Modeling gene regulatory networks for cell fate specification

Abstract

Many cellular systems exist in more than one state. Two of the simplest and best-characterized examples are bacteriophage λ, which can be in the lytic or lysogenic state (Ptashne, 1992), and the lac operon of Escherichia coli bacteria which can be in either an active state, in which lactose is metabolized, or a repressed state, in which it is not (Novick & Weiner, 1957; Ozbudak et al., 2004). These states can be remarkably stable. For example, in the absence of signals which induce the lytic cell fate, the likelihood of any given cell infected with phage λ switching from the lysogenic state is about 10−7 per generation (Aurell et al., 2002). In multicellular organisms, a progenitor cell often selects between terminal cell fates which are even more stable; under natural circumstances, these decisions are irreversible. One complex but relatively well-studied example is the development of the cells of the immune system from a hematopoietic stem cell. The process of lineage commitment in this case involves a hierarchy of binary cell fate decisions in which a cell moves progressively closer to adopting certain terminal cell fates and loses access to others (Orkin, 2000; Medina & Singh, 2005).