↓5 |
The supply of nutrients to eggs is an investment of the mother into the fitness of her offspring. Studies in species of almost all major taxa of egg-laying animals have shown that the amount of provisions contained in the propagule, usually approximated by measuring its size or weight, can influence the size, vitality or reproductive success of offspring (e.g. Kaplan, 1987; Russell et al., 2007). In situations where environmental conditions are stable, oviparous females tend to lay eggs which are all receive similar quantities of nutrients (reviewed in Crean and Marshall, 2009; Marshall et al., 2008). If environmental conditions are unpredictable however, theory predicts that mothers should produce both small and big eggs as a way of “hedging their bets”. If conditions are favourable, a great proportion of offspring both from light and heavy eggs can survive. If they are unfavourable, at least some will likely be able to carry on the genes of their parents (Cooper and Kaplan, 1982; Philippi and Seeger, 1989; Crean and Marshall, 2009). In the honeybee (Apis mellifera), effects of environmental changes on the success of brood rearing are probably less immediate than in other species, because the colony and its food stores provide a buffer between exterior influences and the immature stages in their cells. Nevertheless, there are good reasons why a bet-hatching strategy could be adaptive in this highly social species. One is that diversity of worker phenotypes provides a security against changing conditions for the colony as a whole. This is thought to be one reason why polyandry is common in highly eusocial insects, because it increases phenotypic variation among workers by increasing genetic variation (e.g. Waddington et al., 2010; Mattila et al., 2008; Oldroyd and Fewell, 2007). Differences between workers can also be caused by differences in maternal investment (Borodacheva, 1973), and this could add to the variability resulting from the presence of several patrilines. A second possible role for differential maternal investment could lie in the fact that female eggs can develop into either queens or workers. It has been shown that queen weight and the number of a queen’s ovarioles are strongly influenced by the weight of the egg from which they develop (Borodacheva, 1973). Therefore, the production of some heavier eggs for colony reproduction or as a “precautionary measure” for the case of sudden queen loss could be a selective advantage.
↓6 |
A precondition to studying these interesting possibilities is the existence of variability of the size of egg provisions in the honey bee. The aim of this study is to test whether this precondition is fulfilled. Several studies have already dealt with egg weight variability in Apis mellifera, but they focussed mostly on inter-colony differences and environmental influences. Taber and Roberts (1963) and Roberts and Taber (1965) have evidenced considerable differences between queens concerning egg weights, and found that these differences were partly heritable. Borodacheva (1973) weighed eggs from 200 queens and found that the heaviest egg was more than 2.5 times as heavy as the lightest one. Król (1996) showed that the weather had no sizeable influence on egg weight, but that negative correlations between egg weight and the number of eggs produced can sometimes be found (see also Jordan, 1961, for this relationship). Woyke (1998) showed that the size and weight of bee eggs change during development, which means that egg age should be standardized as much as possible if differences in maternal investment are to be measured.
To create a methodological basis for the study of differential prenatal maternal investment in the honey bees, we started off by determining the repeatability of egg weight measurements. We also identified the optimum age of eggs for measuring differences in maternal investment. In the context of possible effects of differential maternal investment on properties of the offspring, it was mainly differences between eggs laid by the same queen which were of interest. Accordingly, our principal aim was to measure variability within samples of eggs laid by individual queens within a short time frame. Secondly, we studied weight differences between the eggs laid by the same queens either in spring (when the so-called “summer bees” are produced) or in late summer (when both “summer” and “winter bees” are produced). Our hypothesis was that if quantitative differences in maternal investment play a role in the differentiation into “summer” or “winter bees”, this should lead to greater variability of egg weights in late summer.
