The Portrait of a Fly (part 1): Come fly with me

For more than a hundred years, scientists have used the fruit fly (Drosophila melanogaster) to study the fundamentals of developmental biology and genetics.

But as biological understanding and techniques have improved, we are now able to do sophisticated genetic experiments in animals further along the evolutionary scale, such as mice.

What role, then, for the fly today?

At the entrance to the Fly Facility at the University of Manchester, there is a poster on the wall. It proudly proclaims that Drosophila (a genus of fruit flies) has won six Nobel Prizes. All were won in conjunction with human collaborators, of course, who took the glory (not to mention the cash).

The Fly Facility, which is funded by the university and the Wellcome Trust, brings together the university’s collective expertise in fly husbandry and research techniques. It supports a community of 13 research groups who consider the potential of the fly stronger than ever for informing us not just about basic biology but about human diseases as well. I suspect they dream of one day adding a seventh Nobel to Drosophila‘s trophy cabinet.

Dr Andreas Prokop, the head of a laboratory in the Faculty of Life Sciences and an unashamed proselytiser for fruit flies in research, is keen to show them off. He takes me around the Fly Facility’s specialist microscopes and the breeding rooms where the flies are grown. One room is kept at the scientific standard temperature of 25°C, while the other is at 18°C for experiments in which researchers want to slow down Drosophila‘s usual development cycle (this works both ways: some experiments are done at 29°C to speed up development).

Inside each breeding room, the shelves are full of trays and incubators with different fly stocks. The flies live in plastic vials – one genetic stock per vial – with food and a cotton wool plug to stop them flying away. These are no house flies: each adult fly is only a couple of millimetres long, their bodies are pale with black markings and they have bright red eyes.

Fertilised female flies can lay hundreds of eggs over several days. Embryonic development takes 21 hours (at 25°C), at which point each egg hatches to release a Drosophila larva. After a few heady days of continuous feeding at the bottom of its vial, the larva wanders up the side of the vial, away from its food source, and forms a hard shell called a pupa. At this stage, all of its organs are broken down and restructured in their adult forms. Eventually, a mere ten days after the egg was laid, an adult fly emerges from the pupal case.

Pretty fly with a white eye

The first of the Drosophila Nobels was awarded in 1933 to Thomas Hunt Morgan, whose ‘fly room’ at Columbia University laid the groundwork for pretty much all Drosophila research that has followed.

Morgan was interested in problems of development and how heredity might help him understand them. He spent two years searching for Drosophila with clear, heritable variations in characteristics so that he could start to unpick how the differences had been inherited. At last, in 1909, he found white-eyed male flies among his normally red-eyed stock. Crucially, eye colour was a sex-linked variation, which meant the gene responsible had to be on the Drosophila sex chromosome (Drosophila have four pairs of chromosomes, whereas humans have 23; in both species, the sex chromosomes are either XX for females or XY for males). It was the first unambiguous link between a chromosome and a characteristic.

Morgan’s style in the lab was democratic. It spawned a school of pupils and collaborators who spread the techniques he developed around the world, training the next generation to do the experiments that would connect characteristics to chromosomes and, in time, to specific genes.

One of Morgan’s protégés was Hermann Muller. In the 1920s, Muller began looking for ways to change genes artificially rather than relying on spontaneous mutations. His research showed that X-rays caused widespread genetic mutation and chromosomal damage. It won him Drosophila‘s second Nobel, in 1946.

It was some time before Drosophila‘s next Nobel, but then it won three in the same year. Innovations in the 1970s and 1980s meant that DNA could now be manipulated directly. Using such techniques, Ed Lewis discovered genes that controlled the fly’s ‘body plan’, whereas Christiane Nüsslein-Volhard and Eric Wieschaus identified and classified genes that were crucial for early development. At around the same time, studies in other species showed that these genes had closely related analogues in other vertebrates. Such genes are described as being ‘conserved’, and it confirmed that Drosophila research had direct relevance to human biology.

Lewis, Nüsslein-Volhard and Wieschaus shared the 1995 Nobel Prize in Physiology for their work. As news of their success spread around the world, it’s possible Jules Hoffmann would have taken inspiration from it. In 1995, Hoffmann was director of the Institute of Molecular and Cellular Biology in Strasbourg, part of France’s National Centre of Scientific Research. At the same time, he was attempting to unravel the fly’s immune system.

Humans have two lines of defence against infections, one of which is conserved from simpler organisms such as Drosophila. This is the innate immune system. Hoffmann discovered that a gene called Toll is central to the fly’s innate immune system, controlling the response to certain bacterial infections. Soon, related genes – so-called ‘Toll-like receptors’ – were discovered in other species, including humans. As well as contributing to our innate immune system, human Toll-like receptors activate our second line of defence, the adaptive immune system, which is slower to respond but incorporates components such as antibodies that ‘remember’ infections and establish immunity.

