FROM the outside, a honeybee colony appears the epitome of social harmony. The queen produces offspring that are raised by loyal workers who forgo their own reproduction to keep the colony running smoothly. But in reality it is a dark world of conflict, infanticide and thwarted reproductive ambitions, kept in check by brutal policing. And very occasionally, anarchy breaks out. The police lose control and hordes of normally sterile workers abandon their duties to the queen, and instead direct all their energy towards their own reproduction.
“It’s amazing,” says Ben Oldroyd from the University of Sydney, Australia, who is one of the few people to have witnessed it. “The queen ends up in a corner of the colony surrounded by a few loyal workers, and the rest of the colony goes berserk just laying eggs.” Eventually workers become disoriented, barely able to feed themselves. Oldroyd has even seen them trying to rear queens out of male larvae. At that point, he says, total social collapse is inevitable.
Anarchy, in the world of the honeybee at least, is extremely rare. So rare in fact that it took Oldroyd a decade of advertising in beekeeping magazines to track down just two anarchic colonies. His persistence was fuelled by the belief that this aberrant behaviour holds the key to understanding social order, and disorder. Now his tenacity is about to pay off. He and his team are on the verge of pinning down the genes involved in preventing anarchy and in triggering it – what he calls the “ultimate gene for altruism” and its opposite, the truly selfish gene. And the discovery could have implications that go way beyond insects to shed new light on the dynamics of conflict in all animal societies – including our own.
Advertisement
Oldroyd already has a shortlist and predicts that his team will publish the exact identity of the honeybee genes for altruism within a few months. If so, the announcement will come 40 years after British evolutionary biologist William Hamilton resolved the paradox of how such a gene can spread in a population, by showing mathematically that genes for altruism are actually as selfish as any other gene. Hamilton realised that the key is relatedness. His “kin selection” theory states that if a gene confers altruistic behaviour that benefits other individuals who share genes with the altruist, then this will ensure the altruist’s genes are passed on – including the gene conferring altruism itself.
This probably goes a long way to explaining why insects in the order Hymenoptera – bees, wasps and ants – are often so very nice to one another. The fact that these insects have evolved sociality perhaps a dozen times owes much to their curious method of sex determination. Hymenopterans employ a system called haplodiploidy, in which females are diploid – possessing two sets of chromosomes – while males are haploid, developing from unfertilised eggs with just a single set of chromosomes. If you consider relatedness on a scale from 0 to 1, with 0 meaning no relation and 1 meaning genetically identical, mothers and daughters in the haplodiploid system come out at 0.5. Sisters have a relatedness of 0.75, since they share half their mother’s genes and all their father’s, and the value for a sister and brother is 0.25 (see Diagram).
This has important consequences for the colony. Here, the workers – who are all the female offspring of a single queen – are more closely related not only to their own sons, but also to the offspring of sister workers (0.75 × 0.5 = 0.375), than to their brothers, the queen’s sons (0.25). Kin selection theory predicts that workers will therefore favour worker reproduction, and that conflict will arise because every reproducing bee is more related to her own offspring than to anyone else’s.
But everything changes if the queen produces her offspring by more than one male. By diluting the relatedness of workers to each other, she ensures they are marginally more closely related to her own sons than to sons of other workers. It is then in the workers’ interest to devote their energies to rearing the queen’s brood, and to prevent other workers from reproducing. Kin selection theory now predicts that worker policing should emerge, to eliminate the inevitable cheats.
The first suggestion that honeybees might live in this oppressive police state came in the late 1980s from evolutionary biologist Francis Ratnieks, then at Cornell University in New York. Worker honeybees are physically incapable of mating, and their ovaries are usually inactivated by pheromones produced by the queen, but Ratnieks noted that in a normal hive a few workers do try to cheat the system by laying unfertilised eggs. He also found that these rebels are quickly sniffed out and attacked by other workers, and their eggs are eaten.
