Rediscovering Biology: Molecular to Global Perspectives
Genetics of Development Expert Interview Transcript: Nipam Patel, Ph.D.
Professor of Biology and Anatomy; Investigator, HHMI
Patel is a professor in the Department of Organismal Biology and Anatomy at the University of Chicago, and an associate investigator of the Howard Hughes Medical Institute. Using data from Drosophila development, Patel is studying how developmental pathways have been conserved or altered between various arthropods, and between arthropods and other phyla. Insights into the nature of developmental and molecular alterations will help researchers to understand the evolutionary changes in the mechanisms of pattern formation, and provide a molecular basis for analyzing the diversification of body morphologies and developmental mechanisms.
Interview with Nipam Patel, Ph.D. Patel is a professor in the Department of Organismal Biology and Anatomy at the University of Chicago, and an associate investigator of the Howard Hughes Medical Institute. Using data from Drosophila development, Patel is studying how developmental pathways have been conserved or altered between various arthropods, and between arthropods and other phyla. Insights into the nature of developmental and molecular alterations will help researchers to understand the evolutionary changes in the mechanisms of pattern formation, and provide a molecular basis for analyzing the diversification of body morphologies and developmental mechanisms.
To start out with, can you define “homeotic genes”?
Homeotic genes are genes that, when mutated, cause the transformation of one part of the body into another part of the body. It was a term first coined by Bateson, who actually didn’t just go out and make mutants, but he just sort of collected aberrations in the real world. He actually wrote an amazing book a little over a hundred years ago, where he just catalogued what he called “homeotic mutants” or “trans-formants,” where one part of the body was transformed into another. He didn’t know the basis for that, and now we know that — at least in a few cases-there are genes which, when mutated, cause that to happen to an organism and those have been termed “homeotic genes.”
The normal function of a homeotic gene is to define the part of the body that should be in its normal place along its anterior or posterior axis. So one way that you recognize a homeotic gene is that when it’s mutated-when it’s not functioning properly-then one part of the body is transformed into an inappropriate part of the body. They really only can be defined by seeing what happens when they’re not working properly. But their normal function is to make sure that the right things are in the right place.
Is it common for you to refer to those as “master regulatory genes” and can you talk about that?
Many people consider homeotic genes to be master regulatory genes. That’s a phrase that sort of refers to any set of genes that may have lots and lots of “targets,” or other genes that they turn on or off. So these sit sort of at the top, if you would, of a very large hierarchy of genes that are really controlling the morphology and development of an organism. There seems to be a relatively small handful of genes that we give this title of master regulatory gene.
Can you clarify the difference between homeotic genes and Hox genes?
The homeotic genes were first described in Drosophila melanogaster-the fruit fly. When people started finding these genes in other organisms, they started nicknaming them “Hox genes,” which, for some people, stood for “homeotic cognate,” so the abbreviation became “Hox.” Eventually, everyone decided that all of these homeotic genes, whatever organism they were in, one nickname they would give them was “Hox genes” and that made it very simple. So really, the only people that call their genes homeotic genes still are Drosophila people, but even now, more and more people who are working on flies simply call them a Hox gene. And that’s sort of a universal name for the whole family of homeotic genes in all animals.
Can you talk about how Hox genes are fundamental to body segmentation in Drosophila development?
How do you properly pattern an embryo? That question has been studied very intensively in the fruit fly. Embryogenesis in that insect only takes about one day, about 22 hours. In that time, you go from a single fertilized egg of a single cell into a larvae. Of course, they don’t develop directly into adults, they have a larval stage. We understand an enormous amount about how that embryo is patterned. A lot of it occurs in just the first few hours-the first 2 or 3 hours of development. Essentially, there’s a cascade of genes which is forming the axis along both the anterior and posterior-the front and then the sort of dorsal to ventral, or the belly to the back of the animal. And these gene position along the embryo. What the homeotic genes do is they’re one of the readouts of that position. They’re very early in the hierarchy, and they control lots of other genes, so hence, the master regulatory gene idea. They come on in different parts of the animal, and have different sort of patterns that are very precise. Then what they do is they make sure that the cells in that part of the animal essentially know where they are, and they can develop in the appropriate way.