Reflecting on his work at the Euroscience Open Forum conference this summer, Hoffmann said that “no one in the medical field would have believed the systems [of fly and human] were so close in evolution”, but work in both species was needed to uncover the control mechanisms for our immune systems. Hoffmann shared the 2011 Nobel Prize with two of the scientists who had subsequently worked on Toll-like receptors in mice and people.

Despite such an illustrious history, Prokop says researchers began to drift away from Drosophila in favour of other species, most notably the mouse, in which exciting new techniques were continually being developed through the final decades of the 20th century and into the 21st. As genetics became easier to do in mammals, many scientists thought that experiments should be done in as highly evolved a species as possible.

However, Prokop thinks there is more to be gained by combining these new molecular and cell biology techniques with genetics in the fly. Mice, and indeed humans, tend to have multiple genes with overlapping functions, so when you ‘knock out’ a gene to see the effect on the organism of not having it, the results can be ambiguous. Drosophila has little or no such redundancy in its genes, plus the added advantages of having a rapid reproduction cycle and being relatively cheap to maintain.

Return of the fly

Prokop’s own research focuses on the cytoskeleton: the architecture of actin filaments and microtubules woven into structural networks inside all our cells. Far from being static scaffolding, however, these networks are dynamic, constantly rearranging and interacting with other proteins.

“The cytoskeleton is involved in every aspect of every single cellular function you can think of,” he says. Many of its components have been studied for decades but many fundamental questions remain.

“We know what these components do biochemically, but in the cellular context, we are not sure why those interactions are necessary. For example, the protein Tau was discovered in 1975. Today, we know it has a role in Alzheimer’s disease, but we still don’t really know what Tau is doing in the cell in the first place.

“One thing that blocks a lot of this type of research is that we do not have a proper understanding of the cell biology.”

Numerous proteins are known to bind to actin and microtubules, which implies some degree of regulation. But how do all these different functions and regulators combine in one coordinated system of cytoskeletal function? Prokop’s approach to finding an answer is to narrow down the question. He is studying the cytoskeleton in one specific context and, from that, aims to work up a general conceptual understanding that can be tested in other contexts and other species.

“I made three decisions,” he explains.

“First, study axonal growth. This is the best context for understanding what the cytoskeleton does because we know the principal roles of actin and microtubules, and there are drugs we can use to manipulate axonal growth – there aren’t many other places where we can get such meaningful results.”

Axons are the wires and cables of the nervous system. They connect neurons, the most important cells in the nervous system, and carry electrical signals between cells. Problems with axonal development lead to impaired cognition; damage later in life, from strokes or dementia for example, can cause sustained paralysis, permanent loss of basic brain functions such as coordination, or changes in personality.

Drosophila neuron

Axons grow out of neuron cell bodies, led from the front by specialist structures called growth cones. Chemical gradients – varying concentrations of substances that attract or repel the growth cone – determine the direction of growth and help each axon find its target. How these external cues are translated into changes in the behaviour of the growth cone remains unclear. What we do know is that the actin and microtubule structures of the cytoskeleton are constantly rearranging at the axon tip and that they must be regulated in response to certain signals.

“There are probably hundreds of genes that regulate the cytoskeleton,” Prokop says. “But the really essential ones are relatively few. Those genes, and the targets they regulate, present us with a common mechanism dependent on a limited number of factors. Focusing our research on these genes and proteins within the context of axon growth is a good strategy for learning about their function in general. It is simpler than trying to tackle the entire cellular machinery.”

Laud of the flies

His second decision was to work on Drosophila, for the usual reasons: “Efficient genetics, low redundancy. If you knock out one gene, there is less opportunity for compensation than there would be in higher species. Plus you can take out combinations of classes of genes simultaneously and it would be extremely challenging, even impossible, to do that in the mouse. In flies, it takes between a couple of weeks and a few months, so you can create and generate possibilities, test hypotheses.”

Prokop stresses that this is no longer the simple fly genetics of years gone by, when a gene would be knocked out and the gross effects observed. Moreover, combined with modern Drosophila genetics are other sophisticated techniques, developed and well established in mice, for studying what is going on inside cells. Many of these techniques, which have now been adapted for use in fly neurons, visually label specific proteins within cells so that researchers can watch the dynamics of microtubules, actin and other proteins as they operate.

I see this for myself when Dr Natalia Sánchez-Soriano, a postdoctoral researcher in the Prokop laboratory, shows me around. As well as looking at neurons in developing fly embryos, they grow Drosophila neurons in culture. “It takes around 30 fly brains to make four cultures,” she explains, holding out one of many round agar plates, a few centimetres in diameter. “Within six hours, the neurons have grown enough to start analysing them.”

