The metamorphosis of a fertilized egg into an adult organism requires a complex web of gene and protein interactions that scientists have attempted to elucidate for decades. Developmental biologist Kathryn V. Anderson has discovered many of the molecules that lay the groundwork for body plan differentiation, specifically those controlling the dorsal-ventral patterning in the fruit fly Drosophila melanogaster. Such patterning is required for specification of the nervous system, muscles, and other mesodermal cells. These signaling pathways are conserved in higher organisms and, in humans and mice, are critical for the activity of the innate immune system.
Over the past several years, Anderson has used the genetic approaches she helped pioneer in Drosophila and applied them to vertebrate development. In her Inaugural Article in this issue of PNAS (1), Anderson provides a progress report on her ongoing strategy to phenotypically identify genes that control early neural development and mesoderm formation in the mouse embryo.
Anderson's interest in developmental biology was first piqued during a high school science fair project as she peered through a microscope at the “mermaid's wine glass.” She marveled at these single-celled algae, formally known as Acetabularia crenulata, with their slender stalks and umbrella-shaped caps, but most intriguing were the compartments within the cell. “I thought this was great that the different parts of the cell had different roles,” she says.
Born in La Jolla, CA, in 1952, Anderson's interest in development was nurtured by her schooling and her supportive parents. At San Diego's Point Loma High School, a “great biology teacher,” Michael Lorch, fostered her interest in life sciences. Moving north to attend college at University of California, Berkeley, Anderson decided to pursue biochemistry. “I was pretty much a reductionist biologist, and biochemistry seemed the hardest of the biological sciences and the most objective,” she says. After graduating with her bachelor's degree in biochemistry, Anderson hoped to apply the rigorous mechanistic approaches she learned to understanding brain development. In 1973, Anderson began graduate studies in neurodevelopment at Stanford University (Stanford, CA). But, unhappy with the program, she left after 2 years with a master's degree in neuroscience.
While at Stanford, Anderson quizzed her colleagues about the future of biological research. “At this point, I had a pretty fuzzy idea of what I wanted to do,” she says, and she returned home to San Diego where her desire to do “something for humanity” led her to begin medical school at the University of California, San Diego (La Jolla, CA). However, she quickly realized that her true interests lay in basic science. “The clinical work wasn't my cup of tea,” Anderson explains. “I like the dissection of basic mechanisms much more than the kind of interpersonal skills you need as a physician. The lab was where I felt most at home.”
In consulting with her peers while at Stanford, Anderson agreed with their consensus view about the importance of genetics research over the long run. Trusting their advice, Anderson decided to study development in Drosophila. She left medical school and restarted her graduate career at the University of California, Los Angeles (UCLA), in 1977, with biologist Judith Lengyel. “I sought out a woman adviser because I thought the communication would be more straightforward,” says Anderson of Lengyel. “She was a dynamic young assistant professor, full of energy and full of ideas, and I was her first student. I think that most first student-mentor relationships are much more rich and complex. She was an extremely supportive adviser.”
Anderson's work at UCLA focused on the biochemistry of early Drosophila development, specifically examining how DNA replication correlates with histone mRNA synthesis (2, 3). She found that, during the first two hours after fertilization, the development of the Drosophila embryo is under maternal control, where maternal RNA and proteins stored in the egg direct cell division and differentiation. This is then followed by an abrupt shift when gene transcription begins in the embryo.
In the mid-1970s, the first connections were being made between molecular biology and developmental genetics. In 1975, David Hogness and his colleagues at Stanford were the first to clone Drosophila genes. “There was a history of developmental genetics, but it was all pretty complicated and sort of mystical,” says Anderson. Researchers such as Ed Lewis had elucidated how genes in the bithorax complex directed development, but the proofs were genetic and rather abstract. “It was pretty hard to imagine how the phenomena [Lewis] described would work at the molecular level,” says Anderson.
In 1981, Anderson moved to the Friedrich Miescher laboratory in Tübingen, Germany, to work with Christiane Nüsslein-Volhard, one of the few scientists who was studying genetic control of embryonic development in Drosophila. Nüsslein-Volhard combined embryological manipulation with developmental genetics to identify the actual molecules guiding developmental processes. “It turned out to be a wonderful combination of tools that made it possible to figure out how the fly embryo knows its back from its belly,” says Anderson. Just prior to Anderson's arrival, Nüsslein-Volhard and Eric Wieschaus initiated massive genetic screens for mutations that disrupted fly development. These experiments exploited the strengths of Drosophila genetics to identify the genes that controlled the development of specific regions of the body, eventually earning Nüsslein-Volhard and Wieschaus the Nobel Prize for physiology in 1995.
Kathryn V. Anderson
Anderson joined the Tübingen laboratory just one month after its launch. The atmosphere was intense, and Anderson believes Nüsslein-Volhard influenced every aspect of her future research. “She's brilliant,” Anderson says. “And she's extremely creative and tough and insightful, and she sort of generates this attitude that if you aren't going to do science this way then it's not worth doing. Science demands an intense commitment to the biological question you are asking.” Anderson participated in large-scale genetic screens for maternal effect mutants—embryos whose development was disrupted because of mutations in the maternal genome. Between 1984 and 1986, Anderson and Nüsslein-Volhard coauthored five publications in which they identified genes involved in components of the dorsal-ventral signaling, such as Toll, easter, tube, pelle, spätzle, pipe, nudel, and snake (4-8). Many of the mutants that Anderson would study over the next 14 years were discovered in this initial screen.
