Out on a Limb
This essay was written by James Briscoe and was first published in the 2002 Mill Hill Essays.
The mechanisms by which fertilized eggs turn into embryos and eventually into full-grown animals has fascinated biologists for generations, not only because of the fundamental nature of this process but also because this knowledge can help to explain the causes of many diseases – from birth defects to cancers. As recently as thirty years ago, however, we had only a superficial comprehension of embryo development and today there is still much to discover and understand. An example is limb development, the process by which arms and legs are formed. Many researchers have endeavoured to discover how this process works. I was reminded of this at a recent lecture in my Institute.
It was the usual Tuesday afternoon seminar. At 4.30pm, as on most Tuesdays, we filed into the lecture theatre on the ground floor of our building. The room was filled with the familiar hushed chatter of gossip and the smell of coffee from the machine outside. I took my regular seat, near the back, just in case the speaker was boring and I needed to make a quick exit. The hands on the clock at the back of the room ticked the customary three minutes past the advertised start time and the conversation began to die down as the host of today’s speaker rose to make the introduction. As is usual, the speaker was a visiting scientist invited to talk about their latest research. Today it was a professor from Harvard University in Cambridge, USA. He had made a name for himself a decade ago being among the first of a new generation of scientists to study the genes involved in embryo development. Before that, in order to study development, researchers had to perform intricate surgery on tiny embryos to remove or replace small pieces of tissue, and then look at the consequences of the surgery through a microscope as the embryo develops.
The professor began his seminar with descriptions of some of these surgical experiments, introducing his particular interest in the field – limb development. His first slide showed how limbs form. It was an image that I instantly recognized from a textbook I had studied as a student. Limbs start off as small buds growing out of the main body of the embryo at the places where shoulders and hips will eventually be. These buds appear first like small boils on the side of the animal. Over time the cells that make up these buds divide and increase in number and the buds elongate (looking like sausages), growing away from the body. Later still, as the bones develop, a recognizable limb emerges, the upperlimb bone closest to the body, then the forelimb bones and furthest from the body the bones that make up the hand and foot. Two things must be in place for babies to be born with the bones in their arms and legs in the right order. First, the embryo must have full control over the growth process; second, the cells making up the developing limb must have some way of distinguishing where they are within the limb. Working out the way that these two systems work is a conundrum that has intrigued scientists for years.
The first ideas of how limbs grow were developed thirty years ago. After studying chick embryos scientists concluded that when limbs grow it is only those cells in a region at the tip of the limb, the part furthest away from the body, that multiply, rather than all of the cells in the developing limb. In order to divide and grow, the cells need to receive a chemical signal that is only produced at the tip of the limb. If this signal can only travel a short distance then only the cells near the source of the signal will get the message to divide and grow. In order to test this theory researchers removed the cells that produced the signal to grow, before an embryo’s limbs had developed very much. As they expected, growth of the limb stopped and the embryo only developed an upper arm. If they waited a little longer before removing the signal-producing cells, both the upper and the lower arm had formed but the digits did not develop. The researchers concluded that the different regions of the limbs grow in a distinct order: starting at the shoulder or hip, first the upper arm or leg develops then the lower arm or leg and finally the fingers and toes. This was a simple and elegant mechanism and became known as the ‘progress zone model’. It has underpinned research in this field of developmental biology.
As the speaker’s introduction continued my mind began to wander. I started thinking about the experiments underway in my own laboratory at the moment. I noticed a colleague across the room who had promised me a reagent and I wrote her name on my notepad to remind myself to speak to her after the seminar. I looked back up at the speaker and saw that he had displayed a new slide. The professor had now moved on to talk about some new research his team were engaged in. I scanned the slide quickly, impatiently trying to anticipate what he was going to say – I always like to do this, trying to make my own interpretation of the data to see if my perceptions match the speaker’s. But this time the slide was a surprise: what I was looking at was quite unexpected. Had I missed something when I drifted off? Someone behind me whispered to himself “Huh? Look at that.” The slide didn’t make sense because the research results it showed did not fit with the theory of the progress zone model.
The professor explained that a researcher in his lab had repeated the 30 year old experiments but this time had used new molecular biology techniques to analyze the results in much closer detail than had been possible before. What he saw was striking – the new techniques revealed something that the old researchers could never have seen. When they removed the cells at the tip of the bud that produced the growth signal, a small group of cells just underneath the tip also died. The size of the group of cells that died was the same regardless of the stage of the embryo’s development. When the limb was small, at the early stages of the embryo’s development, the cells that died accounted for a larger proportion of the limb than when the limb was larger. To explore further what was happening the researcher placed a tiny spot of a fluorescent dye onto a small group of cells near those that produced the growth signal. The dye enabled him to follow these cells as the embryo grew and to see where they ended up. The results agreed with his previous discovery: the marked cells, which started off close to those producing the growth signal, ended up in the fingers and toes. If however he removed the signal producing cells, the marked cells disappeared. The established theory, the progress zone model, would have predicted that these cells would have ended up in the upper arm or leg, but instead, it looked as if they were dying.
The lecture theatre was silent. The professor then suggested a new model to explain these findings. Perhaps all of the cells making up the limb were produced at the same time and then multiply at different rates to form a complete limb, instead of the limbs forming in distinct order from the shoulder outwards, with the upper arm or leg forming first and the fingers and toes last. This was a radically different model. I looked around the room and it was obvious the speaker had everyone’s attention. Some people were writing furiously, others were staring at the data on the slides, thinking intently. For all of us, the textbooks were being re-written. Looking at the speaker’s calm manner it was clear that in his laboratory the textbooks had been thrown out many months ago. To substantiate his idea, the professor showed some more cell marking experiments. Cells had been marked by the researcher at particular positions in the limb at different points in time. The progress zone model predicts that cells marked early on should end up spread throughout the limb, as they would have all been close to the growth signal. Cells marked later however should be confined to only one region, either upperlimb, forelimb or digits, as they would have already been exposed to different amounts of growth signal as they moved away from the cells producing it. In the new experiments all of the marked cells, regardless of when they were labelled, were only found in one region.
The professor and his team were not alone in questioning the established progress zone model. A research group in San Francisco had used gene technology to breed mice that did not produce the growth signal. They found that if the growth signal was missing entirely, no limbs formed. But if the growth signal was produced for a short time, a shortened limb grew which, on careful inspection, appeared to have all bones of all of the regions – upperlimb, forelimb and digits. The progress zone model would predict that only the upper arm or leg would be formed. These results were consistent with the model that the professor was proposing. The professor finished his talk and invited questions from the audience. A couple of people asked technical questions about experimental details. Then after a short pause, someone sitting on the other side of the aisle asked the question that I, and I’m sure half the room, wanted to hear answered. “If your results are correct, where does this leave the progress zone model – is it wrong?” The professor nodded and smiled. He clearly expected this question and had thought of the answer well in advance. “It might mean the progress zone model is wrong”, he started, “but it might mean that it needs to be modified to account for our findings. Or we could be wrong, of course.” He laughed at his own comment. “But if we are right, then we have to understand the mechanism and the genes that control the process if we want to understand genetic defects that affect limbs in humans as well as the impact of drugs such as thalidomide. I don’t know the answer, but what I do know is that we have more questions now than we thought we had a few months back”.
The audience of the seminar left the room, talking animatedly about what they had just heard. The rug had been pulled from under an established theory, but as we talked in the coffee room afterwards it was clear none of us were sure whether what we had heard was enough for the foundation of a new theory. It felt surprising, yet also liberating and exciting. And in the next Tuesday seminar, I sat closer to the front of the room.