Manuscript 1
↓7 |
Variability of prenatal maternal investment in the honey bee ( Apis mellifera )
Saad Naser AL-Kahtani, Jakob Wegener, Kaspar Bienefeld
(submitted for Journal of Apicultural Research)
↓8 |
Summary
The unequal distribution of resources among offspring by females can be an important adaptation to changing or unpredictable environments in many taxa. In the honeybee, a potential role for differential maternal investment could lie in the amplification of intra-colony phenotypic variability, which is an important factor in stress resilience. Differential maternal investment could also influence polyethism, if it led to physiological adaptations that are important for fulfilling different tasks. Here we asked whether egg weight variability in Apis mellifera is great enough to justify such hypotheses. We first determined the repeatability of weight measurements, and the optimum stage at which eggs should be sampled in order to minimize imprecision due to unavoidable intra-sample age differences. We then weighed eggs from fourteen colonies, and assessed the relative level of weight variability. We also compared the means and variabilities of eggs produced in spring and late summer, to test the hypothesis that eggs destined to become a mix of summer and winter bees should be more variable than eggs all destined to become summer bees. The results show that the optimum age for sampling eggs is 48 h. No systematic difference was found between spring and summer samples, but sample means from the same queens differed by up to 22%. Weight of eggs laid by the same queen within six hours varied by up to 58%. Egg weight variability was clearly sufficient to expect phenotypic differences at the adult stage. We conclude that a considerable level of differential maternal investment exists in the honey bee. The study of its consequences and evolutionary significance in social species could be an exciting field for future studies.
Colonies of social insects make numerous group decisions with respect to foraging, nest maintenance and reproduction. Queen production, for example, is of extreme importance because fitness of the colony strongly depends on decisions made during this process (Tarpy and Gilley, 2004). In honey bee (Apis mellifera) colonies, new queens are produced during reproductive swarming, supersedure and emergency queen replacement. In the latter case, only a few larvae from the many available are selected by workers to be reared as gynes (Winston, 1987).
↓9 |
Nepotism has been hypothesized as the underlying reason for the selection of individual larvae to be reared as queens (Tarpy et al., 2004). However, the mechanism that initiates this group decision is uncertain (Tarpy and Gilley, 2004). In this study, aimed primarily at quantifying the factors that determine which individual larvae are chosen as gynes, we observed that some workers presented their Nasonov glands after inspecting larvae that were offered to queenless colonies in artificial queen cups. The Nasonov pheromone is known to be involved in the outdoor orientation of honey bees (Sladen, 1901). The fundamental decision-making dilemma for groups is to turn individual preferences into a single choice for the group as a whole (Seeley et al., 2006). This observed behavior may be essential to this process. Using video records and observations of the initial phase of the queen rearing process, we sought to determine whether exposure of the Nasonov glands is linked to the recruitment of honey bees for individual larvae to be reared as queens.
Manuscript 2
The Nasonov Gland Pheromone is Involved in Recruiting Honey bee Workers for Individual Larvae to be Reared as Queens
↓10 |
Saad Naser AL-Kahtani and Kaspar Bienefeld
(accepted for Journal of Insect Behavior)
Summary
↓11 |
Colonies of social insects make numerous group decisions, and queen production has a high impact on their fitness. However, the mechanism which initiates selection of individual larvae to be reared as queens remains uncertain. Infra-red video technique and a modified observation hive provide us insight into the behaviour of worker bees at the initial phase of the queen rearing process. Some workers were found to present their Nasonov gland (ENG) after inspecting larvae which were offered to queenless colonies in artificial queen cups. To our knowledge, these are the first data which show an exposure of Nasonov glands within the hives of honey bees. ENG was found exclusively at cells containing young brood (< 48 h) suitable for queen rearing, and a highly significant concentration of ENG was observed at cells later developing into queen cells. This supports the assumption that ENG is not a general trigger for brood care, but that it is involved specifically in campaigning for larvae to be reared as gynes. ENG is able to attract other bees and successfully recruit specific larvae. However, I also found ENG at cells which were not accepted for gyne production. The duration of ENG (assumed indicator of the intensity of recruiting) did not differ significantly at the cell which was accepted or not accepted for queen rearing. It is likely that not every act of recruiting behaviour results in a group decision. The lack of a significant difference between duration of ENG and a highly significant worker density after recruiting for cells accepted and not accepted for queen rearing after ENG, respectively, provides strong evidence that differences in the intensity of recruiting do not influence the final group decision, but that differences in the attractiveness of the campaigned larvae do. If the decision of the recruiting bee for an individual larva corresponds with the notion of the group, then the larva’s superior attractiveness is likely to fine tune the search for it within the pheromonal sphere of the recruiting worker.
