This is the text of the first test that I wrote for P.Z.'s class. The test is supposed to be free-standing essay written for a popular audience, so it seemed perfect for the blog. It's long, but the topic is pretty cool, so you should read it. The title is Bicoid: Evolution of novelties.
Now and again, when the compost bin in my kitchen has not been taken out recently enough, or some fruit has gotten a little too ripe, we get some unexpected visitors. Fruit flies, those persistent, minute and irritating insects, are all too common in university housing, are also common in university research labs. Fruit flies are a model organism, something that biologists use as a proxy, an animal that can be studied easily with results that can hopefully be generalized to other animals. It is not surprising, then, that the development of these flies, and the molecular basis of that development, has been studied in exacting detail. One of the best studied genes that affects fly development is called Bicoid. Although this gene is very specialized and is not found in many other organisms, it can still inform us about the evolutionary acquisition of novel traits.
Bicoid (bcd) is not a normal gene. We typically think of genes being used by cells to be translated into a protein that can then perform various tasks throughout the cell. Bcd does not follow these rules. In fact, during the normal life of a fruit fly bcd is never produced. The DNA that encodes this gene is almost always silent. The exception to this rule is in the ovary of a pregnant female fly. Here there are cells that produce all of the important proteins and other chemicals that the egg will need to develop. These “nurse cells” are the only ones that transcribe the bcd gene into mRNA, but the nurse cells never translate those mRNAs into proteins. Instead the bcd mRNA is shuttled out of the nurse cells and into the cytoplasm of the ovum. Once it is there bcd will perform its single immense task, to inform the developing embryo of its orientation, telling certain cells to
Logan Luce October 14, 2010
build a head and cells on the other end to build an abdomen. To understand how bcd imparts this knowledge onto other cells, we must first examine some basic fly embryology.
The very early fly embryo does not consist of a multitude of differentiated cells. Instead there are a plethora of nuclei all floating near the outer membrane of a large sac of cytoplasm. This is the environment where bcd takes effect. Because there are no membranes dividing cells, signaling molecules like the bicoid protein (BCD) can flow freely throughout the organism, reaching all of the nuclei of the embryo without having to cross any membrane barriers. This particular embryonic organization allows for BCD to form a concentration gradient. The mRNAs that were produced by the nurse cells are localized in the anterior portion of the embryo, but the protein that is then translated can flow freely around the embryo until it is broken down into its component pieces. This means that there is a high concentration of BCD in the anterior of the embryo, where new protein is constantly being produced, but much lower concentrations near the posterior of the embryo, because BCD is being broken down without being replaced. This is the key to anterior-posterior patterning in fruit flies. By reading the concentration gradient of BCD, a nucleus can determine its location within the embryo. If BCD is high, that nucleus must be close to the front of the embryo and should therefore start making a head. If BCD is low, then the nucleus must be close to the rear, and should begin making an abdomen. It is important to remember, however, that cells are not intelligent, and “reading a concentration gradient” is not as easy as it may sound. To understand how BCD concentrations actually transfer information about what body parts to build, we will need to delve into some elementary molecular biology.
Multicellular organisms such as fruit flies and humans have to have to create differentiated tissues from the same genome. This is accomplished through the use of proteins called transcription factors. Transcription factors are proteins that attach to DNA within the nucleus and encourage (or discourage) the transcription of nearby genes. They are the notes written in the margin of each cell’s copy of the giant genetic cookbook, the markers that show cells the proteins they need to make.
As you may have guessed by now, the protein BCD is a transcription factor. It is, however, a particularly complex one, which affects the transcription of dozens of genes by attaching to DNA in hundreds of places. There are some genes (ones that lead towards head development for instance) that are stimulated by BCD, while there are others that BCD represses. In some instances the situation is even more involved. For instance there are genes where BCD can either stimulate or repress transcription, depending on the concentration. By combining the BCD concentration gradient with the varied reactions and sensitivities to BCD as a transcription factor, a single molecule can create varied effects across the length of a developing fly. This complex and Byzantine system allows for fruit flies to get a patterned body that stars with a head and ends with an abdomen. But how did this system come to be? How did evolution create the many interconnected processes that are required for fruit flies to develop normally?
