fertile foreign queens as well as foreign queens from incipient colonies. In such incipient colonies, queens do not yet express the cuticular hydrocarbon profile that is present in highly fertile queens, so their workers do not yet recognize the fertility signal and thus attack foreign individuals (Moore and Liebig 2010b). Interestingly, such workers do not reject eggs whether they originated from highly or weakly fertile foreign queens or from foreign workers (Moore and Liebig 2013).  

Preventing workers from laying eggs is a way to regulate worker reproduction. In the ant Aphaenogaster cockerelli, this is achieved in two ways. Workers that activate their ovaries produce a hydrocarbon profile that becomes more and more similar to that of a queen the longer they are reproductively active (Smith et al 2009). Nestmate workers recognize the presence of pentacosane (C25) in the profile of fertile workers and attack them which inhibits their ovarian activity. This behavior (worker policing) can be induced by applying small amounts of pentacosane to infertile workers which triggers worker attacks (Smith et al 2009).

Pentacosane treatment of infertile workers induces attacks by nestmate workers while treatment with nonacosane or the solvent does not. Pentacosane only occurs in reproductive queens and fertile workers. The attacks of pentacosane-treated workers suggest that such fertility-indicating hydrocarbons are used for the regulation of reproduction in this species.

In A. cockerelli, the queen gets directly involved in the regulation of worker reproduction as well. Similar to workers, queens also recognize workers with active ovaries through the presence of pentacosane on their cuticles (Smith et al 2011). Once identified, the queen marks reproductively active workers with chemicals from her Dufour’s gland which subsequently makes the marked worker the target of intense attacks by nestmate workers (Smith et al 2012a). The many and efficient ways of controlling worker reproduction may be the reason worker reproduction is absent, or at least relatively rare, in this species (Smith et al 2012b).

Fertility signaling in termites

Termites evolved eusociality as well, but their social structure shows major differences when compared to the eusocial Hymenoptera, represented by the ants, some bees, and some wasps. While the  eusocial Hymenoptera consist of female societies, termites have an equal sex ratio within the colony. In addition, a king is present next to the queen with whom he regularly mates. In the case of the termites, the queen and the king need to advertise their presence and fertility to their nestmates. We found that in the termite species Zootermopsis nevadensis, reproductive individuals of both sexes produce long-chained poly-unsaturated alkenes in their hydrocarbon profiles (Liebig et al 2009) making these chemicals candidates for fertility signals in this species.

Reproductive females and males show 4 polyunsaturated alkenes not present in larvae.

Alternative perspectives to within-colony conflicts

In ants and other eusocial Hymenoptera, female larvae can either develop into reproductively degenerate workers or into queens that can found a colony. If the founding success of queens is sufficiently high, larvae may receive more fitness benefits if they become queens instead of workers, which gain only indirect fitness benefits through their raising of brothers and sisters. In this situation, larvae should try to develop into queens. However, this may deplete the colony’s workforce, which is not in the interest of their nestmates, and thus produces a conflict between larvae and workers.

In the ant Harpegnathos saltator, larval fate is associated with levels of juvenile hormone (JH) in their 4th instar. Applying JH analog (JHA) to larvae in their late 3rd or early 4th instar leads to their development as queens (Penick et al. 2012). If queen development is induced in a colony that does not currently produce queens, workers start biting the manipulated larvae. The same happens if queen-determined larvae are transferred into a colony not currently producing queens. The biting response by workers significantly decreased queen development in JHA-treated and naturally queen-determined larvae (Penick and Liebig 2012). Given that Harpegnathos workers can replace a queen in a colony, the average inclusive fitness of workers may be equal to that of alate queens which leave the nest to found their own colonies. In this case, the biting behavior of workers may not be an indication of conflict but instead may represent a signal to larvae regarding when it is best for them to develop into workers rather than queens . The interests of larvae and workers converge in this case, and since workers are likely to have better information about colony status and environmental conditions than larvae, they are in a better position to decide what developmental trajectory larvae ought to take.

Endocrine basis of reproductive regulation

Acquiring a reproductive position in social insects is obviously associated with many physiological changes. Biogenic amines or other hormones may either drive transitions from non-reproductive to reproductive status, or changes in the levels of these compounds might result as a consequence of such transitions. The ant Harpegnathos saltator is especially suited for such studies, because workers can become reproductive like queens (i.e. become so-called gamergates) and it is easily possible to follow such transitions.

Juvenile hormone (JH) and ecdysteroid levels often increase with the acquisition of a reproductive position in social insects, which made them candidates for investigation in H. saltator. Interestingly, levels of JH and ecdysteroid did not differ between reproductive queens, gamergates, and non-reproductive workers that stay inside the nest. However, levels of JH were increased and ecdysteroid levels were decreased in foraging workers compared to the other three groups. Even though JH may still be involved in triggering ovarian activation, it seems to have lost its function in the regulation of reproduction in H. saltator and instead appears to play a role in the regulation  of foraging among the worker caste (Penick et al. 2011).

The transition from worker to gamergate was, on the other hand, associated with an increase in dopamine levels in the brain, while levels of octopamine, serotonin, and tyramine stayed the same. In fact, dopamine levels correlated with ovarian activity. Dopamine levels in the brain decreased when workers that were in the process of acquiring a gamergate position were policed (i.e. attacked by nestmates and reproductively inhibited), and octopamine and tyramine levels were associated with foraging activity. This suggests a role of dopamine in the regulation of reproductive behavior while octopamine and tyramine may be involved in the regulation of foraging (Penick et al 2014).

Dopamine levels in the brain changed in regarding to the social context (i.e. establishment as gamergate associated with dueling or getting policed)

Sensory basis of olfactory communication

Pheromones are a major component of reproductive regulation in insect societies and are detected by receptor organs on the insect antenna. Cuticular hydrocarbons and other semiochemicals are detected by sensilla basiconica, which are thick, hair-like structures on the antenna. These hairs have very small pores through which semiochemicals enter the sensillar lymph where they are transported to receptors on the olfactory receptor neuron (ORN).

With single sensillum recording ORN responses to specific semiochemicals can be visualized and analyzed. We are currently focusing on the responses of social insects to the hydrocarbons typically present on their cuticles.

These results are paired with olfactory conditioning to allow the analysis of the discrimination ability of these insects.

Electrophysiological approach to single sensillum recording in our model systems.

Genetic and epigenetic basis of insect sociality

The Liebig lab obtained the draft genome sequences of our two ant and one termite model systems, Harpegnathos saltator, Camponotus floridanus, and Zootermopsis nevadensis, in collaboration with other groups (Bonasio et al, 2010; Terrapon et al 2014). These genomes are used as the basis for the analysis of genetic and epigenetic interactions that are associated with major features of insect sociality, such as developmental and behavioral plasticity, differential aging, and sophisticated olfactory discrimination.