Let me give you an example. In an insect, there are three thoracic segments. There’s first a head, then these thoracic segments, then a series of abdominal segments. Of course, it’s very important that those segments be in that precise order. So there’s a set of homeotic genes that are defining the segments of the head, another set that define the thorax, and others that define parts of the abdomen. These genes have to be expressed at the right time and place for the cells to know that that’s the part of the animal they’re supposed to make. When they’re expressed at the wrong time, then you get the wrong sort of structure being made. So for example, if we were to remove or mutate some of the homeotic genes that control the abdomen, then the abdomen would be transformed into a thorax and you would have an embryo with more thoracic segments. The Hox genes are very fundamental to the animal to get the right pattern of segments. There’s another hierarchy that generates the repeating pattern, but the homeotic genes are what give identity to the segments.
Another really nice example of that is they continue to have this function all through development. They’re also responsible for giving you the right pattern of appendages and morphology in an adult fly. So if you’re familiar with an adult fly, it has a single pair of wings, and the pair of wings is on the second thoracic segment. On the third thoracic segment, just one segment back, are little structures called halteres which are used as gyroscopes, you have to actually look, probably, in a microscope to see them there. But they’re related, in a sense, to the wings. But of course, they don’t look like wings, and they’re used for a completely different purpose. But there’s a particular homeotic gene, the ultrabithorax gene, which is responsible for making the second thoracic segment different from the third thoracic segment. If you remove the expression of the ultrabithorax gene from that third thoracic segment, then you suddenly now get a fly with two pairs of wings. So now, essentially the cells of that segment think they’re the third thoracic segment, and you get a repetition of that segment.
But of course, this mutation is not good for the fly. In fact, these particular mutants can’t fly properly, because they’re not wired up to have these two pairs of wings like this. But it’s a very nice example of what the homeotic genes can do. Again, the idea is that by mutating them, you can infer what they do normally.
Can you talk about the findings of the research of Hox gene expression in fruit flies and relate it to other species?
When homeotic genes were first studied in Drosophila, many people thought that they would just maybe be an oddity of flies. Understanding the genetics of fly development was great for understanding how you made a fly, but it wasn’t going to be really relevant-directly-to understanding, say, how a human develops. It turns out that we’ve learned over the past several decades that nothing could be further from the truth. Understanding how flies develop is amazingly relevant-directly-to understanding how humans develop. And the Hox genes were actually one of the very first examples we had of how close the relationship really was, and how humans and flies and all animals develop.
It turns out that you can, in fact, find these same sets of genes in vertebrates, and in a vertebrate such as a human. Amazingly enough, the genes do very similar things. They control regional identity along your body plan. So a very nice way to see what they control, for example, is the kind of vertebrae you have in your spinal column. You have certain vertebrae to hold up your skull, very flexible ones for your neck, other ones for your chest, and then for your back. It turns out that Hox genes-just as they control these kinds of repeating patterns, or the identity of pattern in fly-control those same sort of identities in a vertebrate. These genes not only are conserved, but they do very similar things in organisms that are separated by hundreds of millions of years of evolution.
Can you talk about how Hox genes are conserved in other species besides the ones you mentioned?
When the Hox genes were cloned, or first molecularly characterized in Drosophila, the leap was actually made relatively quickly to finding them in other model vertebrate systems such as the mouse, the frog, and eventually, the zebra fish. It’s from that that we sort of extrapolate to what they do, to some extent, in humans.
There are a lot of good data now in humans as well, so we have cloned or isolated those same genes from humans. In fact, the number of genes and their organization is identical to the way it’s in a mouse. There are people who essentially have diseases caused by mutations in their Hox genes. So we have some insights.
Can you explain your current research?
One of the things that my lab is interested in doing is studying the possibility that Hox genes and changing their expression plays a role in changing the morphology of animals. If we look at inappropriate expression — or loss of expression — in any of those kinds of phenotypes from homeotic genes, we see these sort of monstrous changes, if you will, in an animal.
For example, if we take away the function of the UBX gene in the second thoracic segment of a fly, we suddenly get this four-winged fly. That fly can’t fly very well, and it’s sort of a lab mutant, in that it’s not going anywhere. But when that mutant was first made by Ed Lewis, he realized that this actually might be telling us something about evolution. Because a fly-a normal fly-has only a single pair of wings. But if you look at another insect — say a butterfly, or a dragonfly — you know that it has two pairs of wings. So it occurred to scientists that maybe, in fact, what they were doing was essentially mimicking the kind of changes that had gone on in evolution. And that a gene-like the ultrabithorax, an example of a homeotic gene, could actually play a fundamental role in changing the body plan of an organism.