To answer these questions, scientists began to search the genomes of other animals for something similar to bicoid. If they could discover how the gene is different in other animals they could perhaps determine what the gene looked like in the last common ancestor, and therefore how it has changed over time. When scientists began to look for bcd in the genomes of other insects they were faced with a surprise: it wasn’t
there. In fact, the entire process of development outlined above is only present in certain flies. Most insects, such as beetles, have cell membranes separating their nuclei. They do not pattern their entire anterior-posterior axis all at once with a concentration gradient. And most importantly, they do not have bicoid. It almost appears as though the entire system was created at once. A closer look, however, reveals the true evolutionary ancestry of bcd. Although the bicoid gene is unique, the process by which it evolved is not, and the story of its evolution sheds light on the common mechanisms by which evolution creates novelty.
Although there is no gene in beetles that functions in the same way that bcd does in flies, there could still be genes with similar structure but different functions. Such a gene was found by a group of German scientists in 1999. This gene, Hox3, was known to function in a completely different way than bcd. Moreover, there was already a fruit fly gene that was similar to Hox3, called zen. How could this be? The researchers determined that there must have been a genetic duplication in the evolutionary lineage leading to fruit flies. This duplication left the flies with two copies of the Hox3 gene. One copy, zen, maintained something close to its original role, while the other copy became more and more derived. Hox3 is sometimes shared from maternal cells into the developing embryo, and bcd is now only expressed this way, while zen no longer has any maternal expression. By duplication and specialization, bcd has gained new roles while the functions of the ancestral gene are still present through zen.
This idea of duplication and specialization is a motif in biology. Again and again scientists have discovered that many new genes are specialized duplications of older genes. For instance, the hemoglobin that moves oxygen through our blood stream is
composed of two subunits that are duplicates of ancestral hemoglobin. After the duplication, they specialized until now both are dependent upon the other to function normally.
Other pieces of fruit fly development that seem novel at first glance also have precursors. Hox3 is a transcription factor, so when bcd diverged from its precursors, it could already affect the transcription of many other genes. As it continued to diverge, the ways in which it affected those genes changed slowly until it assumed a role similar to the one it now plays in fruit flies. Similarly, the process by which bcd mRNAs are localized in the anterior of the embryo is not a new innovation, but instead a process found in numerous other organisms that has simply been used again for a new purpose in fruit flies. By reusing old mechanisms for a different goal, bcd has managed to create an entirely new mode of development without having to invent anything out of whole cloth.
This repurposing of old genetic equipment is a process that is seen again and again in evolution. Some structures have been repurposed so many times in so many different ways that they no longer seem related. Who would guess that the bacterial flagellum and the syringe that makes some bacterium dangerous would use the same genetics? Although it is hard to fathom, such recycling of genetic elements is quite common. Again, the uniqueness of bicoid in development illuminated general evolutionary patterns for creating unique structures.
Model organisms such as fruit flies are used in the hopes that they may prove analogous to other species- that they will provide a model of shared patterns within a larger group. The developmental properties of the bicoid gene fail in this respect. They are novel in nearly every possible aspect. However the evolutionary history of this novel
gene does illuminate the ways that evolution typically produces genetic novelties. Even though fruit flies appear to be unique developmentally, they have at least shown us how uniqueness can evolve. Not bad for a cloud of bugs above a compost bin.
Bonneton F. 2003. Extreme divergence of a homeotic gene: the bicoid case. M S- Medecine Sciences 19: 1265-1270. McGregor AP. 2005. How to get ahead: the origin, evolution and function of bicoid. Bioessays 27: 904-913.
Peel A, Chipman A, Akam M. 2005. Arthropod segmentation: beyond the Drosophila paradigm. Nature Reviews Genetics 6: 905-916. Stauber M, Jockle H, Schmidt-Ott U. 1999. The anterior determinant bicoid of Drosophila is a derived Hox class 3 gene. Proceedings of the National Academy of Sciences of the United States of America 96: 3786.