ComBio: DNA methylation is the queen bee’s knees

As Dr Ryszard Maleszka likes to say, honey bee queens are made, not born. When a new queen is needed by a colony, a chosen larva will be fed exclusively on a diet of royal jelly, the magnificently rich substance produced by glands in the heads of nurse bees, young workers who never made the cut themselves.

Quite why the workers choose a particular larva is impossible to ascertain, but once she is chosen, that larva will go on to a life radically different to her sisters, all of whom are clonally identical. She literally swims in a sea of royal jelly, Maleszka says, while her siblings are given a brief taste upon birth to get them started in life but then are fed with a normal mixture of pollen and nectar.

Not only is she larger than her sisters and is the only one able to reproduce, but the queen can live up to three years while the workers fade out after just over a month. She keeps sperm from one or many mates in her spermatheca, a storage organ, and chooses when to fertilise and not to fertilise her eggs, the minority unfertilised eggs going on to hatch as males. It is a lifecycle that has fascinated people for generations, and with the sequencing of the honey bee genome in 2006, there has been an explosion in understanding of the bee, an insect essential for the functioning of nature as we know it.

Before the honey bee genome was sequenced, the insect models available for genomic studies consisted of the old favourite, Drosophila melanogaster, and the malaria mosquito, Anopheles gambiae. There are now 21 insect species sequenced and two other arthropods, although 12 of the insects are Drosophila relatives, carried out for an evolutionary comparison of one of biology’s most studied organisms. The other nine represent both social and solitary species and different orders.

“The honey bee has been done in the proper way – we wanted to sequence the genome but to get as much biological information as possible,” Maleszka says. “From the moment of publication, which isn’t even two years ago now, there have been over 65 papers published on the bee genome and many of them are in high impact journals.”

One of the most important findings from the project is the existence in the honey bee of a full complement of genes for DNA methyltransferases, the enzymes that set in motion epigenetic changes in organisms. Epigenetics is now an enormously interesting field of study, highlighting that while nature provides the code, nurture is of the upmost importance.

Maleszka, from the Molecular Genetics and Evolution group in the Research School of Biological Sciences at the Australian National University, will address the ComBio conference on the subject of insect genomes and the genomic revolution. In particular, he will stress that in order to bridge the gap between genotype and phenotype, studying genetic and epigenetic systems on a global scale – and how the environment modulates and controls those systems – is essential.

---PB--- DNA methyltransferases

And the honey bee can help out. As a member of the honey bee genome consortium, he and his colleagues, especially Hugh Robertson from the department of entomology at the University of Illinois at Urbana-Champaign, noticed during the annotation process that the set of DNA methyltransferase genes found in mammals was also there in bees. In 2006, Maleszka, along with Robertson and Gene Robinson from Illinois and a team from Barcelona, published a paper in Science on a functional CpG methylation system in bees.

(Last year, Matthias Schaefer and Frank Lyko from the division of epigenetics at the German Cancer Research Centre in Heidelberg published a commentary in Bioessays on the fully active DNA methylation system in bees, highlighting the importance of this finding to human cancer research. As shown by several studies, DNA hypermethylation plays a role in certain types of cancer. An example is Sydney’s Robyn Ward and Megan Hitchens’ work on germline inheritance of hypermethylation of the MLH1 DNA repair gene in colorectal cancer.)

Of course, most suspected that many of the phenotypic alterations in social insects, such as those found in the queen bee versus the worker bee, were due to environmental factors under epigenetic control, but at the time there was no hard evidence.

One problem was that the model invertebrate genomes – Drosophila and C. elegans – seem to have lost many of the genes for DNA methylation. “So everyone thought insects don’t have DNA methylation but it turns out the Drosophila is an exception,” Malescka says.

“The Drosophila genome is in some aspects very similar to the bee genome, but it is a fast-evolving species that lost some of those genes because it didn’t need them anymore. It didn’t have this phenotypic plasticity noticeable in bees and other social insects. Drosophila is somewhat of a misleading organism in some aspect for broad insect studies.”

To the researchers, it came as a nice surprise that bees have a full complement of DNA cytosine-5-methylatransferases (Dnmts), as do some other arthropod genomes. In fact, it seems that many of the non-dipteran insects sequenced so far have the full complement, and that cytosine methylation occurs at cytosine-phosphate-guanine (CpG) sites, just like in mammals.

It is slightly different in insects – CpG methylation is found in transcription units but not at the three- and five-prime ends, and methylation seem to be less common in insects. Nevertheless, the discovery that honey bees use DNA methylation offers the opportunity to study epigenetics in a model insect, particularly as it affects development.

---PB--- Elegant experiment

In an elegant experiment published in Science in March this year, Maleszka’s team decided to see what would happen if one of those methyltransferase genes, in this case Dmnt3, which is the enzyme that adds new methyl tags to DNA strands and is a key driver of epigenetic global reprogramming, was silenced in newly emerged larvae.