Now, many people objected, saying that these kinds of changes that you would see in the lab were so drastic, they were, of course, useless to an organism and wouldn’t be the kind of thing that evolution would use to gradually change the morphology of a species. But of course, scientists also had found all sorts of mutations in homeotic genes that have very, very subtle effects. They only slightly change the expression of the gene. For example, instead of having a four-winged fly, you’d have a fly that still had a single pair of wings, but now the halteres would be a little bit bigger.
So it occurred to scientists that maybe in fact these were good candidates. What my lab has been interested in doing is really delving into this problem in a deeper way and ask, “Well, if we compare organisms that have different morphologies-that in some way that we can relate to homeotic-type transformations-would we, in fact, see that there were changes in the expression pattern of homeotic genes?” And could these changes might underlie the actual evolutionary differences in the morphology.
So we chose not to do our comparative work in insects. Because insects actually have a fairly stable pattern of their appendages and their basic body plan. What we decided to look at were crustaceans. Crustaceans are things like lobsters and shrimp and they’re relatively closely-related to insects, but they have a lot of variation in their morphology of appendages. In fact, if you compared two crustaceans like a lobster-which everyone is familiar with from eating them-and then something like a sea monkey, or brine shrimp-which many people have seen because they feed them to their tropical fish: Something like a brine shrimp has eleven pairs of appendages-but they’re all essentially the same. If you look at a lobster, it’s actually got ten pairs of walking legs, but on its thorax-the main part of its body-up towards its head, it has other appendages which are part of that main thorax but are used instead for feeding. And morphologically, they actually look a lot like the appendages of the jaw.
So in this example, we wondered, “Well, could these, in fact, be essentially natural kinds of homeotic transformations?” So, in collaboration with others, we looked at where one of the homeotic genes-ultrabithorax-was expressed in different species of crustaceans and asked, “Did that expression correlate with differences in the morphology between different types of crustaceans?” And we found a quite remarkable answer-that in fact, there was a very good correlation with one of the changes. So in something like Artemia or brine shrimp, where all of the segments of the thorax look the same, the ultrabithorax gene expression is all throughout the thorax, starting at the very first segment and going all the way posterior. But if you look at something like a lobster or a crab, the expression actually begins much more posterior. It can begin, for example, at the fourth thoracic segment. In these animals, the first three thoracic segments don’t have the big legs that the animal’s using to walk on. In fact, it has very little appendages, which have the same morphology as some of the appendages of the jaw. And functionally, they’re used for feeding instead of locomotion.
We actually did this sort of comparative analysis in a wide range of species of crustaceans and we saw that, basically, if you knew where the UBX gene was expressed in the very early embryo, that was telling you where you would have the transition from the jaw-like appendages that were used for feeding to the other appendages that were used for walking or swimming. We proposed that, in fact, it was the evolutionary change in where UBX was expressed that actually was responsible for these different species of crustaceans having these different morphologies. What we’ve been trying to do now is to functionally test this idea. We’ve correlated the expression of UBX with this morphology, which is a very exciting result, but now what we’re trying to do is actually inappropriately express the UBX gene in a crustacean and ask if we get the kind of transformation that we would expect.
So for example, in a lobster, if we were able to express UBX throughout the thorax, we would expect that instead of having the five pairs of walking legs, we would actually get more walking legs, and they would be up towards the head, where they would normally be small appendages for feeding.
Do you see this research relating at all to humans?
Well, the kind of work that we do, again, is very focused on trying to understand the evolution of organisms. But it’s very relevant to thinking about humans as well. Because we’re actually addressing the functions of these genes. We know that these genes are serving roles and patterning all vertebrates, including humans. So the more we understand what these genes do-albeit in our case, it would be in a crustaceans-we know that that data are really relevant to understanding also what these genes do in a human. By these sort of increasingly sophisticated comparative studies in different organisms, it gives us actually more insight into making predictions about what these genes do in humans, about how their expression is controlled, and what’s the outcome of changing their patterns of expression.
How is the Hedgehog gene important in the development of fruit flies?
Genes like the homeotic genes control regional identity in the embryo. But the embryo already has, basically, a setup of segments in it, at about the same time as the homeotic genes are coming on. There’s another cascade of genes that is responsible for first generating the basic repeating pattern, and homeotic genes give identity to those units.
But genetic studies in Drosophila have identified a huge number of genes which really are responsible for setting up the segments, per se. One of the genes that’s involved in that process is a gene called Hedgehog. Hedgehog is a signaling molecule that allows cells to communicate with one another. Through that signaling, cells are able to set up boundaries. You get the pattern of segments that you see. So if you look at, for example, larvae of a fly, there are actually grooves that are there between segments. Hedgehog is one of the genes that allow the communication to go on that allows patterning system to set up that pattern of segments. It’s very important to have those genes on, otherwise the segments lose their sort of separation, and you get these phenotypes where the segmentation is altered. They still try to carry out particular identities, but the nice orderly separation of segments gets destroyed.
