Arthur Lander
Research Interests
|
|
My lab is interested in how cells communicate with each other to coordinate the complex behaviors that underlie development, regeneration and cancer. Our research touches on many fields, including Cell Biology, Developmental Biology, Neurobiology, Cancer Biology and Computational Biology. We employ a range of techniques from cell culture to binding studies, in vitro mutagenesis, the generation and analysis of transgenic and "knockout" mice, and computer modeling.
|
|
Many of the projects in the lab focus on the roles of co-receptors in growth factor and morphogen signaling. Co-receptors are molecules at the cell surface that don't themselves transduce signals, but cooperate with receptors and their ligands to make binding and/or signaling possible. One major class of co-receptors is the cell surface heparan sulfate proteoglycans (HSPGs). These molecules contain long carbohydrate chains that can bind a wide variety of growth factors and morphogens, such as Fibroblast Growth Factors, Wnts, Transforming Growth Factor-Beta family members, Hedgehogs, Epidermal Growth Factor family members, and several others.
|
|
One of our goals is to understand the molecular mechanisms by which co-receptors act. Using cell culture approaches, we have been working out how HSPGs (and another class of co-receptor) promote the assembly of signaling receptor complexes by Bone Morphogenetic Proteins (BMPs), which are growth factors of the Transforming Growth Factor-Beta family. Our work points to an important role for HSPGs in the assembly of type I and type II receptor subunits.
|
|
A second goal is to elucidate the roles of HSPG co-receptors in mammalian development. We do this by "knocking out" genes in mice. A major family of cell surface HSPGs is the glypicans (there are six glypicans in mice), and we have knocked out glypicans 1 and 2, and are also working with a mutant in glypican 4. Glypican 1 is the major HSPG of the adult brain, and glypican 1 mutants exhibit delayed development and, ultimately, reduced brain size; this effect is enhanced when animals are mutant for both glypicans 1 and 4. By combining glypican mutations with mutations in the enzymes that synthesize heparan sulfate, we are able to generate animals with phenotypes of different severity, in order to identify sites of glypican action. By combining glypican mutations with mutations in signaling pathways (e.g. FGFs), we are testing hypotheses about the types of signaling pathways that glypicans control during development.
|
|
A third goal is to understand the roles of HSPG co-receptors in cancer. We have recently shown that glypican 1 is strongly induced in human pancreatic and breast cancer, and removal of glypican 1 from such cancer cells renders them insensitive to numerous growth factors, and poorly able to grow in animals. Such cancer cells frequently also express high levels of syndecans (the second major family of cell surface HSPGs). Curiously, the syndecans are not sufficient to act as co-receptors in these cells even though they clearly do so in non-cancer cells. We have shown that this has do to growth-factor mediated shedding of syndecans. We are also in the process of using our glypican-1 knockout mice to test whether glypicans are required for the formation of breast tumors by targeted deletion of the p53 tumor suppressor gene.
|
|
One of the most puzzling aspects of co-receptors is why they exist at all. Certainly, growth factors and receptors exist that get together and signal without the benefit of any other "helper" molecules, so why did evolution select for a situation in which so many factors require co-receptors? The answer may come from the study of morphogens, growth factors that pattern tissues during development by directing cells to adopt different fates depending on the level of morphogen-receptor occupancy they experience. Every known morphogen utilizes a co-receptor (usually a HSPG). Because morphogens are secreted at discrete places and diffuse away to form concentration gradients, developmental patterning depends on achieving morphogen gradients that are accurate (i.e. the right shapes and levels) and robust (i.e. relatively insensitive to changes in activities of genes, environmental factors, etc.). We use computational methods to calculate the shapes and dynamics or morphogen gradients that would be formed by known processes (diffusion, receptor-binding, degradation, transcriptional feedback, etc.), in order to determine how co-receptors affect gradient shape and robustness. These results are compared with a variety of in vivo experiments in fruit fly, frog and zebrafish embryos. The data support an important role for co-receptors in the robustness of morphogen gradients to environmental influences.
|
|
A final project in the lab concerns the developmental biology of Cornelia de Lange Syndrome (CdLS), a human birth defects syndrome that affects up to 1 in 10,000 live births, and is characterized by limb abnormalities, small body size, gastrointestinal malformations, characteristic facial features, and mental retardation. We recently helped identify the NIPBL gene as the site of mutations that cause CdLS. Apparently, loss of function of only a single copy of NIPBL is sufficient to produce the syndrome. Using targeted ES cells, in which the NIPBL locus is disrupted by a gene trap insertion, we have produced the first mouse model of CdLS, which is providing us with new ways to work on the developmental etiology, diagnosis, and potential treatment of this disorder.