"When textbooks explain the basics of how vertebrates develop, they normally use amphibians as their core material," says Professor Jonathan Slack at the University of Bath. "That's where the knowledge is, as it's relatively straightforward to do the experiments with high precision." While many species of amphibian have been studied, one stands out: the African clawed frog, Xenopus laevis.
Xenopus leaps to the fore
The history of experimental embryology stretches back to the 1880s, when German embryologist Wilhelm Roux destroyed one cell of a two-cell frog embryo. Half-embryos resulted - left or right side only - indicating that the two cells had different fates.
German scientists were the undeniable world experts in embryology in these early years, using newts, salamanders, frogs and sea urchins. Some fundamental insights into development emerged, in particular the demonstration of 'induction' by Hans Spemann. He showed that tissue contacting the developing eye socket, the 'optic cup', was induced to form the lens of the eye. Most famously, in his 1924 'organiser' experiment, he showed that if a specific piece of tissue was moved from one side of the embryo to the other, a complete second body axis could be induced.
Yet embryological research before World War II was hampered by a lack of eggs, which had to be collected in the wild. "In the spring, researchers would go and find frogs or newts, take their eggs and do a mad rush of experiments for a few weeks," says Professor Slack. "They would then spend the rest of the year dissecting the outcome."
Xenopus laevis was to be the saviour of the egg-starved scientists, but it rose to prominence for another reason altogether. In the 1930s, it was discovered that a female Xenopus would ovulate if injected with the urine from a pregnant woman - the hormone chorionic gonadotropin being the active ingredient. For a while in the 1940s and 50s, this was the only available pregnancy test, and many hospitals kept Xenopus. Not all were vigilant in keeping the frogs, however - "they are very good at escaping!" says Professor Slack. A number of wild populations were established in California, South America and, in colder climes, in South Wales near Swansea.
"The pregnancy test got people used to the idea that you could get eggs whenever you wanted them," says Professor Slack. "We still use human chorionic gonadotropin to induce ovulation, and we can get lots of eggs, all year round." Coupled to the hardy and robust nature of the frogs, making them easy to keep, Xenopus laevis has proved an ideal model organism.
From the 1950s onwards Xenopus gradually become the organism of choice for developmental studies. "As soon as you start to do molecular things - and in the 1980s molecular biology became essential for studying development - you absolutely have to agree on one species," says Professor Slack. "You need standard reagents and protocols, and to pool knowledge; to have more than one amphibian being studied would dilute the effort."
Xenopus laevis, the African clawed frog, has proved to be an ideal model organism
The large size of the eggs and embryos make them ideal for microsurgery - the core techniques of experimental embryology. Grafting, first developed in the 1900s, involves taking a piece of tissue and putting it somewhere else in the embryo, while 'explant cultures' involve culturing fragments of embryonic tissue. "These experiments are very easy to do with high precision in Xenopus, and are still useful today," says Professor Slack. "They tell you a lot about when cells become committed to become a certain type of tissue, and about the signalling centres that drive commitment."
While Hans Spemann had discovered a signalling centre back in 1924, and the continuing activity of killed tissue convinced people that induction must be due to chemical signals, it was not until the mid-1980s that the first signals were actually identified. Using Xenopus embryos, Professor Slack and Professor Jim Smith (now at the Wellcome Trust/Cancer Research UK Institute in Cambridge) showed that 'inducing factors' called fibroblast growth factors and activins are secreted by signalling centres, leading to the patterning of the surrounding regions of the embryo.
Other families of signalling molecules have subsequently been identified (the Wnts, bone morphogenic proteins, hedgehog proteins and other curiously named factors). Each family contains several members, and the patterning mechanisms - and factors involved - have turned out to be more or less the same in other vertebrates as they are in Xenopus.
"We now know that the signals work by forming gradients, developmental states are combinations of transcription factors, and the expression of those factors is turned on by the signal gradients," says Professor Slack. "It's been good to be in a period of science when you start out with a black box and 30 years later you have the answer."
A world without genetics
Perhaps surprisingly, Xenopus laevis has 'no genetics'. This is partly because it takes a year for females to reach sexual maturity, making breeding experiments impractical. Worse, unlike diploid flies, worms and fish (where each cell contains two copies of each gene), Xenopus is allotetraploid (four copies of each gene).
This would make it incredibly difficult to knock out a gene. Xenopus researchers have therefore developed methods for producing 'dominant negatives' - specific inhibitors of proteins within the frog. These can be more informative than loss-of-function mutations because they tend to block a whole family of proteins rather than just one gene product.
Another popular technique is to force Xenopus to make too much of a particular protein, by injecting messenger RNA into early embryos. And in 1996, Enrique Amaya and Kris Kroll found a way to introduce new genes into embryos through the sperm. This has proved very useful, enabling molecular studies of organ development, regeneration and metamorphosis.
In recent years, a close cousin of X. laevis, Xenopus tropicalis, has also entered the developmental fray. Tropicalis is smaller than laevis, has a shorter life cycle (it matures in about six months), and has a small diploid genome (which is being sequenced). There are hopes that tropicalis will have all the advantages of laevis and genetics as well. On the downside, Xenopus tropicalis is trickier to keep and do experiments with.
"The essence of development is the creation of complexity from a simple starting point - a fertilised egg," says Professor Slack. "We now have a good idea of how the interaction of different signals patterns a three-dimensional object - the embryo. We don't know so much about how the body organs develop, how the tadpole metamorphoses into the frog, or how limbs or tails regenerate. But these are all areas to which Xenopus and other amphibians can contribute, complementing research on other organisms."
Moreover, there are still questions of development that Professor Slack regards as almost completely unaddressed. "I'd really like to know how the timing of development is controlled. Is there a master clock, or does one event cause another? How are the relative proportions of different structures controlled? And why does the whole animal have a certain preset size? Just look at dogs - why is a chihuahua 80 times smaller than a St Bernard? So you see, despite all the progress in developmental biology, there are still lots of black boxes out there."
Other discoveries using frogs
Although embryology has proved to be the frog's tour de force, many other fields have benefited. Here are a few highlights.
In 1952, Robert Briggs and Thomas J King cloned northern leopard frogs using a method of nuclear transfer. Briggs and King's experiment was similar to that envisioned - and piloted using salamanders - by Hans Spemann in his 1938 proposal for a 'fantastical experiment'. Later, John Gurdon extended this work and showed that nuclei from differentiated cells could support development, although less well than those from early embryos.
Verification of messenger RNA
While the existence and role of messenger RNA (mRNA) was known in bacteria, in the 1960s it was still debated whether it also existed in vertebrates. Taking haemoglobin mRNA from immature red blood cells and injecting it into a Xenopus oocyte, John Gurdon showed that the haemoglobin protein was indeed produced. Producing proteins in Xenopus oocytes has proved to be extremely useful in cell biology, in particular for the study of receptor proteins.
As they develop outside the mother, frog eggs are well stocked with the proteins needed to drive the development of the embryo. Studies of these processes has shed considerable light on the processes involved in cell division - termed the cell cycle.
Professor Jonathan Slack research page