Image courtesy of Miriam Rosenberg
Long before they pack lunch boxes with peanut-butter sandwiches or insist that vegetables be finished before dessert, mothers make sure to stockpile their young with what they need to grow.
Such motherliness begins well before birth, when a fetus in the womb survives on nutrients delivered through the umbilical cord.
Like their mammalian counterparts, fruit-fly moms cram eggs of their young with care packages of factors essential to proper development.
Studies of these so-called maternal-effect gene products offer crucial clues to the earliest stages of growth. But the mother's contribution to the developmental program creates a unique challenge for geneticists, said Dr. Susan Parkhurst, whose Basic Sciences Division graduate students have discovered functions important for early development of the fruit fly Drosophila.
Geneticists typically work like aspiring auto mechanics, pulling out a piece of the engine to see if the car still runs. In the research laboratory, this translates to making and analyzing mutations - genetic changes that alter gene function - in specific genes to determine their biological role.
Info deposited by mother
During early embryonic growth, though, many vital functions are specified not by the embryo's own genes but rather by processed genetic information deposited by the mother before the egg is even fertilized.
Mutations in the embryo's genes (known as zygotic genes) thus yield no clues to the function of maternal contributions. So scientists must turn back to the mother, looking for hidden mutations that have little or no effect on her outward appearance but may have dire consequences for her offspring.
The problem, Parkhurst said, is that many such mutations in the mother are lethal to her offspring. Such defects provide little insight into the function of these genes, other than the fact that they are required for life.
As a postdoctoral fellow at the Imperial Cancer Research Fund in London more than 10 years ago, Parkhurst set out to find a way to identify a large collection of maternal-effect genes for analysis.
"Back then, if you looked at what people knew about early development in the fruit fly, about 85 percent of mutations that had been identified were in the genes of the embryo itself," she said. "Only about 15 percent of mutations were in maternal-effect genes. We wanted to look for maternal effect mutations, but the genetic screens to do this were brutal. They were incredibly labor-intensive."
Parkhurst came up with a genetic trick of her own. Using strains of flies that harbored a mutation that reduced but didn't eliminate the function of many maternal-effect genes, she conducted a screen that identified hundreds of new genes.
"This allowed us to take a huge step forward," she said. "People couldn't look for these kinds of mutations until there was a practical way to do it."
Parkhurst hoped to find - and did find - three kinds of mutations. One class contains known genes that performed known functions previously identified in other systems. Another class consists of previously undiscovered genes involved in known biological pathways. They also discovered about 100 unknown genes with completely unknown function.
Two graduate students in the Parkhurst lab, Craig Magie and Miriam Rosenberg, study two maternal-effect genes of the first class. While the genes they work on have previously been identified in other organisms, Magie and Rosenberg have uncovered previously unknown functions important for early fruit-fly development.
Protein called Sir2
Rosenberg studies a protein called Sir2, which has analogous counterparts in organisms as diverse as yeast and humans. Sir2 is best known for its role in gene silencing, a form of gene regulation in which whole regions of chromosomes are inactivated. Though silencing is required for many normal processes, it also has been implicated in some forms of cancer.
Both expected and unusual functions for Sir2 emerge from Rosenberg's studies.
"The surprising thing we found was that in Drosophila, Sir2 plays an additional role in gene repression that is distinct from silencing," she said. "Sir2 seems to be important for shutting off the expression of genes that are not located in chromosomal regions where silencing occurs."
One result of the Sir2 mutation is to perturb the normal process of sex determination, Rosenberg said. "We noticed that flies with Sir2 mutations weren't too happy," she said. "There were far too few males in the fly stocks."
By examining flies that produced too much or too little Sir2 protein, Rosenberg determined that Sir2 plays a key role in regulating one of the master control switches for fly sex determination, a gene called Sex-lethal.
Understanding exactly how Sir2 works can now be studied by looking more closely at processes that require Sir2. As with other genes, an important question they hope to answer is how Sir2 is selected to function in some processes and not others.
Gene called Rho
Magie studies the gene for a protein called Rho, which enables cells to modulate their shape and movement in response to outside signals. These processes occur repeatedly through early development and are also likely to be important in cancer, as tumor cells frequently display abnormal morphology as they acquire the ability to invade surrounding tissue.
Though the Rho mutation was discovered in Parkhurst's initial hunt for maternal-effect genes, Magie found mutations in the embryo's own genes for Rho exhibit developmental defects as well.
Using electron microscopy, a technique that allows him to greatly magnify and visualize the developing embryos, Magie discovered that embryonic mutations lead to several ultimately lethal defects in the earliest stages of development. One striking abnormality is the failure of the primitive brain to be internalized into the embryo; instead, it remains on the surface of the embryo.
More recently, Magie developed an antibody that specifically targets Rho to determine the protein's location in the embryo. He observed that Rho accumulates between adjacent cells in specialized adhesive structures known as adherens junctions, which are important for cell-to-cell communication.
Using an antibody that specifically targets Rho, Magie was the first to observe that Rho binds directly to two forms of a protein called catenin, which are components of adherens junctions.
"The finding that Rho is located in adherens junctions makes sense in terms of early development because both cell shape and cell adhesion are likely to play a role in dorsal closure, a process in which an outer layer of the embryo seals together."
During his studies, in which he uses a fine needle to inject different compounds into embryos, Magie made an unexpected discovery: Rho protein accumulates around the injection site, suggesting that it may be involved in wound healing.
Last year, Parkhurst received a Hutch pilot grant to investigate the genetic basis for this process, an approach that has never been attempted.
"It seems that while Rho is recruited to wounds inflicted very early in development, when the embryo is essentially one cell, it is not present at wound sites in older, multicellular embryos," she said. "We now plan to look for genes involved in all stages of the process as a way to begin to dissect these pathways."