There are a whole series of genes that are involved in that and they form a sort of cascade, and they’re genes such as Hedgehog and Wingless, which are important in setting up these boundaries between segments and allowing patterning to go on.
Can you talk about the function of the Hedgehog gene in flies and humans?
In the fruit fly, Hedgehog functions as a signaling molecule that plays a role in establishing the boundary between segments. When Hedgehog is not there, the correct patterning does not occur, and you, in fact, get a defect in every single segment of the animal and they sort of fuse together.
When you look at the related genes, the Hedgehog in other organisms, Hedgehog serves the same biochemical function. It’s a secreted molecule used to signal from one cell to another. But its context in the embryo can be quite different. So in the case of vertebrates, for example, Hedgehog has been intensely studied in model systems such as mice and in zebrafish and it plays many roles in development. It does not play an equivalent sort of role in setting up boundaries between repeating segments of a vertebrate. But it plays the same role in signaling from one population to another. So for example, it’s responsible for signaling between cells during the development of your spinal cord. And it makes sure that you have certain types of neurons in one part of your spinal chord and other types of neurons in another part of the spinal cord. So at first, that may seem very different from its function in the phenotype that you get by mutating the Hedgehog gene in Drosophila, but of course, they’re linked together, because biochemically, the Hedgehog gene is still doing the same thing. It’s acting as a signaling molecule, and in fact, the receptors that it’s acting through are conserved — the cascade of genes that’s allowing the reception of the signal in the cell that’s getting it is the same, and so you see a lot of biochemical similarities. But it’s a little bit different from homeotic genes, where there’s very, very similar sort of developmental roles. Hedgehog can have very different developmental roles. In fact, within an individual animal, Hedgehog can do many different things. It’s always a signaling molecule, but what goes wrong when you change that signaling is different in different parts of the animal and in different species.
Can you talk about the colinearity of the Hox genes?
In the case of Drosophila, there are eight Hox genes or homeotic genes. It was discovered that these genes were actually arranged on the chromosome in the same order that they’re expressed in along the axis of the embryo. That means that the gene that’s expressed in the most anterior part of the fly, up in its head, is located at one end of a sort of cluster of these Hox genes. And the gene “Abdominal B,” which is expressed in the most posterior part of the fly, is at the other end of this little cluster of genes. And so people called this “colinearity.” There was a relationship between the position on the chromosome and the position on the embryo where the gene was expressed. In Drosophila, the complex is actually split into two, but this colinearity is clearly still there.
One of the fascinating discoveries that was made when homeotic genes were cloned out of other organisms, such as the mouse, was that in fact, the genes maintained this colinearity. They were arranged on the chromosome in the same order that they were expressed along the axis of the animal. A lot of research has gone on to ask, why is this? Why do the genes stay in this order? Because the genome rearranges itself so quickly during evolution that if there’s no real reason for two genes to stay together, they’ll eventually get split apart and end up on different chromosomes. So there must be some reason that the genes maintain this colinearity. One of the properties of homeotic gene expression in vertebrates is not only that they’re collinear in terms of spatial expression, but that the Hox genes in vertebrates also show something called temporal colinearity. They also come on in a precise order — the ones at the most anterior of the animal come on before the ones in the most post part. A variety of work that’s been done suggests that in fact, it’s this temporal property of a Hox gene expression that requires the Hox genes to stay in this nice organization, and that’s what’s actually the force, essentially, that’s keeping them together like that. But this is a very active area of research, where people have always been very fascinated by this property displayed by homeotic genes.
Can you talk about the concept of the “genetic toolkit”?
Genetic screens that have been done in model organisms such as Drosophila, or the nematode C. elegans and mice and in zebra fish have revealed to us a number of genes that are responsible for controlling-very early on-the patterning of the embryo. Surprisingly, of course, we know that there’s a very close relationship between a lot of these genes that are used over and over again in a wide range of organisms to get this patterning done. Geneticists now think that there is essentially a “genetic toolkit:” a sort of family of genes that is used over and over for sort of patterning functions in the embryo. So we’ve come up with this idea that there’s this genetic toolkit that the animal uses to set up its body plan. But then it can use that toolkit to make novel structures, or to change structures.
