The Portrait of a Fly (part 2): Fly on the wall
Drosophila: the model model organism; the humble fruit fly with a noble (not to mention Nobel) place in the history of science. Having learned about its importance in genetics and developmental biology, I wanted to see Drosophila in action.
At a lab in Manchester, I did just that and discovered that the relevance of such research to human health can be unexpectedly direct.
“What do you want to see?”
No idea. “What do you have?”
“I could show you first instar dissection, third instar dissection, electric shock….”
Dr Richard Marley is a research technician in the Baines lab at the University of Manchester’s Faculty of Life Sciences. In the face of my obvious ignorance, he takes a Drosophila larva (a pale, segmented, maggoty creature about one millimetre long), jiggles it in mounting fluid to wash off the leftovers from its feed – “They’re fantastic eating machines,” he says – and then, in two deft movements under the microscope, pulls off the larva’s body and head, leaving only its intact nervous system. Watching on a digital camera linked up to the microscope, BBC Radio 2 playing inoffensively in the background, the operation reminds me of nothing so much as shelling a prawn.
“This is the tricky bit,” he says. “You want to get it down first time.” He stops talking to put a small tube in his mouth. It is a pipette that he uses to put glue on the isolated nervous system, which he has laid prone on a slide, ready for the next part of the demonstration.
The slide is placed under a different microscope and I am encouraged to have a look through the eyepiece. At 60x magnification, I can clearly see the larva’s neurons – the most important cells in the nervous system. The neurons have been labelled with a green fluorescent marker, and I can see thin green processes coming off each cell body: these are axons, the long extensions that neurons send out to communicate with other cells.
I look up from the microscope to find the whole lab – some half a dozen scientists – watching me as intently as I was observing the neurons.
I have come to Manchester to visit research groups that use Drosophila to explore how neurons grow and develop. Drosophila has been a staple of biological research for more than a hundred years – their rapid reproductive cycle and relatively simple genetic make-up made them the go-to organism for genetic studies in particular. More recently, techniques in cell and molecular biology developed in other animals have been adapted for use in Drosophila, adding another dimension to its scientific usefulness.
In the Baines lab, they use electrophysiology – measuring the electrical currents flowing in and out of cells – to understand how Drosophila’s neurons function. Although a human being is more complex than a fruit fly, the biological mechanisms that underpin neuronal development are remarkably well conserved across species, so findings in Drosophila are usually significant to human physiology as well.
For example, Dr Wei-Hsiang Lin, a postdoctoral researcher in the lab, is using Drosophila to study epilepsy. I wouldn’t have thought that fruit flies would be particularly prone to epilepsy, but a bit of genetic engineering makes them respond to an electrical stimulus with a spasm that is a good model for the human disorder.
Neurons use a combination of electrical signals and chemical neurotransmitters to communicate with each other and with other types of cell. This system is dependent on the movement of electrically charged ions across the cell membrane. They move through ion channels: carefully controlled gates in the membrane that are usually specific to one type of ion, such as sodium or potassium, barring passage to all others.
As with all of the proteins in our cells, the instructions for making ion channels are held in our genetic code. Rogue mutations in genes that code for ion channels can lead to epilepsy, although in most cases the precise cause is never identified. There are several types of ion channel – even several types of sodium channel – so a fault in one gene could potentially be compensated for by related ion channel genes adapting their activity. Most epilepsy is probably caused by a small, hard-to-detect irregularity in the ion channels which, in certain circumstances, sets off a cascade of electrical signalling events that amplify the irregularity until it causes a seizure.
Drosophila only has one gene for sodium ion channels but its instructions can be interpreted in different ways, resulting in different ‘isoforms’. The balance of two particular isoforms of sodium ion channel present in a cell’s membrane determines how that cell behaves in response to electrical stimulation. Increasing the proportion of one isoform relative to the other in Drosophila neurons makes the insect susceptible to prolonged seizures.
To investigate epilepsy, therefore, Lin uses larvae with genetic mutations that tip the balance of sodium ion channel isoforms towards prolonged seizures. A live wire placed on a larva induces an extended period of seizure-associated paralysis. Normal Drosophila recover in about 30 seconds; the genetic mutants take much longer. Feeding the mutant larvae with phenytoin – a human antiepileptic drug – rescues them from this response, however, and they behave more like normal Drosophila again.
A gene called pasilla controls the balance of these isoforms in Drosophila neurons. The human equivalent of pasilla is Nova1, which has been implicated in epilepsy. That confirms that this model of epilepsy is relevant and potentially very useful.
In a different experiment, Lin feeds adult flies phenytoin and studies the behaviour of their offspring. These larvae, which have also inherited the mutations that make them prone to seizures, respond to the electrical stimulus as normal larvae would. The phenytoin from their parents’ diet must be helping them, even though the amount in their bodies is far less than would be required to rescue them by putting the drug in their own food. This suggests that their exposure to trace amounts of phenytoin was in exactly the right place at the right developmental time to overcome the effects of the mutations.