“The larvae only have a very short time when they are responsive to the royal jelly,” Maleszka says. “It is up to about 48, maybe 60 hours, and then after that there is no way of changing the development trajectory. So we silenced the gene at the moment they were born to see what happened.”

What happened was that about 80 per cent of the larvae developed as queens, with fully developed ovaries, a situation completely reversed in the control group, where about 80 per cent emerged as sterile workers.

The experiment directly mimicked what would happen if the larvae were fed exclusively on royal jelly, suggesting that DNA methylation in bees is used for storing epigenetic information. It also seems that the use of that information can be differentially altered by what the bees eat, their nutritional input. As the team wrote in the paper, the study suggests that the flexibility of epigenetic modifications underpins profound shifts in developmental fates, with massive implications for reproductive and behavioural status.

Maleszka believes the honey bee is a beautiful system to study not only the functional overlap of DNA methylation in mammals and invertebrates, but also to try to understand the nutritional and environmental basis of epigenetic re-programming.

“I am assuming, and I think I’m right, that mechanistically, at the level of biochemistry, it is very similar. It has been shown already in other systems that you have the fundamental biochemical similarity in different species and then you have other levels of complexity that make the difference. But at the very basic molecular level I think we will see enormous overlap with the bees and mammals, and that makes the bee very useful.

“It is probably the only system where we know exactly the nutritional ingredients used for global genomic reprogramming and we can use this information for manipulating development and to study all of those events from the moment the process is triggered and development is changed.”

---PB--- Biological meaning

At ComBio, Maleszka will argue that while large-scale genomic projects are here to stay, single gene and single network analysis is largely heading for obsolescence. He believes that with so many genomes being sequenced, technology developing so fast and mountains of information so available, biologists are in danger of drowning in data.

“We have to begin to convert that massive amount of raw data into real biological meaning,” he says. “For many years people have been studying biology gene by gene, taking various genes and trying to connect them to various phenotypes and behaviours, but I think that is no longer an interesting approach. It is not going to generate the understanding of how such complex biological systems work.

“The work on the bees is a good example of where you are getting information from the genome and you identify individual genes, but now there is a global genome-wide control of development, through methylation and epigenetics for example, so what I will argue is that we not only have to try to bridge this gap from genotype to phenotype, but we also have to bridge the gap between genotype and environment. In many cases you can see evidence that the genome is being modulated by the environment and epigenetic controls are responsive. I think we need to look into those massive amounts of data and look at systems biology more than individual genes.

“The genomic is useful of course but looking at those genomes for the sake of sequencing them is not necessarily helpful. Reshuffling and channelling our resources into addressing specific questions and looking at global genetic and epigenetic systems and how environment modulates and controls it, that is helpful. We have problems now with obesity and diabetes and clearly they are environmental. That is why insects, and especially the honey bee in this case, are very useful as a model to identify some of those mechanistic switches that also exist in humans.”

Royals and Anarchists

Royal jelly makes the queen, and the so-called Major Royal Jelly Proteins (MRJPs) are probably responsible. Maleszka’s research has shown that MRJPs are encoded by a set of nine genes found in the same region as the yellow family, which play a diverse role in pigmentation, development and sexual maturation in insects. Yellow genes are found only in insects and bacteria, and it is thought they came from symbiotic bacteria by horizontal transfer.

Maleszka’s team has also shown that royal jelly proteins retain some of the ancestral roles associated with the yellow proteins and may serve as developmental regulators or activators of biochemical pathways.

In queen bee development, the selected larva receives an enormous amount of royal jelly, which is sensed by the gut epithelium and is metabolised by the fat body, the insect version of a liver. This then activates the insulin pathway, which then activates the juvenile hormone that controls many metabolic sub-networks in bees. This leads to a rapid demand for more nutrients and then the astonishingly rapid growth of the young queen.

In addition to looking at honey bee epigenetics and proteins, Maleszka and his team have been studying honey bee genes involved in learning and behavioural development. A decade ago, ANU put up some money to create a project in which the molecular and the behavioural levels were studied side by side. “The idea was to try to combine the power of genetics and molecular biology with the well-established behavioural studies on the honey bee at ANU and see if we can bridge those two levels of biological complexity and get a new idea of how behaviour is generated from the genome,” he says.

“The idea was to understand some of the genes involved in learning and memory but also those genes that generate social behaviour. The main attraction of the honey bee is that it is a social organism, so there is this extra level of biological complexity in comparison to the famous vinegar fly, Drosophila. Everyone was very anxious to understand how very similar genomes can generate such different outputs in social and solitary species.”

Another very interesting project his team is involved in is helping Professor Ben Oldroyd from the University of Sydney with technical aspects of his work on the mechanisms of social cohesion in bee colonies. Oldroyd has created a strain of “anarchistic” bees in which workers are able to lay eggs, and lots of them.

The anarchist bees will be useful for investigating how worker sterility is maintained in normal colonies, and to isolate and characterise the genes that control worker sterility in social insects. These genes may prove to be the semi-mythical genes for altruism, Oldroyd says.