It’s a highly effective system: although around 7 per cent of male eggs may be laid by workers, only 1 in 1000 males have workers as mothers. Worker policing is now turning up in more and more social insects, including many species of wasp, bee and ant. “Wherever we look, we find it,” Ratnieks says.
“It looks like the evolution of anarchy requires a mutational double whammy”
Ratnieks has always believed that worker policing must be regulated at some deep level by genetic relatedness, but for a long time this idea was impossible to test. Then along came microsatellite analysis. Microsatellites are stretches of non-coding DNA containing many repeats of a short sequence of the chemical bases that make up the genetic code. They are not genes, but they make ideal markers of parentage because they are easy to find in the genome, the number of repeats is highly variable, and individuals often inherit two versions of different lengths from their parents.
In 2000, Ratnieks, now at the University of Sheffield, UK, and colleague Kevin Foster used the technique to test Hamilton’s rule in the wasp species Dolichovespula saxonica, whose queen can mate either once or several times (Nature, vol 407, p 692). They showed that in a colony where the queen mated just once, almost all worker-laid eggs developed successfully into adult males. But when the queen mated several times, something was happening to ensure that workers’ eggs did not get a chance to develop. That something, Ratnieks and Foster argued, was worker policing.
It followed that in honeybee colonies, where the queen mates on average 17 times, worker policing should be a force to be reckoned with. The next obvious question was: how do policing workers distinguish queen-laid eggs from worker-laid eggs? The clever money is on some sort of chemical marker, but pinning this down in a hive awash with pheromones and other chemical signals is difficult. “It’s proving to be a challenging piece of research,” admits Ratnieks. Now, though, they have uncovered some tantalising clues.
Stephen Martin from Ratnieks’s group, working with organic chemist Graeme Jones of Keele University, UK, has just completed an analysis of secretions from the Dufour’s gland of honeybees. This gland empties into the bees’ reproductive tract, and so is the most likely source of any egg marking scent. Queenly secretions, they found, contain high levels of esters but lack an alcohol called eicosanol, whereas workers secrete eicosanol but not the esters. Even workers that start laying after the death of the queen produce the alcohol that marks them out as workers, though they also produce small amounts of esters.
But that’s not all. The study also hints at how anarchistic worker bees can lay eggs that evade detection: analysis of some of Oldroyd’s anarchistic bees showed that their Dufour’s secretions are rich in the esters that normally signify a queen. However, it seems that these bees haven’t perfected the deception. “If you take anarchistic worker-laid eggs and place them in a healthy colony that still has its queen, the workers can still distinguish the worker-laid eggs and remove them,” says Jones. “The anarchistic workers only appear to be able to lay more acceptable eggs in anarchistic colonies.” Jones concludes that anarchy doesn’t just require individuals to scramble their egg markers; there must also be a breakdown in policing as the workers somehow lose their nose for transgressors.
Really selfish genes
Oldroyd’s own research also indicates that the evolution of anarchy requires a mutational double whammy. The two original anarchistic honeybee colonies he tracked down each had a dozen or so laying workers, and from these he has selectively bred colonies in which 40 per cent of the workers lay eggs. Experiments with these anarchistic bees clearly show that egg laying and egg detection are under separate genetic control. Oldroyd has bred colonies in which workers lay eggs that are consistently detected and removed by fellow workers, and other colonies where workers do not lay, but neither do they remove workers’ eggs that have been transplanted into the colony. Now, with the help of the honeybee genome sequence, which was published last year, he and his colleagues are moving in on the genes involved.
The team has used two different genetic approaches. The first is based on the observation that anarchistic workers can override the pheromones that usually suppress their ovaries. If this is under genetic control, then normal workers should have a gene associated with sterility, while anarchistic ones will have a slightly different version of that same gene, allowing their ovaries to be more active.
To search for the gene or genes involved, the researchers have produced workers with different combinations of anarchistic and conformist genes by mating normal males with queens that produce anarchistic workers, and then mating their offspring with the son of an anarchistic worker. The team then identifies individuals with particularly active ovaries and others with particularly inactive ovaries, and looks for correlations between the frequencies of certain alleles – different versions of the same gene – using microsatellite analysis. “Microsatellite alleles that co-segregate with ovary activation must be very near the anarchy gene,” says Oldroyd. So these alleles direct the researchers to the right part of the genome.