Lin is doing sophisticated genetic experiments to hone in on the crucial time and place where the mutations cause the problem. Through this work, he is pinpointing possible mechanisms for the development of vulnerable neuronal networks in the larvae. Comparing the developmental timeline of Drosophila and people may then indicate the crucial time and place at which epilepsy can develop in the human brain. Eventually, this understanding of the development of epilepsy could help in the design of treatments to prevent it.
A more immediate application is being sought by Kei Hattori, an undergraduate student at Nagoya University, Japan, here in the Baines lab on a placement. She is using Drosophila as a screen to identify effective antiepileptics. Hattori has 26 candidate drugs – some already in use for epilepsy, some in use for other conditions such as cancer, and some experimental compounds. She uses a chemical called picrotoxin to induce epilepsy-like behaviour in Drosophila larvae. She then treats them with different concentrations and combinations of drugs to see which are successful in reducing the severity of the response.
Some of the candidate drugs on the list are already known to be effective but are too toxic to be used clinically. Using more than one drug at a time makes it possible to use a lower dose of each drug, which often reduces their combined toxicity while retaining overall effectiveness. The Drosophila screen quickly picks this up: “You couldn’t test all combinations of all 26 drugs in mice,” Hattori tells me, “but you can do it in the fly.”
The focal point for neuroscientists at Manchester University is the newly built AV Hill Building, located towards one end of the main campus. Bridges connect some of the buildings here to protect researchers from the North-west’s infamously damp climate. Coming in to the AV Hill Building from a campus that despite bright sunshine is covered in large, cliché-confirming puddles, the first thing I encounter is the geekily named Synapse Café. Waiting for my coffee, I look up through the light, airy central atrium and see the open plan offices that fill one side of the floors above, with most of the labs on the other side.
The Baines lab, however, is in the basement. I’m assured the location is no reflection on the sociability of electrophysiologists but more to do with the need to minimise vibration and electrical interference. The lab is split into two distinct halves, as Dr Verena Wolfram, a postdoctoral researcher, explains: “We have loads of equipment – we do some molecular biology, but that’s the same as any other biology lab. Mostly we do electrophysiology. We have imaging microscopes to look at the neurons and, on the other side, we have our rigs.”
Everyone in the lab has their own ‘rig’ – a workstation with a microscope and all the kit for conducting electrophysiology experiments. “We each do recordings every day, pretty much the whole day,” says Wolfram. The recordings are of the electrical currents that flow across neuronal cell membranes.
As a neuron responds to incoming signals, it opens and closes ion channels in its membrane. The resulting flow of ions creates electrical currents. Ordinarily, this would change the voltage across the membrane, but in a type of experiment called voltage clamping, an electrode is put on to the neuron where it provides the current necessary to maintain a fixed voltage. The current from the electrode is easily measured and can be used to determine the current that the cell’s actions would have produced in real life but which cannot be directly measured.
Back on the imaging side of the lab, Wolfram shows me some Drosophila larvae that have been labelled with the green fluorescent marker. The marker is used to identify which ones have inherited the genetic profile a scientist wants to study. “Perhaps the mother has a mutation on one chromosome and the father another on a different chromosome. You want the offspring with both, but not all of them will inherit both mutations. We can label the ‘healthy’ chromosomes green and so we know it is only the larvae without any green that have both mutations. Or you can do it so the ones you want are green. Either way, you can then just pick out the ones you want.”
In the past, researchers only knew which mutations had been passed on when the insects reached their mature stage, the fly. “They had to add in additional, linked mutations that caused obvious physical changes that identified the flies of interest,” Wolfram explains. “For example, a mutation that produced short hairs, curled or scissored wings, dark bodies or different eye colours. But of course you can’t see these effects in the larvae and they’re what we need for our recordings.”
Looking down one of the imaging microscopes, I see two wriggly larvae, mouth parts chomping away as they feed on a recipe of cornflour, glucose, yeast and agar. Wolfram guides me through their anatomy. Each has two long sacs running backwards from its head – these are the salivary glands. Between them, about halfway down the larva’s back, is the nervous system, which in these larvae is covered in fluorescent green dots – the neuron cell bodies.
Within Drosophila’s simple nervous system, it is possible to identify specific neurons from their position in the body – the spatial organisation of neurons is remarkably consistent between each insect and each neuron has a name, much like the bones in the human skeleton. Even so, Wolfram admits that sometimes she is not quite sure which neuron is which, especially as the larvae get older, so she often labels them with fluorescence too.
Not all research in the Baines lab is on epilepsy – at one rig, Francesca Cash is working towards her PhD with a project to uncover the mechanism by which a potential new insecticide works (a rare example of Drosophila being studied not as a model organism but as an insect in its own right); at another, a scientist visiting from France is learning some of the lab’s dissection techniques so he can apply them to studying Drosophila’s rudimentary heart. However, everything here is founded on using electrophysiology to understand the fundamental principles of electrical signalling across all species, including people.