In the honey bee Apis mellifera, mating of queens with several haploid males results in societies with large variations in genetic relatedness. Subfamilies originating from the same drone share on the average 75% common alleles, while among members from different subfamilies, only 25%. Workers may benefit in a nepotistic manner by raising super sisters instead of half-sisters as new queens, thereby increasing their inclusive fitness (Tarpy et al., 2004). A precondition for selective queen rearing is the ability of workers to recognize kinship, as was demonstrated by Breed (1981) and Getz and Smith (1983), but results of studies to prove the presence of nepotism in queen rearing were inconsistent (Page and Erickson, 1984; Breed, 1981; Koeniger et al., 1996; Breed et al., 1984; Visscher, 1986; Woyciechowski, 1990). However, by pooling data of all available studies, a significant preference towards related larvae (n = 14148 larvae) was found (χ2 = 21.3, P < 0.0001). But the most common methods in these studies, in which young brood was reciprocally transferred between colonies are criticised (Tarpy and Fletcher, 1998) because recognition may not be based on kin but rather on colony odour, which is partly environmental and can be differentiated by worker bees (Breed, 1994).
In the current study, we try to avoid this methodological pitfall, by artificial rearing of transferred larvae and we include, beside relatedness, another factor, which may impact the acceptance for queen rearing. Different prenatal maternal investment was found to affect fitness of the resulting offspring (Sinn et al., 2008), consequently this differences in maternal supply may provide a suitable selection criterion at this early stage. Prenatal investment was estimated by measuring egg weight at 48 h from which the tested larvae originated from.
↓12 |
Manuscript 3
Strength surpasses relatedness – how honeybees decide which larvae should be reared as new queens
Saad Naser AL-Kahtani and Kaspar Bienefeld
↓13 |
(In preparation for publication)
Summary
Natural selection favours efficient cooperation within eusocial colonies. However, in polyandrous species, queen rearing might provide some conflict. Worker bees are assumed to play a nepotistic role during the queen-rearing phase by preferring more closely related larvae (Getz and Smith, 1983). Honeybee workers have been found being able to discriminate between differently related individuals (Page and Erickson, 1984), although the published data are inconsistent (Tarpy et al., 2004). Here, we show that standardized reared larvae are not significantly preferred by related worker bees, but that a higher egg weight from which the larvae originate significantly increases its chances of receiving royal treatment and shows a tendency positively to correlate with the fitness traits of the resulting queen. Queen rearing results from a colony decision involving many workers. Consequently, the lack of significantly establishing nepotism might be because the relatedness-driven kin preference of individuals cannot be transferred into a colony decision. We have therefore also tested, by DNA fingerprinting, the relatedness of the larvae to be reared as queens and the nursing worker bee that initiates the queen rearing process. In five colonies, each inseminated with three drones of three different origins, we have not observed any preference of worker bees for their closely related sibling for queen rearing but we have found a significant preference of workers for larvae of a special drone origin. Our data provide evidence that honeybees have the ability to discriminate genetic differences but that they do not use this skill for nepotism. In the naturally genetically diverse but always (at least 25%) related framework within a colony, genetically or environmentally driven attractivity and not relatedness are the key criteria for selecting larvae for gyne production. With regard to factors such as cost and benefits in Hamilton’s equation (1964a,b), the preference for higher relatedness is not necessarily the best option for inclusive fitness and, consequently, our data do not support the new hypothesis of Nowak et al., 2010 proposing that relatedness is irrelevant for eusociality.
© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme. | ||
DiML DTD Version 4.0 | Zertifizierter Dokumentenserver der Humboldt-Universität zu Berlin | HTML generated: 05.12.2014 |