For example, there’s a genetic toolkit of genes that are used to make appendages. So during the course of their evolution, if they’re changing their patterns of appendages, putting appendages in new places, organisms have that toolkit to use, to pull from, to develop that structure. And often the genes are related to one another, also, in that one gene will then turn on a whole other set of genes.
The toolkit is sort of interconnected, and pulling one member of that toolkit often pulls all the members of that toolkit along with it. But we think, then, that there’s basically a sort of definable family or a set of genes, they’re not necessarily all related to one another, some of them are transcription factors, some of them are signaling molecules and things like that-but they’re sort of basic genes that are used any time you want to build any sort of structure or pattern any sort of structure in a developing animal.
Can you talk about the concept of regulatory genes in general?
“Regulatory genes” is the term we use for genes that essentially regulate the expression of other genes. They can do this in a variety of ways. A number of genes are transcription factors. They encode proteins which then bind to DNA and turn on or off other genes, and so we think of these as regulatory genes. There are other genes that are regulatory in nature, but they do quite different things. They can signal from cell to cell — so for example the Hedgehog gene can be thought of as a regulatory gene because it’s involved in cell-cell signaling and it causes the cell receiving the signal to respond in a particular way and turn on and off other genes in that cell. So these are genes that basically are regulating development, or having usually fairly large-scale effects on what goes on in the organism and regulating the expression and function of other genes.
What are still the big questions unanswered questions in your field?
The big unanswered question that I think my research is sort of trying to get to-and it’s a far-off goal-is really understanding how evolution works. Understanding what are the changes that go on in an organism that allow it to diversify in terms of its morphology or patterns of development. The questions that we’re really focused on are trying to understand how segmentation has evolved, and how it’s changed, and how the process occurs differently in different groups of organisms, and then how morphologies have changed-and, again, specifically addressing whether homeotic genes play a role in that.
But the really big question is really trying to understand how does evolutionary change occur. We’re most interested in understanding that on a macro scale-how you get differences between species; how in both relatively closely-related species-say, two species of crustaceans-are much more divergent species, such as trying to compare a mouse, for example, to a fly. And understanding how did you get these kinds of changes in the course of evolution? It’s, of course, a very difficult question, because what we’re trying to do is go backward in time and ask how organisms change. We can do that by looking at the endpoints of evolution and asking how the organisms, now at these endpoints of these lineages, are different. If we can look at enough organisms and understand their development in some detail, basically we can extrapolate backward, bit by bit.
If we look at organisms that are relatively closely related, we can extrapolate a short distance back. But if we go further and further apart, our conclusions may not be as sort of solid all the time, because we’re reaching really far back in time, but ultimately what we’d like to do is understand how the process of evolution works on a sort of grand scale. I don’t think we’re going to come to answers immediately on these kinds of questions, despite all the tools and techniques and genetics that we have because it’s a very difficult question. But I think we’re making incredible headway into this issue now.
You mentioned a “hierarch of separation.” Can you briefly talk about that?
Geneticists working on Drosophila melanogaster carried out genetic screens to find all the genes that played a role in setting up the early embryonic pattern of the larvae. They found that there were five classes of genes that actually were controlling this pattern. There were what we call “maternal coordinate” genes, “Gap genes, “Pair-Rule” genes, “segment polarity” genes, and “homeotic” genes. They basically act in that kind of hierarchical order to set a pattern of the embryo. What is happening is that the mother builds an egg. And in that egg, she localizes RNAs to the two ends. And then these RNAs are localized-the machinery to localize those are maternal coordinate genes. Those RNAs get translated and form gradients of protein in the early embryo. Then the first genes that read out that gradient are called Gap genes. They come on in large regions of the embryo in response to that gradient. The overlap interaction of those Gap Genes turn on the first periodic genes-called Pair-Rule genes, which are thoracic segments that come on in a seven-stripe pattern in the early embryo only about two hours after fertilization. And then the overlap of those Pair-Rule genes turns on the segment polarity genes, which are in a stripe in every segment. They’re responsible for patterning individual segments and maintaining their boundaries. An example is the Hedgehog gene. And then the same Gap genes are also activating homeotic genes, which come on in large domains, and they give identity to segments that are being set up by the rest of the hierarchy. So you have this hierarchy of genes that is sequentially subdividing the egg into smaller and smaller units until you get the pattern of segments, and then you have the homeotic genes giving identity to those. So those five classes of genes actually are then the hierarchy that controls the early pattern of the Drosophila embryo-forming the axis, setting up segments and giving identity to those segments.