Back upstairs in his office, Professor Richard Baines explains his approach: “There are billions of neurons in the human brain but very few share specific sets of characteristics. Neuronal behaviour is all about frequencies and patterns of action potentials, the electrical signals that neurons send along their axons in order to communicate with other cells. But 12 action potentials close together produce a different response to 12 action potentials firing less rapidly, so clearly different neurons have different requirements.
“Ion channels play a part in determining what subtype a neuron is, but there are over a million subtypes and the entire human genome is only a few tens of thousands of genes. How is this diversity generated?
“Fundamentally, just a few types of ion channel can give rise to the huge diversity we see among neurons, the same way that four drinks in a bar can make numerous cocktails, depending on how you mix them.”
All neurons have the same types of ion channel available but they can have different combinations, locations and numbers of each. With relatively few genes needed to produce the ion channels, another small set of genes could determine how those ion channels are distributed in each cell. This level of regulation usually comes from transcription factors – genes that control many other genes – which are already known to be vitally important in determining properties of neurons such as which neurotransmitters they deploy, how they grow and what other neurons they connect with. Transcription factors could be the bartenders in Baines’s ion channel cocktail lounge.
However, once a neuron makes connections with other neurons, its place and role in the neural circuitry influences how it responds to the signals it receives. This capacity for change is called Hebbian plasticity, often summed up in the phrase ‘Cells that fire together, wire together’. Early in development, neurons compete to form signalling connections with each other. If a connection is made but never used, the axon will retract and try to form a more useful connection with a different cell. Similarly, a neuron firing action potentials at maximum capacity is no use because it cannot ever increase its firing rate, so it will change its properties to damp down its response.
Because neurons can adapt when part of a ‘live’ circuit, it has been unclear whether the cocktail of ion channels in a particular neuron is determined mainly by transcription factors or by its subsequent activity.
A recent study on which Wolfram and Baines were lead and senior author, respectively, suggested that the same transcription factors that control other neuronal properties do establish the initial mix of ion channels. They had found that a single transcription factor determined whether a neuron produced an electrical current associated with the movement of potassium ions. This was true even when they had blocked signalling activity throughout the nervous system. It means that this transcription factor controls the expression of a certain type of potassium ion channel.
While a lot of neuronal diversity could be predetermined by transcription factors, Baines is not sure that this would be enough to establish functioning circuits in the nervous system. His theory is that it gives the system a ‘head-start’, speeding up initial development that can later be fine-tuned in response to activity at the connections each neuron makes. The story might be slightly different in people, though.
“Normally, what we find in flies, we find in people – although it may have been built upon and adapted,” he says. “Drosophila may need more predetermination than humans because flies have such a short gestation period: embryogenesis is only about 21 hours in the fly.”
Intriguingly, transcription factors remain active in neurons after they have matured and no one is quite sure what they are doing. And, of course, there is the question of what is regulating the regulators: what determines how the transcription factors will configure each neuron? The complexity is daunting, but Baines is optimistic.
“One of the beauties of invertebrates is the small number and consistency of neurons,” he says. “Having the same layout of cells in each animal means we can compare them directly. We know they have different combinations of transcription factors and we are starting to interpret what that means for ion channels. Other groups are looking at how those same transcription factors determine a neuron’s choice of neurotransmitter or the way it connects with other cells.
“My contribution is functional electrophysiology. We had to take electrophysiology techniques, which were developed in very large cells, and combine them with genetics in Drosophila’s very small cells. I was the first to start doing that about 15 years ago. Now we’re getting a good understanding of how you get from neural stem cells to neurons to diversity in neurons.”
As I prepare to go, Baines leaves me with a parting shot: “There are brain surgeons operating on people with epilepsy,” he says, “who have little or no idea that what we’re studying in flies might one day help their future patients.”
It is a plea for greater appreciation of Drosophila as a model organism in science. Of course, while the fly continues to prove itself an elegant and valuable research tool, the credit really belongs to the scientists who do the experiments and interpret the results.
I was keen to see how fruit flies are actually used in scientific experiments; I’m grateful to everyone in the Baines lab for letting me come and be reminded that the power of any research model lies in the ways it is applied by smart, inquisitive and dedicated people.
Lin WH, Günay C, Marley R, Prinz AA, & Baines RA (2012). Activity-dependent alternative splicing increases persistent sodium current and promotes seizure. The Journal of neuroscience, 32 (21), 7267-77 PMID: 22623672
Wolfram V, Southall TD, Brand AH, & Baines RA (2012). The LIM-homeodomain protein islet dictates motor neuron electrical properties by regulating K(+) channel expression. Neuron, 75 (4), 663-74 PMID: 22920257