Cultured neurons lack the external signals that would normally direct their growth in the fly but their cell machinery is intact and axons grow, although in a less organised way. Sánchez-Soriano puts a plate under a microscope and focuses on a small area in the middle. At 10x magnification, I can see clusters of neurons on the screen.

“These are young cultures,” she says, showing me the cell body of a neuron and the axon growing from it. We zoom in. The neurons have been marked with fluorescent markers. “Looking in green light, you see the microtubules. If we look in red, you can see the actin.

“I always find it amazing to look at one of these cultures where they have formed a network, a mature network. It looks very messy because they are cultures but you can see how they connect. Now you have this complex neural network with lots of synapses, which are the contacts between the neurons.”

Microtubules are found in abundance at synapses, so the fluorescence also reveals the neurons’ connections. I can see synapses all the way along the axons, not just at the tip, which is how I remember having seen them portrayed in simplified textbook diagrams.

We are supposed to be going to lunch with other scientists from the lab. They hover at the doorway, but Sánchez-Soriano waves them away – she wants to show me one more slide.

B0002442 Axons in living embryo, confocal image

Axons in a living embryo

“This is a Drosophila embryo. It’s completely different to the cultured cells – the axons of the neurons are all organised. These axons connect to the muscles; their cell bodies are in the spinal cord. We can see where these axons are growing, and we can be very specific about developmental time. We can look at embryos after 21 hours of development and the axons are always in these positions, so we can detect tiny changes in axonal growth.

“The beauty of this system is that we can go back and forward all the time between neurons in the organism and in culture, and that’s what makes it very attractive for me. And it’s easy to have expertise in both. That makes it a really nice model.”

Prokop’s third decision was which component of cytoskeleton regulation to focus on. He chose a protein called Shot, which, he explains, “sits at the heart of the machine”. Shot is one of a family of proteins called spectraplakins, which have so many different functions that they are sometimes also called ‘the Swiss Army knives of the cell’. In this context, Shot regulates microtubules during axonal growth.

In a normal growing axon, microtubules are constantly joining together, elongating and pushing against the cell membrane in the direction of growth. Actin networks provide some control over microtubules by blocking microtubule extension in the wrong direction. Proteins like Shot that bind actin and microtubules are therefore likely to be involved in regulating that interaction and, indeed, without Shot, microtubules become disorganised and axons are severely shortened (and when they do grow, it is often in the wrong direction).

The very model of a research model organism

Having shown that axon growth cone behaviour, cytoskeletal dynamics and the underlying functions of actin and microtubule regulators in Drosophila are well conserved with higher animals, Prokop’s group is using the Shot system to discover the mechanisms by which it all hangs together. They work mostly in the fly but regularly do analogous experiments in mice to confirm the general relevance of their findings.

They have already begun to show how Shot performs its role in cytoskeletal regulation during axon growth. In research published this summer, Prokop and his team showed that Shot links the tips of microtubules to actin networks, providing a physical connection that points the microtubules in the right direction. It also binds along the length of the microtubules, providing structural stability as they push out along the path of the growth cone.

Understanding how Shot works at a molecular and cellular level in one biological process in Drosophila will inform research on many other aspects of spectraplakins’ functions. For example, these proteins have roles in neurodegenerative diseases, skin blistering, and cell migration during wound healing and brain development.

In humans, the equivalent of Shot is dystonin, which causes neurodegeneration if it is impaired. In mice, loss of the dystonin gene in neurons produces exactly the same effect as loss of Shot in Drosophila: the microtubules are disorganised and unstable. Studying Shot may therefore be relevant to explaining the processes of neurodegeneration.

“We are beginning to study neurodegeneration in the Petri dish,” says Sánchez-Soriano. “We can cultivate neurons for a month, two months – they make connections and have active synapses, so we can look at matters that affect more mature neurons and study how they become disorganised and degenerate.”

Prokop’s model system certainly seems to be on the right track. “If you can establish a concept in Drosophila neurons, it is easier to verify it in other cell types,” he says. “Drosophila is an efficient system to narrow down what you are looking for. Then you can go and look for it in higher animals, including humans. Our system can be used to develop the whole concept rather than piecing together evidence from lots of different models and trying to construct a concept that way before seeing if it maps on to other species.”

When, a little later than planned, I go to lunch with members of the lab, I discover that the postdocs and PhD students are as evangelical as Prokop in championing Drosophila. These researchers are the latest generation of TH Morgan’s intellectual descendants, eager to celebrate the fruit fly’s contribution to science and persuade you that it is the pre-eminent model organism.

Prokop sums up the philosophy: “The task of Drosophila is to produce ideas that can then stimulate research in mammalian systems. That’s what it has been doing for a hundred years, and that’s what our system can do too.”

In part 2, ‘Fly on the wall’, which will be published on Friday, I visit a second Drosophila lab at the University of Manchester to see how the fly is actually used in experiments, and how relevant they can be for understanding human diseases.