In 1985, Anderson returned to the United States to accept an assistant professorship at U.C. Berkeley's Department of Molecular and Cell Biology. Over the next 8 years, she continued to tease apart the genes involved in dorsal-ventral pattern formation while rising through the ranks of associate and then full professor. During this period, Anderson completed the experiments that she rates as her finest and most influential to date. Scientists in Anderson's laboratory and several other laboratories identified about a dozen genes affecting cell fate along the dorsal-ventral axis, including dorsal, cactus, gastrulation defection, and windbeutel. They delineated the genetic pathway demonstrating how the proteins work together to specify all cell types such as muscle, nerve, and epidermis.
Anderson and her colleagues also cloned the key developmental genes Toll, spätzle, and easter. Toll encodes a transmembrane protein and, together with its ligand, Spätzle, controls the entire dorsal-ventral pathway in Drosophila development (9-12). “The genetic ordering of the Toll pathway is probably the thing that has had the biggest impact because of the importance of Toll-like receptors in mammalian immunity,” says Anderson. In mammals, Toll mediates innate immunity with each member of the Toll-like family of receptors dedicated to recognizing a different bacteria or virus.
“You don't need to do a million mouse experiments, and you can still find remarkable, interesting, new levels of biology.”
As the conservation of Drosophila genes in vertebrates began to be more widely studied in the early 1990s, Anderson became “very excited about taking what we had learned from flies and applying that to the mouse.” This hope led to a sabbatical at the National Institute for Medical Research (Mill Hill, U.K.) in 1993-1994, in the laboratory of Rosa Beddington. “Rosa was the world's best mouse embryologist,” says Anderson. “She was completely fearless embryologically and did amazing transplantation experiments.” Under Beddington, Anderson learned about the possibilities of mouse genetics, and her experiments in England helped reveal that the Toll receptor plays a different role in mice than it did in flies. In flies, Toll is involved in both cell fate specification and innate immunity—playing a critical role in response to fungal infections. Toll, however, did not seem to play a role in early mouse development. This finding triggered a new path of research for Anderson, studying innate immunity in flies, and marked the shift in her research from Drosophila to mice (13-19).
When Anderson returned to Berkeley, she convinced a graduate student, Andrew Kasarskis, to use forward genetics, rather than the traditional reverse approach, to identify new genes vital for mouse embryogenesis (20). “In reverse genetics you start with the gene. You guess that a molecule is going to be important, you inactivate it, and the mouse tells you whether you were right or not,” explains Anderson. “In forward genetics, you start with a process. You mutagenize at random with a chemical mutagen and then pick mutants because they have an interesting phenotype.”
In 1994, chemical mutagenesis was being used to look for lethal mutations in a genomic region of interest, but no one, according to Anderson, was performing an unbiased screen of the entire genome for genes that, if mutated, altered a specific process in embryonic development. In 1996, as Anderson's research increasingly focused on the mouse, she accepted a position in the Molecular Biology Program at the Sloan-Kettering Institute in New York—lured to the city in large part by geneticist Timothy Bestor at Columbia University (New York, NY), whom she married last year.
Over the past 5 years, during which time Anderson became the Chair of the Developmental Biology Program at Sloan-Kettering, she and her colleagues have used forward genetics to examine the progeny of about 400 mice carrying new mutations. She and her team have screened approximately 12,000 mutations and selected 43 that produce gross physical abnormalities in morphogenesis and patterning halfway through embryonic development. This research, described in her accompanying Inaugural Article in this issue (1), shows that many of the mutants affect nervous system development. A second class of mutants demonstrates defective mesoderm patterning.
”This [article] goes back to my real old interest in how the very early body plan of animals is established,” says Anderson. “If the Toll pathway isn't specifying cell fate [as in Drosophila], then how does the mouse embryo make its tissue types?” Many biologists are working on the same question, but Anderson feels her approach provides a unique starting point: her team is picking mutants in which tissue organization is disrupted in the early mouse embryo. Many of the molecules identified control cell migration and epithelial organization, which are not the same processes that other developmental biologists are working on, explains Anderson. These findings make sense, she says, because, in the mouse embryo, cells are being specified at the same time that they are interacting and moving. “That's very different than in the fly,” says Anderson. “So there is this interesting coupling between a cell's fate determination and these cellular migrations.”
Anderson describes her PNAS Inaugural Article as a progress report on forward genetics in the mouse. “I hope what people take away is that you can do this type of genetics in the mouse like you can do it in the fly and zebrafish,” she says. “There are different constraints, but the same strategy will identify more key players for a particular biological process.” Mouse genetics has traditionally been performed with knockout strains examining one gene at a time. “The fact we identified a set of genes with related phenotypes that are important to neural patterning, not just one random gene, is a good yield,” says Anderson. “My world view is this kind of screen can be done cottage industry style—you don't need to do a million mouse experiments, and you can still find remarkable, interesting, new levels of biology.”
As far as applications to human biology are concerned, many birth defects are due to problems with cell migration and intercellular signaling. “My lab won't directly pursue the human biology connections, but they are so obvious that it will make it easy for other people to follow up on them,” she says. With a laboratory bursting with intriguing mutants, Anderson has her hands full, but she says she is particularly excited about learning more about the relationship between cell fate specification and cell migrations in the early mouse embryo. “That's a big puzzle and something where I hope we can make some contributions.”
This is a Biography of a recently elected member of the National Academy of Sciences to accompany the member's Inaugural Article on page 5913.
