Human cells receive a constant bombardment of signaling information, and they have evolved strategies to generate proper physiological responses. The cell processes this signaling information, correctly choosing which of the approximately 20,000 genes will be switched "on" or "off." Using genetic, biochemical, and biological approaches, Michael G. Rosenfeld is deciphering on a genome-wide scale how cells control gene expression through the integration of the output to these diverse signals, which is crucial to the body's development and its smooth operation. His studies have revealed surprising new strategies that precisely regulate the pattern of genes that are turned "on" and "off"—processes that are linked to other key cellular responses, such as DNA damage and repair. This knowledge is providing the backdrop to developing new treatments for diseases that occur when gene expression goes awry, such as diabetes, atherosclerosis, cancer, and growth defects in children.

Rosenfeld's enthusiasm for research was initially sparked during numerous discussions he had with his father, a physical chemist, on scientific topics ranging from astronomy and mathematical models to biology and evolution. "The joy he transmitted in his love of science was very infectious," Rosenfeld says. When he entered college at the Johns Hopkins University, Rosenfeld entertained thoughts of becoming an archaeologist. But a biochemistry course taught by the renowned William McElroy, who made groundbreaking discoveries in bioluminescence and later became director of the National Science Foundation and chancellor at UC San Diego, inspired Rosenfeld with the "creative fire" to shift his major to biology and biochemistry. A number of mentors in his medical and postdoctoral training helped Rosenfeld to focus his energies on understanding the mechanisms that underlie critical aspects of mammalian development and homeostasis.

Over the years, Rosenfeld has developed a comprehensive picture of cell signaling and gene transcriptional events that occur during embryonic development of the hypothalamus and pituitary gland, and an understanding of when and how these signals go wrong, resulting in disease. The hypothalamus links the nervous system to the endocrine system, and the pituitary gland secretes hormones important for metabolism, growth, homeostasis, and reproduction. These studies have served as a model for increasing the understanding of mammalian organ development and have made important contributions to comprehending the way gene expression is controlled during the development of the central nervous system and the neuroendocrine system. These lines of research have led to insights into growth defects that occur in humans. For example, Rosenfeld and his colleagues have identified the molecular basis for three forms of dwarfism, including the type based on mutations in the Prop-1 gene, the most common pituitary-based growth disorder in the world.

Study of this system inevitably led the Rosenfeld laboratory to delineate new rules by which molecular mediators called transcription factors bind to DNA to activate or repress gene expression. The expression of each gene is controlled by evolutionarily conserved sequences immediately adjacent to the coding region, referred to as a "promoter," or more distant regions, referred to as "enhancers." When the correct transcription factors dock to these sequences, a process is initiated that involves the interactions between these proteins and a network of cofactors that ultimately switches genes "on" or "off."

Rosenfeld also has elucidated a series of integrated molecular strategies that link cohorts of experienced genes to detection of specific signaling pathways. He is now exploring how missteps in the machinery integrating transcriptional events can lead to disease, including insulin resistance, which often develops into type 2 diabetes, and resistance to drugs used to treat several common forms of cancer. "One of the most exciting aspects of our current research efforts is to explore the vast unknown territory that links gene transcription to the architecture of the nucleus in which these events occur," says Rosenfeld. "This should provide critical insights over the next five years that have broad implications for human disease."