The second approach is to extract RNA from the cells of normal and anarchistic bees. This indicates what proteins they are producing, since RNA is the middleman in the formation of protein from genes. By screening the RNA against a library of 7200 honeybee genes, the researchers can then identify genes that are over-expressed as protein in anarchistic bees compared with normal ones, and vice versa. Using this method, Oldroyd’s team has come up with a list of 60 genes. “The top five or six candidates are to do with reproduction or pheromone receptors,” says Oldroyd. “If we’re right, the genes for selfishness or anarchy are in that group.” He is convinced they will find that anarchy arises when two or more unusual genetic mutations coincide in the same colony. This could explain the rarity of anarchy in nature. Looked at another way, it also suggests why social cohesion is so enduring, despite the potential conflicts of interest within societies.
The findings are eagerly awaited, but not everyone agrees that anarchy is as rare as Oldroyd suggests. Laurent Keller of the University of Lausanne in Switzerland thinks it is probably widespread in nature, but beekeepers mistake anarchistic colonies for ones that have lost their queen. Keller also points out that anarchy is just one of a range of selfish strategies that colonial insects have developed to beat the system (see “How to get ahead in a hive”). He argues that policing exists simply to keep the colony running efficiently, regardless of its kin structure, and that cheats will inevitably evolve. “I think the same must have occurred in all other social insects that have worker policing,” he adds.
If Keller is right, then the tension between conflict and harmony in insect colonies runs very deep. What’s more, he thinks there are parallels with other biological systems, from genomes – where policing mechanisms must have evolved to prevent individual genes from selfishly inserting themselves into more than their fair share of eggs or sperm – to social structures throughout the natural world in which conflict between individuals with different priorities must be resolved.
Whether conflict resolution and anarchy among honeybees can tell us anything about human relations is another matter. After all, we have an overlay of culture to curb our biological instincts, and we cooperate with genetically unrelated strangers we are never likely to see again – seemingly violating Hamilton’s theory. Nevertheless, says Keller, conflict is ever-present in human societies.
“Conflict occurs when interests vary and resources are limited, and that applies to humans as well as bees,” says Ratnieks, who has just embarked on a fellowship at the Institute for Advanced Study in Berlin, using the honeybee as a model to study conflict resolution. Already he has found that the insect societies with the most effective policing also have the lowest levels of worker egg laying. In other words, good policing acts as a deterrent. Obviously insect policing does not have an exact parallel in human societies. “But we do have policing systems that are highly analogous to the honeybee, systems that prevent individuals from disrupting the society as a whole,” Ratnieks says.
How to get ahead in a hive
“In all the social insects, whether you become a queen or a worker doesn’t depend on your genes, it depends on how you’re reared,” says Francis Ratnieks from the University of Sheffield, UK. Queens are fed more, so they grow bigger, and their diet is rich in royal jelly, as opposed to the pollen fed to workers. Among some species of tropical stingless bee, female larvae sometimes manage to eat more than their fair share, turning into queens rather than workers – albeit dwarf ones. “There is a clear advantage to being a queen in that others will help promote the survival of your offspring,” says Ratnieks. “So some of them beat the system by becoming queens.”
Bumblebees (Bombus terrestris) take a different route to the top, incorporating something similar to honeybee anarchy into their natural life cycle. Bumblebee colonies are founded in the spring by a queen, and grow through the summer before a new queen emerges in a peculiar contest. During the summer there comes a “switch point” at which the queen sends a chemical signal to her female larvae to start developing into queens instead of workers. About a week later all hell breaks loose. The workers start laying eggs and being aggressive to one another, the queen attacks the workers and eats their eggs, they retaliate and fight among themselves. The violence goes on until just one queen remains – the only member of the old colony that will survive. She flies off, mates just once and then hibernates for the winter.
